Automatic Train Control in Rail RapidTransit

May 1976

NTIS order #PB-254738

OFFICE OF TECHNOLOGY ASSESSMENT

CONGRESSIONAL BOARD

Representative Olin E. Teague, Texas, ChairmanSenator Clifford P. Case, New Jersey, Vice Chairman

SENATE HOUSE

Edward M. Kennedy Morris K. UdallMassachusetts Arizona

Ernest F. Hollings George E. Brown, Jr.South Carolina California

Hubert H. Humphrey Charles A. MosherMinnesota Ohio

Richard S. Schweiker Marvin L. EschPennsylvania Michigan

Ted Stevens Marjorie S. HoltAlaska Maryland

Emilio Q. Daddario, ex officio

— — —For sale by the Superintendent of Documents, U.S. Government Printing Office

Washington, D.C. 20402- Price $3.15

ii

OFFICE OF TECHNOLOGY ASSESSMENTDIRECTOR’S OFFICE

Emilio Q. Daddario, Director

Daniel V. De Simone, Deputy Director

URBAN MASS TRANSIT ADVISORY PANEL

George Krambles, Chairman, Chicago Transit Authority

Walter J. Bierwagen Bernard M. OliverAmalgamated Transit Union Hewlett-Packard Corporation

Robert A. Burco Simon ReichOregon DOT Train Control Consultant

Jeanne J. Fox Thomas C. Sutherland, Jr.Joint Center for Political Studies Princeton University

Lawrence A. Goldmuntz Frederick P. SalvucciEconomics and Science Planning Massachusetts DOT

Dorn McGrath Stewart F. TaylorGeorge Washington University Sanders and Thomas,

OTA TRANSPORTATION PROGRAM STAFF

Gretchen S. Kolsrud, Program Manager

Mary E. Ames Larry L. JenneyV. Rodger Digilio Bev JohnsonThomas E. Hirsch III Teri Miles

Inc.

TECHNICAL CONSULTANTS

Battelle Columbus Laboratories

. . .111

P r e p a r e d b y t h e O f f i c e o f T e c h n o l o g y A s s e s s m e n t w i t h t h ea s s i s t a n c e o f i t s U r b a n M a s s T r a n s i t A d v i s o r y P a n e l , t h i sr e p o r t d e s c r i b e s t h e t e c h n o l o g y o f a u t o m a t i c t r a i n c o n t r o ls y s t e m s a n d a s s e s s e s t h e o p e r a t i o n a l , p l a n n i n g , a n d p o l i c yi s s u e s a r i s i n g f r o m t h e u s e o f a u t o m a t e d d e v i c e s t o c o n t r o la n d d i r e c t r a i l r a p i d t r a n s i t v e h i c l e s . T h e r e p o r t a l s oc o n t a i n s b a c k g r o u n d m a t e r i a l u s e f u l f o r u n d e r s t a n d i n g t h ea p p l i c a t i o n o f a u t o m a t i o n t e c h n o l o g y i n u r b a n r a i l t r a n s i ts y s t e m s .

S i n c e r e l y , S i n c e r e l y ,

T h i s r e p o r t w a s p r e p a r e d b y t h e O f f i c e o f T e c h n o l o g yA s s e s s m e n t w i t h t h e a s s i s t a n c e o f i t s U r b a n M a s sT r a n s i t A d v i s o r y P a n e l , c o m p o s e d o f r e p r e s e n t a t i v e so f t h e t r a n s i t i n d u s t r y , e n g i n e e r i n g f i r m s , p l a n n i n ga n d d e v e l o p m e n t o r g a n i z a t i o n s , u n i v e r s i t i e s , o r g a -n i z e d l a b o r , a n d c i t i z e n p a r t i c i p a t i o n g r o u p s .

v

Preface

This report, prepared by OTA at the request of the Senate Committee onAppropriations on behalf of the Transportation Subcommittee, is an assess-ment of the technology of automatic train control in rail rapid transit systems.Automatic train control (ATC) is the general designation for a variety oftechniques by which machines regulate the movement of rail rapid transitvehicles for the purposes of safety and efficiency. Functionally, ATC in-cludes:

● Train Protection . Tra in Superv i s ion● Train Operation ● C o m m u n i c a t i o n

The use of the term “automatic” does not imply that train control or anyof its constituent functions is carried out wholly without human involvementin operat ing the equipment or in overseeing automated devices. Rather ,automatic is used to denote systems in which machines perform a substantialpart of the routine functions and there is minimal reliance on man as anoperat ional element. Man’s role in such systems is to monitor the per-formance of automatic elements and to act as the ultimate safety backup.

The history of train control technology has seen extensive, but not com-plete, replacement of the human operator by machines. The number of peo-ple required to run trains, operate wayside equipment, and supervise traffichas been reduced by automation to the point where the newest t ransi tsystems now have only a single on-board operator for the train, regardless ofits length, and a small cadre of centrally located supervisors.

The increasing reliance on automation, both in existing transit systemsand those under development, raises several basic issues about this applica-tion of technology. The importance of these issues was recognized by theSenate Committee on Appropriations Transportation Subcommittee who re-ques ted the Of f i ce o f Techno logy Assessmen t to s tudy au tomat ion infederally supported rail rapid transit projects. Correspondence relating to therequest is contained in Appendix I of this report; the following is a paraphraseof the fundamental questions posed in the letter of request:

How does reduction of man’s responsibility for directoperational control affect the safety of transit systems?

What operational advantages are to be gained fromautomation?

Is automation cost-effective, considering both capitaland operating costs?

vii

Does the p lann ing , deve lopment , and t e s t ing o fautomatic train control systems give adequate attention tothe safety, performance, and cost implications of automa-t ion?

Are there policy and inst i tut ional factors that in-fluence the selection of a level of automation or that condi-tion the application of automatic train control technology?

Because of the number and complexity of the issues to be addressed, thetechnology assessment was divided into three separate, but coordinated,studies dealing with (1) the planning process, (z) automated small vehiclesystems, and (3) automatic train control in rail rapid transit. Reports on thefirst two topics have been published in separate volumes. 1 This report dealswith the third topic, specifically the degree of automation which is tech-nically feasible, economically justifiable, or otherwise appropriate for railrapid transit.

The technology assessment presented here is the product of a combinedeffort of the OTA Urban Mass Transit Advisory Panel and the staff of theOTA Transportation Program. Major assistance was received from BattelleColumbus Laboratories in col lect ing data and providing technical back-ground information. These materials and other information collected inde-pendently were combined by the panel and staff to prepare this report. Thepanel and staff are also indebted to the urban transit system officials andrepresentatives of the transit industry who gave access to their records andparticipated in numerous technical discussions.

Since this report is the result of a joint effort, the findings should not beconstrued as the view of any individual participant. Divergent opinions areincluded; and, where the subject matter is controversial, an attempt has beenmade to present a balanced treatment.

The OTA staff members participating in this study were: Dr. Gretchen S.Kolsrud, Program Manager; Larry L. Jenney, Project Director; V. RodgerDigilio, Thomas E. Hirsch III, Bev Johnson, and Teri Miles.

lsee An Assessment of Community Planning for Mass Transit, February 1976 (Report NOS. OTA–T–16through OTA–T-27) and Automated Guideway Transit: An Assessment of PRT and Other New Systems (ReportNo. OTA-T-8), June 1975,

,..Vlll

CONTENTS

PagePREFACE . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viiLIST OF TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ... ... ... ...xiLIST OF FIGURES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ... ..o. o..xii

CHAPTER 1: FINDINGS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........1Definitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........2Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ........3Policy and Institutional Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3The Planning, Development, and Testing Process . . . . . . . ........4Operational Experience . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. ......5Assessment of ATC Technology . . . . . . . . . . . . . . . . . . . . . . ........6

CHAPTER 2: BACKGROUND. . . . . . . 0 . . . . ....0.... . . . . 0 . . ........9

Rail Rapid Transit. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11Objectives ... ... t.t.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......12Scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ......0...........012Study Method.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......13Organization. o. . . . . . . . . . . . . . . 0 . . . . . . . . . . . . . . . . . . . . . . . .......13

CHAPTER 3: AUTOMATIC TRAIN CONTROL . . . . . . . . . . . . .......15Train Control System Functions . . . . . . . . . . . . . . . . . . . . . . . . .......17Automation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......20Automatic Train Control Technology . . . . . . . . . . . . . . . . . . . .......22A Walk Through a Transit System . . . . . . . . . . . . . . . . . . . . . . .. ....33Levels of Automation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......40

CHAPTER 4: TRANSIT SYSTEM DESCRIPTIONS . . . . . . .0 . . . . . . . . .43Bay Area Rapid Transit (BART) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ 45Chicago Transit Authority (CTA) . . . . . . . . . . . . . . . . . . . . . . .......49Massachusetts Bay Transportation Authority (MBTA) . . . . .......55New York City Transit Authority (NYCTA). . . . . . . . . . . . . .......60Port Authority Transit Corporation (PATCO). . . . . . . . . . . . .......65Systems Under Development . . . . . . . . . . . . . . . . . .. .. ... ..s ...1..69

CHAPTER 5: OPERATIONAL EXPERIENCE . . . . . . . . . . . . . . . ...0...75Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....77Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ....0.... . . . . . . ...0...77

Issue O-1: Train Protection . . . . . . . . . . . . . . . . . . . . . . .......78Issue O-2: Train Operation . . . . . . . . . . . . . . . . . . . . . . . .......81

Issue O-3: Design Safety . . . . . . . . . . . . . . .. * . . . . . . . . .......84

Issue O-4: Passenger Security . . . . . . . . . . . . . . . . . . . . .......87

Performance. . .. .. .. .. .. .. .. .. . . $ . . . . . . . . . . . . . . . . . . . . .......90Issue O-5: Ride Quality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .90Issue O-6: Level of Service . . . . . . . . . . . . . . . . . . . . . . . .......93Issue o-7: Reliability . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......99Issue O-8: Maintainabili ty . . . . . . . . . . . . . . . . . . . . . . . .0. . . . .105

ix

Page

C o s t . . . . ......o . . . . . . . . . 0 . 0 0 . . 0 0 . 0 + 0 0 . . 0 . 0 . 0 . 0 . . , o , * . , * , . . . 1 0 9Issue O-9: Capital Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110Issue O-lo: Operational Cost. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113Issue O-11: Workforce Reduction. . . . . . . . . . . . . . . . . . . . . . . . . 115Issue O-12: Workforce Distribution, . . . . . . . . . . . . . . . . . . . . . . 116Issue O-13: Energy Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120

H u m a n F a c t o r s . . . . . . . . . . . . . . . . , , . , , , . . . . . , , . , ,,, ,,, ....,,,,. 1 2 1Issue O-14: The Human Role . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122Issue O-15: Effects of Automation on

Employees and Passengers, . . . . . . . . . . . ,..,.. 127

CHAPTER 6: THE PLANNING AND DEVELOPMENT PROCESS . . . 133Introduction . . . . . . . . . . . . . . . ., .,. .., .,.O,O, .., ,O. ,,, ,O,,,, ,,, 135Issues

Issue D-1: Design Concepts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137Issue D-2: System Development. . . . . . . . . . . . . . . . . . . . . . . . . 139Issue D-3: Procurement Specifications . . . . . . . . . . . . . . . . . . . 141Issue D-4: Specification of Reliability,

Maintainability, and Availability .,..... . . . . . 142Issue D-5: Equipment Suppliers. . . . . . . . . . . . . . . . . . . . . . . . . 144Issue D-6: Contractor Selection.. . . . . . . . . . . . . . . . . . . . . . . . 145Issue D-7: Contract Management., . . . . . . . . . . . . . . . . . . . . . . 147Issue D-8: Testing, ...,,.,.,,.,., . . . . . . . . . . . . . . . . . . . . . . 148Issue D-9: R&D Programs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150Issue D-lo: Test Tracks..,,,,.,.,., . . . . . . . . . . . . . . . . . . . . . 151Issue D-11: Research Needs .,, , .,, , ...,,..,, . . . . . . . . . . . . 153

CHAPTER 7: POLICY AND INSTITUTIONAL FACTORS . . . . . . ., , . 157I n t r o d u c t i o n . . . . . . . . . . . . . . . . , , , . . . . , . . . , , 0 . .,,...,.,,,,,, ,,, 1 6 1Summary of Existing Legislation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161Issues

Issue P-l: Impact of Existing Legislation . . . . . . . . . . . . . . . . . . 163Issue P-2: Regulation . ., .,, .,,..,, . . . . . . . . . . . . . . . . . . . . . . 166Issue P-3: Acceptance Testing and Qalification . . . . . . . . . . . 168Issue P-4 : S tandard iza t ion , ,,..,. ., ., .. .,, ... OO. ,,O,,., . . 169Issue P-5: Safety Assurance ...., . . . . . . . . . . . . . . . . . . . . . . . . 171Issue P-6: Public Expectations . . . . . . . . . . . . . . . . . . . . . . . . . . . 172

APPENDICESA. Tra in Cont ro l Sys tem Func t ions . ,. .., ,. .., .. .$.....,, ..,,. 177B. Automatic Train Control Technology . . . . . . . . . . . . . . . . . . . . . . 191C. Design Characteristics of Selected Rapid Rail Transit Systems 199D. Glossary of Train Control Terms. , .,, .,, .,.,,.., ,, ..,,,.,, 207E. Chronology of Train Control Development, .,, ,..,... . . . . . . 215F. Persons and Organizations Visited. ., . .,, . ..,,...,. . . . . . . . . 221G. Biographies of Urban Mass Transit Advisory Panel,, .. .,..., 231H. References . ........,.. ,,, ..,,..,,....,.,,,., , ...,,.,,,., 235I Congressional Letters of Request .,,..,,,. , ....,,.,,,..,,., 237

x

TABLES

TABLE 1.TABLE 2.TABLE 3.TABLE 40TABLE 5.TABLE 6.TABLE 7.

TABLE 8.

TABLE 9.TABLE 10.TABLE 11.TABLE 12.

TABLE 13.

TABLE 14.TABLE 15.TABLE 16.TABLE 17.TABLE 18.

TABLE 19.

TABLE 20.

TABLE 21.TABLE 22.

TABLE 23,

TABLE 24.TABLE 25.TABLE 26.TABLE 27.TABLE 28.TABLE 29,TABLE 30.

TABLE 31.

PageLevels of Automation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40BART System Facts. . . . . . . . . . . . . ., . . . . . . . . . . . . . . . 46CTA System Facts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51MBTA System Facts . . . . . . . . . . . . . . . . . . . . . . . ., . . . . 56NYCTA System Facts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61PATCO System Facts . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66Automated Feature of Three Transit Systems Under

Development. . . . . . . . . . . . . . . . . . . . . ., . . . . . . . . . . . 69Train Protect ion Methods in Exist ing and Planned

Transit Systems ... , ., . . . . . . . . . . . . . . . . . . . . . . . . . 78Analysis of CTA Accident Record, 1965-74, ... , , . . . . 80Passenger Injuries Due to Train Operation ., , , ... , , 81Passenger Accident Summary . . . . . . . . . . . . . . . . . . . . . 82Fatalities in the United States by Transportation Mode

During 1973 ...., . . . . . . . . . . . . . . . . . . . . . . . . . . . . ., 83Passenger Assaults and Robberies for Selected Transit

Systems . . . . . . . . . . . . . . . . , . . . , . . . . . . . . . . . .. ....88Train Operation Methods Related to Ride Quality. , ..92Service Characteristics in Typical Transit Systems .. .94Schedule Adherence in PATCO, 1970-74 . . . . . . . . . . . . 96Schedule Adherence on CTA Dan Ryan Line, 1970-74 96Schedule Adherence in BART, August 1973-August

1974. ., . . . . . . . . ● , 00, ..0, , . * ● 0.. ● ● , ... , , , . . . . . . .98

PATCO Car Component Performance, July-October1974...0, .. ..0, . ...00...60, . * . . . . . . . . . . . . . . ● .. 101

Summary of Disabling Equipment Failures in PATCO,August 1973-July 1974 . . . . . . . . . . . . . . . . . . . . . . . . . . 102

Summary of Equipment Failure in BART, 1974 -75.... 102Car Component Per fo rmance on CTA Wes t -South

Route, July-October 1974. . . . . . . . . . . . . . . . , , . . . . . . 104Maintenance Time for Selected PATCO Car Compo-

nents. , . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .106Capital Costs in New Transit Systems . . . . . . . . . . . . . .110Cos t Es t imates fo r ATC Equipment . . . . . . . . . . . . . . . .111Summary of Rail Rapid Transit Operating Costs.. ....113Effect of Automation on Size of Workforce. . ........116Analysis of Operations and Maintenance Workforce. .118Rail Rapid Transit Energy Consumption . . . ... ... ...120Man and Machine Roles in Rail Rapid Transit Systems

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124

Significant Dates in the Engineering Planning, Pro-curement, and Testing of ATC for Various TransitSystems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136

xi

Page

TABLE 32. Type of Specification Used in Recent ATC Procure-ments . . . . . . . . . . . . , . . . ., . . . . . . . . . , , ., . . . . . . . . . 142

TABLE 33. C o n t r a c t o r S e l e c t i o n A p p r o a c h e s . . . . . . . . . . . . . . . . . . 1 4 6TABLE 34. UMTA Budget for Fiscal Year 1976, ,, .,..,. . .......164TABLE A–1. Primary Means of Communicating Information Re-

lated to Train Control., . . . . . . . . . . . . . . . . . . . . . . . . 188

FIGURES

FIGURE 1.FIGURE 2.FIGURE 3.FIGURE 4.FIGURE 5.FIGURE 6.FIGURE 7.

FIGURE 8.FIGURE 9.FIGURE 10.FIGURE 11.FIGURE 12.FIGURE 13.FIGURE 14.FIGURE 15.FIGURE 16.FIGURE 170FIGURE 18.

FIGURE 19.FIGURE 20.

FIGURE 21.FIGURE 22.

FIGURE 23.FIGURE 24.FIGURE 25.FIGURE 26.FIGURE 27.FIGURE 28.

xii

Train Control System Functions, . . . , ., , . . . . . . . . . . . 18Generalized Control Process . . . . . . . . . . . . . . . . . . . . . . 21Simple D.C. Track Circuit . . . . . ... , ., ., , , . . . ... , , .22Three-block, Three-Aspect Wayside Signal System, . 24Cab Signals . . . . . , . . . . . ., , ., , . . . . . . . . . . . , . . . . . . . . 26Three-block, Three-Aspect Cab Signal System. ... , . 26Cab Signal System With Automatic Overspeed Protec-

tion . . . . . . . . . . . , . . . . . . . . . . . . . . . . , . . . ... , . . . . . . 27

Automatic Train Operation System. . . . ... , ... , . . . . 29Typical Interlocking Location. ... , ., . . . . . . . . . . . . . . 30Entrance-Exit Interlocking Control Panel. , ., . . . . . , . 31Model Board and Train Control Console . . . . . . . . . . . 32General View of Rail Rapid Transit Station . . . . . . . . 33Trip Stop . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34Track Circuit Wiring. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34Wayside Equipment Case . . . . . . . . . . . . . . . ... , . . . ., 34Track Apparatus at an Interlocking. . . . . . . . . . . . . . . . 35Central Train Control Facility . . . . . . . . . . . . . . . . . . . . 35T w o V i e w s o f a C e n t r a l C o n t r o l F a c i l i t y W i t h

Electromechanical Equipment. . . . , . . . . . . . . , , . . . . 36Tower for Local Control of Interlocking. . . . . . . . . . . . 36Carborne Receiver Coil for Coded Track Circuit Sig-

nals. . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . 37Train Operator’s Console for System With ATO .. ..37Aerial View of Rail Rapid Transit Yard and Mainte-

nance Facility. , . . . . . ... , . . . . . . . . . . . . . . . . . . . . . . 38Rail Rapid Transit Car Maintenance Shop. ., . . . . . . . 39BART Route Map . . . . . . . . . . . . . . . . . . . . . . . . .......45BART Train in Underground Station ... , . . . .......47Interior of BART Car . . . . . . . . . . . . . . . . . . . . . .......48BART Train Passing Through the Transbay Tube. ..49CTA Route Map. . . . . . . . . . . . . . . . . . . . . . . . . . .......50

FIGURE 29.

FIGURE 30.FIGURE 31.FIGURE 32.FIGURE 33.FIGURE 34.FIGURE 35.FIGURE 36.FIGURE 37.FIGURE 38.FIGURE 39.FIGURE 40.FIGURE 41.FIGURE 42.FIGURE 43.FIGURE 44.FIGURE 45,FIGURE 46.FIGURE 47.FIGURE 48.FIGURE 49.FIGURE 50.FIGURE 51.FIGURE 52.FIGURE 53.FIGURE 54.FIGURE 55.FIGURE 56.FIGURE 57.FIGURE 58.FIGURE 59.FIGURE 60.FIGURE 61.

FIGURE 62.FIGURE 63.FIGURE 64.FIGURE 65.

FIGURE 66.

FIGURE 67.FIGURE 68.FIGURE 69.

PageCTA Train on the West-Northwest Route in the Me-

dian of the Kennedy Expressway. . . . . . . . . .......53Interior View of CTA Train . . . ., . . . . . . . , ... , . . . ..53Lake-Dan Ryan Train Entering the Loop . . . . .......54MBTA Route Map. ... , . . . . . . . . . . . . . . . . . . . . . . . . ..55MBTA Orange Line Train . . . . . . . . . . ... , , ... , . . . ..57Red Line Train Arriving at Wollaston Station. .. ....57The Old and the New. . . . . . . ., . . . . . . . . . . . . .......59NYCTA Route Map. . . . . . . . . . . . . . . . . . . . . . . .......60IND F Train on Elevated Line in Brooklyn , . . . . . . ..63Examples of NYCTA Transit Cars, . . . . . . ., ., ., , .. .63The Graffiti Epidemic. . . . . . . . . . . . . . . . . . . . . . . .. ...64PATCO Route Map. . . . . . . . . . . . . . . . . . . . . . . . . . .. ..65PATCO Train in Lindenwold Yard, . . . . . . . . .......67Outbound on the Benjamin Franklin Bridge . .......68PATCO Train Operator Monitoring ATC Equipment 68WMATA Route Map . . . . . . . . . . . . . . . . . . . . ., ., .. ..71MARTA Route Map. . . . . . . . . . . . . . . . . . . . . . .......72Baltimore MTA Route Map. ., ., , . . . . . . . . . . . . .. ...74Interlocking Control Tower for Train Protection .. ..79CTA Passengers Alighting at Belmont Station .. .....82History of Train Motion Accidents . . . . . . . . . .......83Unmanned Train at Seatt le-Tacoma Airport . . . . . . . .85Fully Automated AIRTRANS Train . . . . . . . . .......86Approach to Brightly Lighted Station . . . . . . . .......89Lonely Station at Off-Peak Hour ., ., . . . . . . . .......89Comfort Features of Modern Transit Cars. . . .......92The Wait and the Rush To Leave . . . . . . . . . . .......95AIRTRANS Availability and Service Interruptions ..97Carbone and Ways ide ATC Equipment . . . . . . . . . . . . 100Maintenance of Transit Vehicle Truck. . . . . . . . . .. ..106Cab Signal Maintenance. . . . . . . . . . . . . . . . . . . . . .. ...107Car Washer. . . . . . . . . . . . . . . . . . . . . . ...............108Inf luence o f Main tenance on Car Ava i l ab i l i ty in

BART, May 1974-January 1975 . . . . . . . . . . . . . . .. ..109Transit System Construction. . . . . . . . . . . . . . . . . . . . . .112Winte r on the Skokie Swif t L ine . . . . . . . . . . . . . . . . . .114Reduction of Train Crew. . . . . . . . . . . . . . . . . . . . . . . . .117Proportion of Operations and Maintenance Employees

in Total Workforce . . . . . . . . . . . . . . . . . . . . . .......118Distribution of Maintenance Force as a Function of

Automation. . . . . . . . . . . . . . . . . . . . . . .............119State-of-the-Art Car . . . . . . . . . . . . . . . . . . . . . . . . . . . , .121Student Conductors Training on the Job. . . . . . . . . . . .122Motorman at Work . . . . . . . . . . . . . . . . . . . . . .........123

,..Xlll

FIGURE 70.FIGURE 71.FIGURE 72.FIGURE 73.

FIGURE A–1.FIGURE A–2.

FIGURE A–3.

FIGURE B–1.FIGURE B–2.FIGURE B–3.FIGURE B–4.

PageLine Supervisors. . . . . . . . . . . . . . . . . . . . . . ...........123Trans i t Sys tem Main tenance Workers . . . . . . . . . . . . . 129D O T T e s t T r a c k , P u e b l o , C o l o . . . . . . . . . . . . . . . . . . . . . 1 5 4ATC Research and Development Priorities and Rela-

tive Cost. . . . . . . . . . . . . . . . . . . . . . ................156Conceptual Diagram of an ATP System, . . . . . . . . . . .178Train Separation in a Conventional Block Signaling

System ... , . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......180Conceptual Diagram of Service Brake Flare-Out Con-

trol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185Simple D.C. Track Circuit . . . . . . . . . . . . . . . . . . . . .. ..192Simple Power-Frequency A.C. Track Circuit . . . . . . .193Simple High-Frequency A.C. Track Circuit . . . . . . . . .194Alternate High-Frequency A.C. Track Circuit . . . . . .195

xiv

. ,

.

-.

.

Chapter 1

FINDINGS

,’ .

\,

1

Definitions 2

Train Control—the process by which the movement of rail rapidtransit vehicles is regulated for the purposes of safety and effi-ciency. The system that accomplishes train control performs fourtypes of functions:

Train Protection-assurance that trains maintain a safefollowing distance, that overspeed is prevented, and thatconflicting movements at junctions, crossings, andswitches are precluded;

Train Operation--control of train movements—specificallyregulating speed, stopping at stations, and opening andclosing doors;

Train Supervision-assignment of routes, dispatch of trains,and maintaining or adjusting schedule;

Communication—interchange of command and status infor-mation among trains, wayside elements, stations, andcentral control.

Automatic Train Control (ATC)—the use of machines to performall or most of the functions of train control in the normal mode ofoperation. Human involvement in ATC systems consists mainly ofmonitoring and back-up. The acronyms ATP (automatic train pro-tection), ATO (automatic train operation), and ATS (automatictrain supervision) denote particular groups of automated func-tions.

Rail Rapid Transit-an electrified rail system operating in urbanareas on exclusive rights-of-way. Rail rapid transit is consideredhere to exclude commuter railroad systems and light rail systems,although the technology of train control is similar for all three.

ZA glossary of train control terms is presented in Appendix D. Explanation of the fundamentals

of train control and descriptions of typical train control equipment are contained in Chapter 3.

INTRODUCTION

In requesting this assessment, the Senate Com-mittee on Appropriations posed four major ques-tions concerning automatic train control tech-nology:

1. What is the state of ATC technology?

Z. What application is made of ATC technologyin existing and planned rail rapid transitsystems?

3. Are the testing programs and methods forATC systems adequate?

4. How is the level of automation selected, andwhat tradeoffs are considered?

These questions served initially as the basicframework for organizing and directing the assess-ment. As the study progressed, it became apparentthat each issue raised by the requesting committeehad many ramifications and that there were corol-lary questions that had to be addressed. Therefore,the study was expanded in scope and detail to con-sider not just the matters enumerated in the letter ofrequest but, more generally, the entire field ofautomation technology in train control systems.The findings of this broader investigation dealingwith policy, planning, and operational concerns aresummarized below. Supporting data and discussionare presented in chapters 5, 6, and 7. At the conclu-sion of this chapter is a brief interpretation of thefindings that responds directly and specifically tothe issues raised by the Senate Committee on Ap-propriations.

POLICY AND INSTITUTIONAL

FACTORS

The development of rail rapid transit systems isinfluenced by three major pieces of Federal legisla-tion: the Urban Mass Transportation Act of 1964,the Department of Transportation Act of 1966, andthe National Mass Transportation Assistance Act of1974. Transit system planning, development, and(since 1975) operation are supported by these actsand the annual appropriations that flow from them.The administrative agency for Federal support oftransit development programs is the Urban MassTransportation Administration (UMTA). Neitherthe existing legislation nor the administrativeprograms of UMTA deal specifically with ATCsystems as such. Research in train control tech-nology and development of individual ATC

systems are carried on within a more generalprogram of activities relating to rail rapid transit asa whole.

Findings pertaining to policy and institutionalconsiderations are as follows:

Regulation

At the Federal level, regulation of rail rapid tran-sit (and ATC specifically) is of recent origin,Regulation is vested in two agencies—UMTA andthe Federal Railway Administration (FRA), whoserespective areas of responsibility are not clearlydefined, It is not surprising, therefore, that so farneither agency has done much to regulate or stand-ardize ATC systems. However, FRA has recentlyindicated the intention to start rulemaking pro-cedures concerning ATP and the safety aspects ofdoor operation.

The National Transportation Safety Board(NTSB) is charged with overseeing rail rapid transitsafety and with accident investigation. Implemen-tation of NTSB recommendations is left to eitherFRA or UMTA or is handled as a matter of volun-tary compliance by transit agencies.

Most regulation of rail rapid transit (and ATCspecifically) is carried out either by State publicutility commissions or by the transit agencies them-selves as self-regulating bodies. The concern ofState regulatory bodies is primarily safety, Little at-tention is given to operational concerns, such asreliability, maintainability, level of service, effi-ciency, and economics.

Advantages in increased Federal regulation, par-ticularly in the areas of safety assurance and equip-ment standardization, must be weighed carefullyagainst the disadvantages of preempting State andlocal authority and raising possible barriers to in-novation.

Institutions

Decisions relating to ATC design and develop-ment are influenced by several nongovernmentalinstitutions or groups. The strongest influence isthat of the local planning or operating authorities,which rely heavily on engineering and technicalconsultants employed to assist in planning anddevelopment activities,

Other institutions and groups acting to shape thecourse of ATC design and development are equip-

72-683 0 - 76 -23

ment manufacturers, industry associations, andorganized labor. Except in isolated cases, only theequipment manufacturers exercise any significantinfluence during the ATC design and developmentprocess. The influence of labor is usually brought tobear only as a new system is being readied foroperation and a contract with the union local isbeing negotiated.

Community planners, public-interest groups, andthe public at large play only a small role in thedesign and development of ATC systems. There issome evidence that these groups may be assumingmore influence, not in technical concerns, but in thearea of establishing priorities and general servicecharacteristics.

Policy Impacts

Federal policy from 1964 to 1974 may havetended to encourage the development of new, tech-nologically advanced transit systems employinghighly automated forms of train control. In part, thispolicy appears to have stemmed from the expecta-tion that automation would lead to increased pro-ductivity-a benefit that, in the case of ATC, hasnot been substantiated. This policy may be in theprocess of change as a result of the National MassTransportation Assistance Act of 1974.3

Transit agencies, when planning new systems,have also been inclined to favor technological ad-vancement-partly as a reflection of how they per-ceived Federal Government policy and partlybecause they or their consultants believed advancedtechnology was necessary to win public support fordevelopment and patronage of the system.

This situation has created a tendency for systemdesigners to turn to highly automated forms of traincontrol as a means of offering improved perform-ance and service. The superiority of automated overmanual methods of train control is not certain,however, except in the area of train protection(ATP).

The cost of automatic train control has negligibleinfluence on the public primarily because it is smallin relation to the total cost of the system (typicallybetween 2 and 5 percent). A question on train con-trol system automation, as a specific issue, hasnever been submitted to the public for decision byreferendum.

THE PLANNING, DEVELOPMENT,

AND TESTING PROCESS

The evolution of a rail rapid transit system fromconcept to start of revenue service may span 10 to20 years. The process has three major phases: plan-ning, engineer ing development , and tes t ing,Research and development to support design areconducted throughout but tend to be concentratedin the middle phase, where detail design anddevelopment takes place. The design and engineer-ing of the train control system, while generally con-current with the development cycle of the wholetransit system, is usually neither the pacing itemnor a dominant technical concern.

Findings concerning the planning, development,and testing process for ATC systems are as follows:

Planning

Formulation of the ATC design concept anddetermination of the extent to which the systemwill be automated are greatly influenced by non-technical factors, notably social and political con-cerns, the prevailing attitude of decisionmakers andsystem designers toward technological innovation,and reaction to the recent experience of other tran-sit agencies.

Cost-benefit analyses conducted during thesystem design process seldom, if ever, includeevaluat ion of al ternat ive ATC concepts anddifferent levels of automation, perhaps becauseATC represents only 2 to 5 percent of total systemcost and benefits are not easily quantified.

The public appears to attach greater importance The comparative operational costs of alternative

to dependability of service and personal securitylevels of ATC are given very little consideration.

than to ATC system performance characteristics.Engineering Development

ATC procurement specifications vary greatly in

.

3The OTA study, An Assessment of Community P]anning terms of approach and level of detail; but the trendfor Urban Mass Transit, February 1976 (Report Nos.OTA–T–16 through OTA–T–27), deals extensively with the

in newer systems is toward a more quantitative

history and current trends of planning and public policy in mass form of specification, particularly for reliability,transit. maintainability, and availability requirements.

4

There is a recognized need in the transit industryfor improvement in the writing of specificationsand in setting realistic requirements for reliability,maintainability, and availability.

In new transit systems, the ATC equipment isprocured as a package through a single contractor.In existing transit systems, ATC equipment is oftenacquired piecemeal as additions or improvements toequipment already in operation.

In most instances, contractor selection is basedon low bid from technically qualified competitors.This procedure is usually required by State law orlocal ordinance. Noncompetitive procurement isseldom used,contract.

Testing

Testing isdevelopment

except for a follow-on to an earlier

conducted at several points in theprocess, generally for one of three

purposes: qualification and validation of compo-nent and subsystem design, assurance of conform-ity to specification, and demonstration of totalsystem performance prior to final acceptance andstart of revenue service.

Performance verification and acceptance testingof train control systems, coming near the end of thedevelopment cycle, may be slighted because ofpressure to open the system for service. The pre-operational test program may be either abbreviatedor deferred until after the start of revenue serviceand often extends into the first year of operation orlonger.

The quality and extent of assurance and ac-ceptance testing vary greatly among transit systems,largely as a function of the qualifications and ex-perience of the organization managing the develop-ment of the system. There is a need for moredetailed and comprehensive test plans, more clearlydefined criteria and methods of measurement, morerigorous procedures for conducting tests, and morecomplete documentation of test findings.

Research and Development

There are no test t racks and experimentalfacilities for carrying out R&D activities related totrain control, except at individual transit systems orat a manufacturer’s plant as part of a productdevelopment program. The Pueblo facility does notpermit detailed study of ATC design and engineer-ing problems in a realistic operational setting.

The state of ATC technology is such that thegreatest R&D need is refinement of existing designsand not development of innovative or more ad-vanced technology. Yet, relatively little R&D effortis concentrated on presently known operationalproblems, such as reliability, maintainability, andavailability, performance testing methods andstandards, and development of a uniform data baseon ATC system performance.

OPERATIONAL EXPERIENCE

No rail rapid transit system now operating orunder development in the United States has a traincontrol system that is completely automatic. Allemploy some mixture of manual and automaticcontrol, and all have at least one person on boardthe train to carry out some control functions. Onlytwo rail rapid transit systems operating in theUnited States at the end of 1975—BART in SanFrancisco and the PATCO Lindenwold Line inPhiladelphia and suburban New Jersey-are auto-mated to the extent that the trainman has little orno direct part in operating the train. In all other U.S.rail rapid transit systems, trains are operatedmanually, with automation employed only for trainprotection and some supervisory functions. Newtransit systems being planned and developed inWashington, Baltimore, and Atlanta show the in-fluence of BART and PATCO with respect to boththe level of automation and the use of advancedATC technology.

A survey of the operational experience withATC leads to the following findings:

Safety

Automatic Train Protection (ATP) systems aresuperior to manual methods of preventing collisionsand de ra i lmen t s , p r inc ipa l l y because ATPsafeguards against human error and inattention.The use of ATP is becoming universal in the U.S.transit industry.

Automatic Train Operation (ATO) offers noclear safety advantages over manual modes ofoperation.

Automatic Train Supervision (ATS) does notproduce additional safety benefits beyond those at-tainable with traditional manual or machine-aidedforms of supervision carried out by dispatchers,towermen, and line supervisors.

5

-.

In conjunction with increased automation, thesize of the train crew is often reduced to one. One-man operation does not appear to have an adverseeffect on passenger security from crime or on pro-tection of equipment from vandalism.

Performance

Under normal operating conditions, the ridequality provided by ATO is comparable to that ofmanually operated trains. The principal advantageof ATO is that it eliminates variation due to the in-dividual operator’s skill and provides a ride of moreuniform quality. Manual operation is considered tobe the more effective mode of control under certainunfavorable weather and track conditions.

Systems with ATC have experienced problemsof schedule adherence during the start-up period,but it is not certain how much of this is a result oftrain control automation and how much is due toother factors such as the complexity and reliabilityof other new items of transit system equipment.

Reliability of ATC equipment has been a majoroperational problem. Failure rates for both waysideand carborne components have been higher thananticipated, but not greater than those of other tran-sit system components of comparable complexityand sophistication (e.g., communications equip-ment, propulsion motors, electrical systems, air-condi t ioning equipment , and door-operat ingmechanisms).

Maintenance of ATC equipment, like otheritems of new technology, has been troublesomebecause of longer repair time, more complicatedtroubleshooting procedures, higher levels of skill re-quired of maintenance personnel, and the lack ofpeople with these skills. A shortage of spare parts.has also hindered maintenance efforts.

On the whole, however, ATC equipment con-tributes proportionally no more to vehicle down-time or service interruptions than other transitsystem components. The problem is that ATC, likeany other new element added to a transit system,has an effect that is cumulative and tends to lowerthe general reliability of the system.

costs

ATC typically accounts for 2 to 5 percent of thecapital cost of rail rapid transit; the variation isalmost directly proportional to the level of automa-tion,

Because of the reduction in train crew that oftenaccompanies ATO and because of the centralizationand consolidation of train supervisors broughtabout by ATS, automated systems are somewhatcheaper to operate than manual systems. These sav-ings are offset, however, by the increased laborcosts of maintaining ATC equipment. In com-parison with manual systems, the maintenanceforce for ATC systems is larger, skill requirementsand the corresponding salary levels are higher,training of technicians must be more extensive andhence costly, and repairs are more frequent andtake longer. The combined operation and mainte-nance costs of automated systems are about thesame as those of manual systems, There is no evi-dence that ATC systems lead to more efficient trainoperation or to any significant change in energyconsumption. Vehicle weight, route layout, andpropulsion system characteristics are far moredominant factors in energy use than automated ormanual operation.

Human Factors

Monotony and light responsibility make itdifficult for operators of highly automated systemsto maintain vigilance. There has also been a tenden-cy for ATC system designers, notably in BART, tomake insufficient use of the human operator toback up or enhance automatic system performance.The designers of systems now under developmentare seeking to integrate the operator more effec-tively into the ATC system, to give man a moremeaningful set of responsibilities, and to makeautomatic equipment more amenable to human in-tervention.

For maintenance employees and train supervi-sion personnel, ATC systems impose new and high-er skill qualifications and more demanding per-formance requirements.

The effect of automation on passengers isnegligible, except insofar as it maybe more difficultfor them to obtain information with fewer transitsystem employees on the train.

ASSESSMENT OF ATC TECHNOLOGY

The following is an analysis and interpretation ofthe findings in light of the concerns expressed in theletter of request from the Senate Committee on Ap-propriations. 4

4Thjs letter and related correspondence are contained in ap-pendix I.

6

The State of ATC Technology

ATC technology is a mature technology insofaras train protection (ATP) and train operation(ATO) funct ions are concerned. The majordifficulties encountered in these areas have arisenfrom the application of new, unproven techniquesthat represent departures from conventional traincontrol system engineering. Train supervision(ATS), except for certain well-established dispatch-ing and routing techniques, is the least advancedarea of ATC technology. Research and develop-ment efforts are now underway to devise computerprograms and control techniques to permit com-prehensive, real-time supervision and direction oftrain movement by automated methods.

Operational experience indicates that automatictrain protection (ATP) enhances the safety of atransit system because it safeguards against colli-sions and derailments more effectively than manualand procedural methods. Performance and servicecharacteristics of ATC systems are as good as, andperhaps better than, manual systems once thesomewhat lengthier period of debugging andsystem shakedown has passed. Reliability andmaintenance continue to be serious problems forsystems using higher levels of ATC and probablyaccount for an increase in operating costs that out-weighs any manpower savings achieved throughautomation.

Application of ATC Technology in New

Systems

In assessing the application of technology in newtransit systems, a distinction must be made betweentrain protection (ATP) and train operation andsupervision (ATO and ATS). All systems-old,new, and planned—rely on automatic devices to ac-complish train protection functions. Two forms oftechnology are employed. One uses wayside signalswith trip stops, the other uses cab signals. The trendin the transit industry today is toward cab signaling,which is the newer technology, because it offerssomewhat more flexible protection than waysidesignaling, and because it provides an evolutionarypath to partially or fully automated train operation.The new systems in Washington, Atlanta, andBaltimore and the recent extensions to existingsystems (e.g., the CTA Dan Ryan extension and theMBTA Red Line) all employ cab signaling and themore automated forms of operation derived from it.

With regard to ATO and ATS, the new systemsunder development and those in the planning stageswill employ more advanced technology and higherlevels of automation than those built and put inoperation before 1969. With some exceptions, suchas door closure or train starting, train operation inthe new systems will be entirely automatic, butsupervised by an on-board operator who will inter-vene in case of emergency or unusual conditions,Central control functions (ATS) will be assisted, orin some cases accomplished entirely, by automaticdevices. Thus, train operation and supervision innew systems will resemble those of PATCO andBART, and the general trend is toward extensiveuse of ATO and ATS.

There is almost no research and developmentnow in progress to produce new ATC technologyfor rail rapid transit. The development work cur-rently underway is devoted primarily to refinementof existing techniques and their application in par-ticular localities. The transit industry has watchedclosely the experience of BART and PATCO. Theresults of the PATCO approach, which made use ofconventional technology, have been compared tothose of BART, where innovative technology andmore extensive automation were employed. Thedesigners of the Washington, Atlanta, andBaltimore systems have generally opted for a mid-dle ground with regard to automation and havefollowed a cautious approach to new technology,inclining more toward PATCO than BART. Par-ticular care has been given to the role of the humanoperator in backing up or augmenting the per-formance of ATO. and ATS equipment. The ex-perience of BART and PATCO has also led thenewer systems to give careful attention to thereliability and maintainability of ATC equipmentand to developing strategies for assuring systemperformance in adverse conditions or degradedmodes of operation. It is certain that WMATA, thenext of the new systems to be put in operation, willbe scrutinized by the transit industry for otherlessons to be learned.

The Testing Process

As train control systems have grown more com-plex, the testing process has been burdened in twoways: there are more elements that must be testedfrom prototype through final installation, and thereare more interrelationships that must be checkedout before the system can be placed in revenueservice. The problem of testing is especially

7

—

difficult in a new transit system, where all theequipment is new and untried and where all theparts need to be tested before initiating passengeroperations.

The experience of BART has underscored boththe basic need for testing and the importance of giv-ing careful attention to test methods, procedures,and documentation of results. The application ofnew technology on a large scale in a transit systeminvolves more than just development and installa-tion of equipment; it also involves the application ofmanagement techniques to integrate the parts of thesystem and to test and evaluate the performance ofthese parts, singly and in the system as a whole.Perhaps the greatest shortcoming in the area of test-ing in the transit industry today is the lack of asatisfactory method for comprehensive evaluationof transit system performance, under realistic con-ditions, in the preoperational period, This is oftencompounded by political, social, and economicpressures to open the system for revenue service assoon as possible, with the result that the testprogram may be truncated or deferred until afteropening day and the full certification of the systemmay not come until months or years later.

The managers of the new systems underdevelopment appear to be mindful of theseproblems. Improved testing methods and pro-cedures are being devised. More complete programsof preoperational testing, even at the expense ofpostponing revenue service, are being planned. Anincremental approach to testing and full systemoperation has been adopted, with each step buildingon the results of earlier phases and with testingtimed to the pace of system growth. Methods oftesting in revenue service, both in regular hours ofoperation and during nighttime periods, are beingexplored. More attention is being given to docu-mentation of test plans and results.

Selecting the Level of Automation

There is no single procedure for selecting thetype of train control system and the level ofautomation. Individual transit authorities followrules of their own devising. Some rely on the adviceof consultants; others draw upon the experience oftheir own technical staff, Only a few generaliza-tions can be made about the nature of this process.

The decisionmaking process does not appear to

be deeply analytical. Criteria of choice are not oftendefined, the rules of choice are not made explicit,and the analysis of alternatives is not documentedexcept in a f ragmentary fashion by internalmemoranda and working papers.

Established transit systems, where extensions ornew lines are being planned, give considerable at-tention to the engineering characteristics of the pro-posed train control system, primarily to assure thatnew ATC equipment can be successfully integratedwith other parts of the existing system, In this case,engineering criteria serve primarily as constraintsupon the type of ATC equipment that can be usedor upon the level of automation to be selected. Theestablished rules and procedures of the transitsystem act in much the same way to limit the choiceof design alternatives. But there is no evidence toindicate that the planning and design process in-cludes studies directed specifically at determiningan optimum train control system or at balancingtrain control system design features against theservice and operating characteristics of other equip-ment or of the transit system as a whole.

In new transit systems, the process for selecting atrain control system is governed even less by systemengineering and trade-off studies. The level ofautomation appears to be selected, more or less ar-bitrarily, early in the system development cycle. Itis treated more as a postulate or a design goal thanas a point for analysis and trade-off. It also appearsthat characteristics of the proposed ATC system arederived more from general, nontechnical decisionsabout the nature of the whole system and its desiredservice features (speed, headways, station spacing,etc.) than from technical considerations of controlsystem design or automation technology.

During the planning process, the developmentand acquisition costs of ATC equipment are con-sidered, but formal cost-benefit studies specific tothe ATC system are usually not conducted. ATCcosts-and, to a lesser extent, benefits—are some-times factored into cost-benefit studies for the tran-sit system as a whole; but the objective of thesestudies is to analyze other aspects of the system orto justify a more general choice regarding transitmode, system size, or route structure. The opera-tional costs of ATC are seldom included in systemcost-benefit studies, and they are not subjected toseparate analysis to determine their potential in-fluence on the life-cycle costs of the transit system,

8

Chapter 2

BACKGROUND

.

9

—.

RAIL RAPID TRANSIT

Rail rapid transits is an old and established partof the national transportation system. It carrieslarge numbers of people at high speeds withincentral business districts and to and from outlyingareas. The patronage in Chicago, for example, isover half a million people on a typical weekday; inNew York City as many as 3-1/2 million riders arecarried daily. Nationwide, rail rapid transit servesabout 2 billion passengers per year. In the newersystems, top speeds of 70–80 miles per hour are at-tained, with average speeds of 30–40 miles per hourfor an entire trip. In cities where there is an existingrail rapid transit system, it is difficult to conceivehow they could function properly, or at all, withoutthis mode of transportation.

Most rail rapid transit systems in this countrywere built over 30 years ago. The New York,Boston, and Chicago systems date from the turn ofthe century. In recent years, other major cities haveturned to rail rapid transit as a solution to theproblems of urban transportation and automobiletraffic congestion. The Lindenwold Line (PATCO)in New Jersey and BART in San Francisco werebuilt within the last 10 years, and rail rapid transitsystems are planned or under construction in Atlan-ta, Baltimore, and Washington, D.C. The majorcities with existing systems (New York, Chicago,Boston, Philadelphia, and Cleveland) have under-taken programs to extend and improve their service.

Along with the new attention to rail rapid transithas come an increased concern with technology.The basic technology of rail rapid transit, whichderives largely from railway engineering, is quiteold. Propulsion and braking systems, for example,are products of the late nineteenth century. Theelectric track circuit, used to detect the presence oftrains and to assure safe separation of trains, wasdeveloped over 100 years ago. The cam controller (amechanism for controlling the application of powerto d.c. propulsion motors) was first used in theChicago subway system in 1914. Cab signalingsystems, functionally similar to those of today, werein use in the 1930’s. While this technology has beenrefined and improved over years of operational ex-perience, many transi t system planners and

5Rai] rapid transit is an electrified rail system operating inurban areas on exclusive rights-of-way, Rail rapid transit is con-sidered here to exclude commuter railroad systems and light railsystems, although the technolog y of train control is similar forall three.

engineers believe that new andforms of technology need to be

more sophisticatedapplied in order to

achieve systems of higher safety, performance, andefficiency,

Generally, two avenues of technological innova-tion are proposed for rail rapid transit: substitutionof electronic for electromechanical components andmore extensive use of automation, One such ap-plication of new technology is in the area of traincontrol, where the replacement of men withelectronic monitoring and control mechanisms isthought to offer several advantages--greater con-sistency of performance, safeguarding againsthuman error, more extensive and precise control oftrain operations, and reduced labor costs in operat-ing the system. However, some transit engineershave misgivings about the ability of the newerautomatic train control systems to perform as safelyand efficiently as manual systems, There is alsosome doubt about the cost-benefit of automation.Automated control systems are more expensive todesign and produce, and their complexity maymake them less reliable and more costly to main-tain. Automatic train control is, thus, a controver-sial matter in rail rapid transit, especially as a resultof the difficulties encountered by the BART systemin San Francisco. BART is the newest and mosttechnologically advanced transit system in theUnited States, but it has not yet lived up to thelevels of performance and service predicted duringits planning and development, or even to the stand-ards set by older and technologically less advancedtransit systems now in operation. Some critics con-tend that problems of BART stem from its extensiveuse of unproven innovative technology for trainoperation and control,

A part of the controversy over automation maystem from a common misconception that it is syn-onymous with computers. Electronic data process-ing is certainly one way to achieve automatic opera-tion, but there are others. The track circuit, theelectromechanical relay, the emergency air brake,the trip stop, and recorded passenger informationannouncements are all automatic devices; and noneinvolves a computer in the usual sense of the term.Another misconception is that automation is some-thing new, a product of aerospace technology.While it is true that automated equipment has beenemployed extensively in advanced aviation andspace systems, the birthplace was certainly notthere. Automation has been with us since the begin-ning of the industrial revolution. All of the

11

————

automaticuse in rail

devices mentioned above have been inrapid transit for many years.

Thus, the issue is not whether automation shouldbe applied in rail rapid transit train control.Automatic train control devices of various typeshave been used in rail rapid transit for many years.The real concerns are where should automation beapplied, how far should the train control process beautomated, and what technology should be used. Asphrased by the OTA staff in planning this assess-ment of automatic train control in rail rapid transit,the central question is: “What degree of systemautomation is technically feasible, economicallyjustifiable, or otherwise appropriate for rail rapidtransit?” The answer, which entails examination ofsafety, performance, and cost, is crucial to thefuture development of rail rapid transit and itsvalue as a public transportation system.

OBJECTIVES

This study was undertaken with the followingobjectives:

1. to examine the design characteristics ofautomatic t rain control systems andevaluate the state of automatic train con-trol technology;

Z. to assess the operating experience andperformance of transit systems whichemploy various forms of automatic traincontrol;

3. to analyze the process by which automatict r a in con t ro l sy s t ems a r e p l anned ,developed, and tested;

4. to examine the policy and institutionalfactors that influence the application ofautomatic train control technology in railrapid transit.

Thus, the emphasis of this report is not on tech-nology as such. While there is considerable atten-tion given to technical matters in the early chapters,it is intended as background for subsequent ex-amination of the results and implications that ensuefrom the application of automation in rail rapidtransit systems. The bulk of this report is devoted toan assessment of the practical results of ATC inoperating transit systems and to the practical resultsof ATC in operating transit systems and to anevaluation of the planning and development proc-

ess by which ATC systems evolve in the context ofpublic institutions and government policy.

SCOPE

The scope of this report is limited to automatictrain control technology in rail rapid transitsystems. No attempt has been made to deal eitherwith rail rapid transit technology as a whole or withthe application of ATC to small-vehicle fixed-guideway systems. 6 The parts of this report thatdeal with the planning and development process areconfined to matters relating to the evolution of thetrain control system. It is recognized that ATCdesign and development does not occur in isolation,but as a part of the larger process by which the en-tire transit system is planned and built. A moregeneral assessment of mass transit planning is thesubject of a separately published report.7

Five operating rail rapid transit systems are ex-amined in this report:

Bay Area Rapid Transit System (BART) in theSan Francisco area,

Chicago Transit Authority (CTA),

Massachusetts Bay Transportation Authority(MBTA) in the Boston area,

New York City Transit Authority (NYCTA),

Port Authority Transit Corporation (PATCO),the Lindenwold Line, in Philadelphia andsuburban New Jersey.

These systems were selected for study because theyembrace a broad range of system characteristics.They vary from a simple one-line system (PATCO)to complex and dense transit networks (CTA andNYCTA). They represent a range of automation,from predominantly manual (NYCTA and CTA) tohighly automated (BART). They differ greatly withrespect to age--NYCTA, MBTA, and CTA beingthe oldest and PATCO and BART the newest. Theyalso employ several forms of train control tech-nology--conventional (CTA, MBTA, NYCTA), ad-vanced (PATCO), and innovative (BART).

eAn assessment of the technology of transit systems employ-ing automatically operated small vehicles on fixed guidewayswas issued by OTA in June 1975 under the title, AutomatedGuideway Transit (Report No. OTA–T-8).

TAn Assessment of Community Planning for Mass Transit,Office of Technology Assessment, February 1976 (Report Nos.OTA-T–16 through OTA-T–27).

12

In addition to these five operating systems,others in the planning and development stage areconsidered in the parts of the report that deal withthe process by which transit systems are conceived,designed, and built. The principal rail rapid transitsystems under development are:

Metropolitan Atlanta Rapid Transit Authority(MARTA)

Mass T rans i t Admin i s t r a t i on (MTA) inBaltimore

W a s h i n g t o n M e t r o p o l i t a n A r e a T r a n s i tAuthority (WMATA)

STUDY METHOD

This assessment was a joint undertaking by theOTA Transportation Program Staff and the UrbanMass Transit Advisory Panel, an 11-memher groupmade up of representatives of the transit industry,State department of transportation, planning con-sultants, organized labor, and public-interestgroups. Battelle Columbus Laboratories acted astechnical consultants and provided major assistancein collecting data and conducting interviews withtransit system officials, planning organizations, andequipment manufacturers. The OTA staff also car-ried out an independent program of visits to inter-view transit system officials at five sites and to col-lect data on their operational experience with ATCequipment. The findings of the Battelle investiga-tion were presented to the panel in a series of back-ground and technical documents. This material wascombined with the results of the OTA staff effort toform the basis for this technology assessment.

ORGANIZATION

This report is organized to accommodate readersof different interests and technical backgrounds,The next two chapters, entitled “Automatic TrainControl” and “Transit System Descriptions,” are in-tended to acquaint the reader with basic train con-trol technology and the operational characteristicsof the rail rapid transit systems selected for study.These chapters are written with a minimum oftechnical detail and provide a general backgroundfor the subsequent examination of operational,planning, and policy issues. Those already familiarwith train control technology and transit operationsmay wish to skim this material or to pass on directlyto chapters 5, 6, and 7, which deal with operationalexperience, planning and development, and policyissues relating to automatic train control tech-nology. As an accommodation to differing readerinterests, these chapters are organized in threelevels of detail. The first level is a summary of themajor issues at the beginning of each chapter. Nextis a presentation of the individual issues, eachheaded by a capsule statement and a synopsis of theprincipal findings and conclusions. The third levelconsists of supporting detail and discussion of theimplications for each issue. Thus, the reader canpursue each topic to whatever depth desired.

At the end of the report are various technical ap-pendices, intended primarily for those who wishmore specific information on train control tech-nology and system engineering features. AppendixD — G l o s s a r y o f T e r m s , a n d A p p e n d i x E —Chronology of Train Control Development, mayalso be of interest to the general reader.

13

.

Chapter 3

AUTOMATIC TRAIN CONTROL

.

15

.—

Train control is the process by which the move-ment of rail rapid transit vehicles is regulated forthe purposes of safety and efficiency. The process iscarried out by a combination of elements-somemen, some machines—located on the train, alongthe track, in stations, and at remote centralfacilities. These elements interact to form a com-mand and control system with four major func-tions:

. Train Protection prevention of col-lisions and derail-ments,

. Train Operation con t ro l o f t r a inm o v e m e n t a n dstopping at sta-tions,

● Train Supervision direction of trainmovement in rela-tion to schedule,

. Communica t ion interchange of in-formation amongthe elements of thesystem.

The train control system is analogous to the sen-sory organs and central nervous system of thehuman body. It senses and processes information,makes decisions, and transmits commands. Also asin the human body, the execution of commands isnot a function of the train control system but ofother parts specialized for that purpose. For exam-ple, the train control system may sense train speed,determine that it should be increased, provide anappropriate command signal to the motors, andmonitor to see that the desired result is achieved.The means by which a speed change is effected,however, are not part of the train control system.All the equipment for getting electric power to thewayside, bringing it into the train, converting it tomechanical energy, and providing tractive effort isexternal to the train control system. Similarly, theequipment to select a route for a particular train andtransmit commands to aline switches accordinglyare within the train control system, but the parts ofthe trackwork that actually move (the switchpoints) are not elements of the train control system.

TRAIN CONTROL

SYSTEM FUNCTIONS

Presented below is a description of the specificfunctions performed by a train control system and

of the way in which functional elements interact.These functional relationships are also illustratedby the diagram in figure 1. Since the purpose is onlyto provide the reader with a general background forunderstanding the nature of train control, thedefinitions presented here are brief and nontechni-cal.8

Train Protection

Train protection is a family of functions whosepurpose is to assure the safety of train movement bypreventing collisions and derailments. 9 rain pro-tection functions and requirements override allother control system functions either throughequipment design or, in a completely manual mode,by rules and procedures. The functions that makeup train protection are:

Train detection—monitoring of the track todetermine the presence and location of trains;

Train separation-assuring that trains on thesame track maintain a safe following distanceto prevent collisions;

Route interlocking—preventing trains on cross-ing, merging, or branching routes from makingconflicting (unsafe) moves that would cause acollision or derailment;

Overspeed protect ion—assuring that t rainspeed remains at or below the commanded orposted civil speed limit10 as to prevent colli-sions resulting from going too fast to stopwithin the available distance and to prevent=

derailments due to excessive speed on curvesor through switches;

Train and track surveillance-observing condi-tions on and in the vicinity of the track aheadof the train and monitoring safety-related con-ditions on board the train.

Train Operation

Train operation consists of those functionsnecessary to move the train and to stop it at stations

F. For more detailed technical descriptions of train controlsystem functions and technology, see appendices A and B.

gThere is no unive~]ly accepted terminology and scheme ofdefinitions for train control system functions within the transitindustry. The terms and classification employed here are basedon several sources and represent the best of current usage.

IOA g]os5ary of train control and rail rapid transit terms is

provided in appendix D.

1 7

PROTECTION

I PERFORMANCEMODIFICATION

k

r I

(TrainPresence)

1

I 1

I L IA

Movement Order

I 1

Movement Report

Operation

OPERATION

J I

“ To simplify the diagram, the functions of Alarmlng and Recordkeepinq are not shown

to board and discharge passengers. Train move-ment, as controlled by train operation functions, isunder the direction of train supervisory functionsand always within the constraints of train protec-tion functions. Train operation involves the follow-ing:

Speed regulation-controlling train speed, with-in the constraints of overspeed protection, tomake the run according to schedule;ll

Station stopping—bringing the train to a stopwithin some specified area in a station;

Door control--opening of doors in stations topermit passengers to enter or leave the trainand closing of doors when the train is ready tostart ; l2

Train starting—initiating train departure from astation after the doors are closed (and pro-vided the train protection system permits it),13

Train Supervision

Train supervision involves monitoring the move-ment of individual trains in relation to schedule androute assignments and overseeing the general dis-position of vehicles and flow of traffic for thesystem as a whole. The train supervision systemmay thus be thought of as making strategic deci-sions which the train operation system carries out

11Speed regulation involves more than matching actual tocommand speed. It also includes control of acceleration, jerklimiting (controlling the rate of change of acceleration), slip-slide control (correction of wheel spinning during accelerationand skidding during braking), and flare-out (gradual relaxationof braking effort as the train comes to a stop). Flare-out is con-sidered by some transit engineers to be a subsidiary function ofspeed regulation, and hence part of the train control system. Ac-celeration control, jerk limiting, and slip-slide control areregarded by transit engineers to be propulsion and brakingsystem functions, but they are mentioned here because of theirrelationship to the train control functions of speed regulationand station stopping.

12The mechanisms that actually open and close doors are notpart of the train control system, but the signals to actuate thesemechanisms and the interlocks to assure that doors are closedbefore starting and that they remain closed while the train is inmotion are generated within the train control system. Because ofthe safety implications of door control, some transit engineersconsider it to be a part of train protection.

13Train starting is sometimes classified as part of the doorcontrol function. It is separated here for two reasons: (1) in someautomated systems, door control is automatic while train start-ing is retained as a manual function; (2) in manual systems, thedoor control and train starting functions are often assigned todifferent persons.

tactically, In addition, train supervision includescertain information processing and recording ac-tivities not directly concerned with train safety andmovement but necessary to the general scheme ofoperations. Train supervision functions are:

Schedule design and implementation—prepar-ing a plan of service in light of expecteddemand, available equipment, and environ-mental conditions and issuing a schedule toimplement the plan;

Route assignment and control--selecting andassigning routes to be followed by trains (andrerouting as necessary);

Train dispatching-controlling train departuresfrom terminals or waypoints in accordancewith the schedule;

P e r f o r m a n c e m o n i t o r i n g — f o l l o w i n g t h eprogress of trains against the schedule by ob-taining periodic updates of train identity, loca-tion, and destination;

Performance modification—adjusting move-ment commands and revising the schedule inresponse to train, traffic, and environmentalconditions.

Alarms and malfunction recording-alerting tomalfunctions, breakdowns, or problems, andrecording their time, location, and nature;

Recordkeeping —maintaining operational logsand records for business and payroll purposes,for scheduling maintenance, for ordering sup-plies and equipment, and for computing tech-nical statistics.

Communication

The communication system is the means bywhich the information needed to carry out all othertrain control functions is transmitted and ex-c h a n g e d . l4 This information may take any ofseveral forms-voice, visual, auditory, and digital

14On the function diagram in figure 1, communication func-tions are indicated by the lines connecting the boxes whichrepresent train protection, operation, and supervision functions,

1 9

or analog electrical signals.l5Unlike other train con-trol functions, which involve information process-ing and decisionmaking, communication is largely afacilitative process-serving to convey informationbut without producing any unique functional out-comes of and by itself. For this reason, thecategorization given below indicates not functionsas such but major classes of information that mustflow throughout the system in order for other traincontrol functions to take place:

Train protect ion—information necessary tolocate individual trains, to assure their safeseparation, to prevent overspeed, and to con-trol movement at route interlockings;l6

Command and s ta tus—information on theoperational state of the system, command sig-nals to control train and switch movement,and feedback to determine the response ofsystem elements to command inputs;l6

Emergency—information on the nature andlocation of emergency events and summonsfor help to elements within the transit systemor to outside agencies (e.g., fire, police, medi-cal, and rescue);

Passenger service—information relating to trainservice and system operation for the purposeof assisting passengers using transit facilities;

Maintenance—information needed to plan orconduct preventive and corrective mainte-nance;

Business operations--operational informationused to maintain a record of (and to plan for)work force allocation, vehicle utilization, pro-curement of supplies and equipment, operat-ing expenses, and system patronage.

15Some transit engineers limit the definition of communica-tion to verbal or visual communication (radio, telephone, TV,and the like). Machine-to-machine communications, since theytend to be very specialized, are considered part of the functionwhich they serve. This seems to be unnecessarily restrictive andmakes an artificial distinction between information exchange byhuman operators and other forms of information exchange in-volved in operating the system (i.e., man to machine or machineto machine). The definition offered here is generic and embracesall types of information flow, regardless of how effected.

16Customarily, this part of the communication system is com-pletely separate from the network used for other types of infor-mation and is considered to be an integral part of the train pro-tection system.

AUTOMATION

At one time or another, all of the train controlfunctions listed above have been performed byhuman operators, and many still are, even in themost technologically advanced transit systems.Theoretically, any of these functions could also beperformed by automatic devices, and more andmore have, in fact, been assigned to machines overthe years, Before examining the technology bywhich train control automation has been achieved,it is first necessary to consider what is meant byautomation and to clarify the terminology used inthis report.

Figure 2 is a generalized diagram of the processby which any train control funct ion is ac-complished. It involves receiving information aboutsome operational state of the system and somedesired state. This information must then be in-terpreted—for example, by comparing the twostates and deriving a quantitative expression of thedifference, Next, an appropriate control response tonull the difference must be selected, and somespecific command message to the controlled ele-ment must be formulated and transmitted. A final,and all-important, step is monitoring the results ofthe control action to ascertain that the desiredsystem state or condition has been achieved. Thislast step, called feedback, provides an input signalto start the process all over again, thereby creating aloop that permits the control process to be con-tinuous and adaptive.l7

If all of the steps in the general sequence shownin Figure 2 are performed by a human operator, theprocess is called manual, even though manual ac-tion in the strict sense may not be involved. Thus,manual denotes a process that may include visual,auditory, and other forms of sensory perception aswell as purely cognitive activities such as in-terpretation, weighing alternatives, and decision-making. The command output might be ac-complished by some manual activity such as press-ing a button or moving a control lever, or it mighttake the form of a voice command or simply a nodof the head. The essential feature of a manual proc-ess, as the term is used here, is that all the basic con-trol steps to accomplish a function are human ac-tivities.

l~his cIescriptiOn overlooks the difference between closed-and open-loop control systems. For a discussion of the applica-tion of each in train control technology, see appendix B.

20

It is also possible for all of the steps in the controlloop to be accomplished by some mechanical orelectrical device. If so, the process is called auto-mated. The device need not necessarily be compli-cated, nor is a computer required in order for the ap-paratus to process information and make a “deci-sion.” A simple junction box with a two-state logiccircuit (ON or OFF) would satisfy the definition ofan automated control device, provided no humanactions were required to receive and interpret inputsignals, select and order a response, and monitor theresult.

Between the extremes of purely manual controland fully automatic control, there are numerouscombinations of mixed man-machine control loops.These are called semi-automated or partially auto-mated—the terms are used synonymously to denotea process (or a system) in which there are bothmanual and automatic elements. Thus, automationis not to be taken in an absolute, all-or-nothingsense. The machine can be introduced by degreesinto a system to perform specific functions or partsof functions. When comparing parts of a train con-trol system or when comparing one system withanother, it is therefore possible to speak of automa-tion in comparative terms and to say that one ismore or less automated than another, depending onhow many specific functions are performed bymachines.

For brevity, acronyms are used to describe cer-tain areas where automation is applied in train con-trol. ATC (automatic train control) refers generallyto the use of machines to accomplish train control

functions. It does not necessarily suggest a com-pletely automated system. It can be applied to asystem where certain functions or groups of func-tions are performed automatically while others areperformed manually. ATP (automatic train protec-tion), ATO (automatic train operation), and ATS(automatic train supervision) are used to designatemajor groups of functions that may be automated.For example, if a system is said to have ATP, itmeans that train protection is accomplished (eithercompletely or mostly) by automatic devices withoutdirect human involvement. If a system is describedas having ATC consisting of ATP and some ATS,this indicates that train protection is assured byautomatic devices and that train supervision is amixture of manual and automatic elements. By im-plication, train operation in such a system would bemanual.

While automation involves the substitution ofmachine for human control, this does not mean thatthe human operator is removed from the systemaltogether. An automated system is not always anunmanned system, even though all functions areroutinely performed by machines. For instance,train protection and train operation may be com-pletely automatic in a given transit system, butthere could still be an operator or attendant onboard the train to oversee equipment operation and,most importantly, to intervene in the event offailure or malfunction. This emergency and backuprole is, in fact, a major type of human involvementin even the most automated train control systems,In all rail rapid transit train control systems now in

2 1

. .

operation or under development, automation isutilized only for normal modes of operation, withmanual backup as the alternative for unusual condi-tions, breakdowns, and emergencies.

In passing, it should also be noted that automa-tion is not synonymous with remote control, eventhough the two may at times go hand in hand. Intrain supervision, for example, many functions areaccomplished manually by controllers who arephysically far removed from the train and wayside.In central control facilities, the operators may neveractually see the vehicles or track and yet perform allor most of the functions necessary to set up routes,dispatch trains, and monitor traffic. Conversely,automated functions are often performed locally,i.e., by devices on board the train or at a station orswitch. In general, the location of the controllingelement in relation to the controlled element is in-dependent of how the functions are accomplished.However, it is also true that automation does facili-tate the process of remote control, and systems witha high level of ATC tend also to employ morecentralized forms of train control, especially forsupervisory functions.

AUTOMATIC TRAIN

CONTROL TECHNOLOGY

The automatic equipment that accomplishestrain control functions is often of complex design,but the basic technology is quite simple. The pur-pose of this section is to provide an acquaintancewith the fundamental e lements of an ATCsystem—track circuits, signaling apparatus, trainoperating devices, interlocking controls, and super-visory equipment, The details of this technologyand the design features of ATC equipment now inuse in rail rapid transit systems are omitted here butare provided in appendices B and C.

Track Circuits

For safety and efficient operation of a transitsystem, it is imperative to know the locations oftrains at all times. The sensing device providing thisinformation is the track circuit, which was inventedover 100 years ago and has remained essentiallyunchanged in principle even though extensivelyrefined and modified in its engineering details.

FIGURE 3.—Simple D.C. Track Circuit

22

The track circuit is an electrical circuit consistingof a power source, the running rails, and a signalr ece ive r ( r e l ay ) .l8 The track is divided intoelectrically isolated segments (called blocks) by in-sulated joints placed at intervals in the runningrails. l9 This forms a circuit with a power sourceconnected to the rails at one end of the block and arelay at the other. The relay, in turn, forms part of asecond electrical circuit which has its own indepen-dent power supply (commonly a battery) and in-cludes a signaling device such as wayside coloredlights,

When no train occupies the block, the relay isenergized by the track circuit battery, causing therelay to “pick up, ” i.e., a movable element (ar-mature) is moved to and held electromagnetically ina position opposed to the force of gravity. Thiscloses an electrical contact in the secondary signalcircuit. When a train enters the block, the wheelsand axles conduct electricity between the runningrails, thereby short circuiting (shunting) the trackcircuit and reducing the current to the relay. Thisweakens the electromagnetic force holding up thearmature, allowing it to drop under the force ofgravity. This action opens the contact that was pre-viously closed and closes a different contact in thesignal circuit. The relay, therefore, acts as a switchin the secondary signal circuit and creates one

electrical path when it picks up and another when itdrops.

Thus, the basic principle of the track circuit is theshunt ing phenomenon produced by the t ra inwheels passing along the electrically energized run-ning rails. The presence of the train is detected inthe track circuit as a reduction of electrical current,which-by means of the relay—is used to controlthe secondary signal circuit and operate varioustypes of track occupancy indicators.

The track circuit is designed according to the fail-safe principle. In order for a clear (unoccupiedblock) indication to be given, the track circuit mustbe in proper working order. If one of the rails were

18Track circuits may utilize one or both running rails, mayoperate on direct or alternating current, and may haveelectromechanical relays or solid-state electronic receivers. Thetype described here is a double-rail dc track circuit with a relay.The other types are similar in principle and operation.

19Block length in rail rapid transit systems varies considera-bly as a function of track and traffic conditions and signalsystem design. Some are as short as 40 feet; others are over half amile long,

to break, the relay would receive no current; andthe armature would drop just as if a train were pres-ent. A broken electrical connection, a failure of thepower source, or a burned-out relay coil would alsohave the same effect.

Wayside Signals

One of the earliest types of signal devicesemployed to control train movement, and one stillwidely used, is the automatic wayside block signal,It consists of a color-light signal, in appearancemuch like the traffic signal on city streets, locatedbeside the track at the entrance to each block, Thissignal is controlled by the track circuit relay, asdescribed above. The signal directs train movementby displaying red, yellow, or green lights (aspects)to indicate track circuit occupancy ahead,

Since it would be impractical for the train tocreep ahead block by block, waiting to be sure eachblock is clear before entering, the wayside signalsare arranged to give the operator advanced indica-tion of speed and stopping commands. Figure 4 is anillustration of a three-block, three-aspect waysidesignal system, This signaling arrangement tells thetrain operator the occupancy of the track threeblocks ahead of the train and conveys threedifferent movement commands (indications)—green (proceed), yellow (proceed prepared to stop atthe next signal), red (stop).

In the illustration, Train A is stopped in Block 4and Train B is approaching from the rear. Sincethere is a separation of at least three blocks betweenthem, Train B receives a green aspect at theentrance to Block 1, allowing it to proceed at themaximum allowable speed. At the entrance toBlock Z, however, Train B receives a yellow aspect,indicating that the train operator should be pre-pared to stop at the next signal because there maybe a train ahead. At the entrance to Block 3, Train Bis commanded to stop by a red signal aspect. WhenTrain A leaves Block 4 and moves on to Blocks 5and 6, the signal at the entrance to Block 3 changesto yellow and then green, allowing Train B to pro-ceed.

The wayside signaling system is made fail-safethrough design and by operating rules. Dual, orsometimes triple, lamps are used to illuminate eachsignal aspect. Redundant power sources are some-times provided. The ultimate safeguard, however, is

23

FIGURE 4.—Three-Block, Three-Aspect Wayside Signal System

procedural. A complete failure of the signal lampsor a loss of power would result in a dark (unlighted)signal, which standard operating rules require thetrain operator to observe as if it were a red signal.

Trip Stops

In the wayside signal system described above,safe train movement depends solely on the com-pliance of the operator with signal indications. Toguard against error, inattention, or incapacitation ofthe train operator, wayside signals can be supple-mented with an automatic stop-enforcing mecha-nism, called a trip stop.

The trip stop is a device located beside the trackat each wayside signal. The type commonly used inthe United States consists of a mechanical arm thatis raised or lowered in response to the track occu-pancy detected by the track circuit. When the arm isin the raised position, it engages a triggering deviceon the train and actuates (trips) the emergency

24

brake.20 A train entering a block in violation of thewayside signal indication would thus be brought toa complete stop before colliding with the train inthe next block regardless of what action the trainoperator took, or

In addition tosions, trip stops

failed to take.

protecting against rear-end colli-can also be used in conjunction

with the track circuits and other signal appliances toprovide automatic protection against overspeed. Forthis application, a timing device is added to the cir-cuit controlling the trip stop. When a train enters a

20An alternative system employing inductive train stops is

used on main-line railroads in the United States and on railrapid transit systems abroad. The device is somewhat more com-plex than the mechanical trip stop, but it avoids mechanical con-tact between a stationary wayside element and a moving trainand is less vulnerable to blockage by snow or debris. Both tripstops and inductive train stops have the inherent disadvantageof requiring strict alinement of wayside devices. Further, ifeither type of device is removed, the system will operate in amode that is not fail-safe.

block, the trip stop at the entrance to the next blockis in the raised position but will be lowered after atime interval corresponding to the minimum time(the maximum speed) permitted for a train to tra-verse the block. This arrangement is commonlyused on curves, downgrades, and other such sec-tions of track where excessive speed could cause aderailment. A variation of this scheme is commonlyused at stations to allow a following train to close inon a leading train, provided the follower moves atappropriately diminishing speed as it approaches itsleader.

Like track circuits and signals, the trip stop isdesigned to operate in a fail-safe manner. The trip israised to the stopping position by gravity or a heavyspring and lowered by a pneumatic or electricmechanism. Thus, failure of the trip stop actuatingmechanism or its source of energy will result in thetrip stop being raised to the stop position.

Cab Signals

Automatic block signal systems with waysidesignals and trip stops, while offering effective trainprotection, have certain operational disadvantages.Sometimes the signals are obscured by fog, rain, orsnow. In such cases, operating rules require that theoperator consider the signal as displaying its mostrestrictive aspect and operate the train accordingly.If the signal is actually displaying a more per-missive indication, time is lost unnecessarily. A sec-ond disadvantage is that wayside signals conveycommands only at the entrance to a block. The trainoperator must reduce speed to the maximum per-mitted by the signal and maintain that speed untilreaching the next signal. If conditions change im-mediately after the train enters the block and itbecomes safe to proceed at a greater speed, the trainoperator has no way of knowing this since the sig-nal is behind him. Again, time is lost. With waysideblock signals there is also the possibility that theoperator will fail to observe the signal correctly,read the wrong signal in multiple-track territory, orforget the indication of the last signal passed. Ifthere are trip stops, these kinds of human failure donot result in an unsafe condition, but the efficiencyof train operation can be adversely affected.

One way to overcome these disadvantages is toprovide signal displays within the cab of the train,This is called cab signaling, A display unit, mountedin the cab within the train operator’s forward field

of view, shows indicator lights similar to those ofwayside signals, e.g., red, yellow, and green aspects.Cab signals can thus convey the same movementcommands as wayside signals, but they do so con-tinuously in response to the instantaneous condi-tion of the track ahead. They can also convey pre-cise speed commands instead of just stop-and-go in-formation, thus providing more flexible operationand paving the way to ATO. The cab signal unit hasan audible warning that sounds whenever the sig-nal aspect becomes more restrictive and continuesto sound until the operator silences it by anacknowledging device. Figure 5 is an illustration ofa typical cab signal.

Transferring the display of information from thewayside to the cab involves an alternate type oftrack circuit technology. To operate cab signals, thecurrent passing through the track circuit (usuallya.c. is not steady, as for conventional wayside sig-nals, but is pulsed (turned on and off) at severaldifferent repetition rates in response to track occu-pancy. Each pulse rate is a code to indicate allowa-ble train speed. This pulsed d.c. energy is passedthrough the rails, picked up inductively by areceiver (antenna) on the train, and decoded toretrieve speed command information, This infor-mation is used to actuate the appropriate cab signaldisplay. Because the train is continuously receivingpulses of energy, a change in the pulse rate of thecoded track circuits indicating a change of condi-tions ahead of the train is instantaneously receivedby carborne equipment and displayed by cab signalsregardless of where the train happens to be within ablock.

Figure 6 illustrates how cab signals control atrain in a three-block, three-aspect signalingsystem. In this example, the code rates transmittedthrough the rails (expressed as pulses per minute)correspond to the following signal aspects:

180 Green (Proceed)

75 Yellow (Proceed at medium speedprepared to stop)

O Red (stop)21

21Note that O code-the absence of a code—is the mostrestrictive, Thus, any failure of the track circuit or the carbornereceiver is a fail-safe condition since it is interpreted by the cabsignal equipment as a command to stop.

25

HOW IT WORKS: Receiver coils, mounted on the train near the rails, receive pulse-coded track signals, whichare decoded and used to pick up relays that energize the cab signal lamp indicating trackconditions ahead.

FIGURE 5.-Cab Signals

FIGURE 6.—Three-Block, Three-Aspect Cab Signal System

26

The situation depicted here is the same as in the il-lustration of wayside signals (figure 4). Train B isapproaching Train A, which is completely stopped,Note that the moment Train A starts to move andclears the block, Train B receives a green signal im-mediately—not at the entrance to the next block, asit would with wayside signals. Note also that a Ocode appears in the part of the block immediatelybehind Train B as it moves along the track and thatTrain B can approachrequired to stop,

Speed Control

closer to Train A before being

With the addition of speed sensing and brakecontrol mechanisms, cab signals can also be used toprovide automatic overspeed protection. Figure 7 isa schematic diagram of such a system. It is the sameas the schematic shown in figure 5, except for theaddition of speed and code rate comparison equip-ment and the direct connections to the propulsionand braking systems.

This arrangement allows the train operator tocontrol speed so long as it does not exceed the com-manded speed shown on the cab signal unit. If thecommanded speed is exceeded or if the block speedchanges to a lower value because of another trainahead, the operator receives an audible warning.The operator has a fixed time (typically 2 to 3 sec-onds) to initiate the required braking manually. Ifthis is done, the brakes can be released when thecommanded lower speed is reached. If not, thebrakes are applied automatically and irrevocably bythe ATC system, and the train is brought to a fullstop before the operator can resume control. This isanalogous to the overspeed control provided bywayside signals with trip stops, except that brakingcan be initiated anywhere within a block not just atthe entrance. Another difference is that trip stopsact to stop the train after an overspeed condition hasoccurred over a measured course, usually severalhundred feet in length. Cab signals do the same, butinstantaneously, thus eliminating the delay in-herent in the preliminary measured course and per-

Train Wheels

/

& Axle

FIGURE 7.—Cab Signal System With Automatic Overspeed Protection

27

—

mit trains to follow one another more closely for agiven block length.

Automatic Train Operation

Basically cab signaling provides carborneautomatic train protection in the form of collisionprevention. With the addition of on-board equip-merit for sensing and comparing command (allowa-ble) and actual speed, cab signaling makes it possi-ble to expand the train protection function to permitspeed regulation. This, in turn, forms the basis forextending automation into the area of train opera-tion.

Several forms of automatic train operation(ATO) are possible, but all have two basicfeatures-automatic speed regulation and stationstopping.

Automatic speed regulation (ASR), as the nameimplies, is basically a comparator circuit for match-ing actual speed to command speed. Speed comandsreceived from coded track circuits are picked up bya carborne receiver, decoded, and compared to ac-tual train speed sensed by a tachometer in the drivemechanism. Up to this point, an automatic speedregulation system is like cab signaling. Thedifference arises in how this comparison is used.With cab signals, the comparison is used to actuatea penalty brake application to stop the train whenactual speed exceeds command speed. With ASR,the comparison is used to control the motors andbrakes in an effort to minimize the difference be-tween actual and command speed. An advisory dis-play of speed commands and train speed may beprovided for the operator. In effect, ASR removesthe human operator from the control loop for run-ning the train and provides for an essentially instan-taneous and invariant response by propulsion andbraking systems, without the delay of human reac-tion time and without the variability and possibilityfor misinterpretation inherent in manual trainoperation.

The other basic element of ATO is station stop-ping, which involves bringing the train to stopautomatically at a predetermined location in eachstation. This is accomplished by special waysidecontrol units working in cooperation with positionreceivers, logic circuits, and automatic speedregulation equipment on the train. One methoduses wayside “triggers” spaced some distance fromthe station as reference points for programed stop-ping. The first trigger, farthest from the station,

transmits a command signal thatboard the train, a velocity-distance

generates, onprofile which

the train is to follow to a stop. Additional triggers,nearer the station platform, correct the generatedvelocity-distance profile for the effects of wheelslip and slide. The ASR system monitors thevelocity-distance profile and controls the brakingeffort to bring the train to a stop at a predeterminedpoint. Another method of programed stoppingmakes use of long wayside antenna to provide aseries of position signals to a carborne controlsystem as the train passes along its length. The car-borne control system determines train position andcombines this with speed and deceleration informa-tion (sensed on board the train), to produce an ap-propriate propulsion or braking command for thetraction control system.

To this basic ATO system, other automatedfeatures may be added. Doors can be openedautomatically after the train is brought to a stop in astation, This requires a circuit to actuate door open-ing mechanisms and appropriate safety interlocks toassure that the train is in fact stopped and at a sta-tion. Door closure may also be automated by addinga timing circuit to measure how long the doors havebeen open and to initiate a door closure signalautomatically after a predetermined dwell time haselapsed. Train departure can also be initiatedautomatically by introducing another control circuitto apply propulsion power after receipt of a signalconfirming that doors are closed and locked.

For each of these levels of ATO, the train opera-tor may be provided with an advisory display toshow what commands are being received and whatresponse is being made by automatic mechanisms.The operator may also be provided with manualoverride controls to inhibit automatic functions orto vary automatic system operation. For example,the operator may intervene manually to adjust thestopping point, to prevent some or all doors fromopening, to vary station dwell time, or to initiate orprevent departure. Figure 8 shows a functionaldiagram of a typical ATO system and a picture ofthe train operator’s console.

Interlocking

An interlocking is an arrangement of signals andsignal appliances so interconnected that functionsmust succeed each other in a predetermined se-quence, thus permitting safe train movements alonga selected route without collision or derailment. An

28

FIGURE 8.—Automatic Train Operation System

29

FIGURE 9.—Typical Interlocking Location

interlocking thus consists of more than justswitches to allow trains to move along crossing,merging or branching routes; it is also made up ofsignals and control devices that automatically pre-vent conflicting or improper movements. Interlock-ing may be manually controlled or equipped withautomatic devices that sort trains through branchesand junctions according to desired destinations,

Several forms of automatic interlocking are inuse. One of the oldest and simplest is an arrange-ment of hand-operated switches, each of whichcontrols an individual signal or track turnout. Theswitches are mechanically or electrically intercon-nected such that once a particular route is selected,the switch points locked in place, and the signalscleared, no other route for a potentially conflictingmove can be established until the train bound forthe cleared route has safely passed, This arrange-ment represents a semiautomated form of move-ment control, Manual operation is required to selecta route and move the control levers, but all else

follows automatically, including inhibition offurther switch movement until the train has tra-versed the limits of the interlocking.

A more advanced, but still not completely auto-mated, type of interlocking is a system that permitsa towerman or central supervisor to select theentrance and exit points for a train to pass throughan interlocking, with the switches and signals fort h e a p p r o p r i a t e r o u t e t h e n b e i n g s e t u pautomatically by an arrangement of electricalrelays, Figure 10 shows such a control panel for asystem called entrance-exit route interlocking, Thetower operator moves the control knobs to desig-nate a desired route. Internal logic circuitsautomatically select the best available nonconflict-ing route, aline and lock switches, and activate theappropriate wayside signals to allowing train move-ment while holding other signals at stop to preventconflicting moves. This level of automation may becharacterized as automatic execution in response tomanual inputs,

30

FIGURE 10.—Entrance-Exit Interlocking Control Panel

Fully automatic interlocking are also in use. Inaddition to track circuits, switch operation, and sig-nal control elements, the automatic interlockingmust have some device for identifying a specifictrain in order to create the necessary input to thelogic circuits.22 One method to identify trains is bymeans of wayside optical device that scans a panelon the lead car which gives destination, route, andother needed information. Another method makesuse of a carborne transponder that is interrogated bya wayside device. With either technique, however,train identity becomes the substitute for manual in-puts that allows trains to be sent along predeter-mined routes without human involvement.

22A rudimentary form of automatic interlocking is one thatuses a simple in-out logic circuit to switch trains from one trackto another. This device is commonly used at terminals and oper-ates to switch each entering train from the inbound to the out-bound track and thus does not require train identity information.

Train Supervision Equipment

Train supervision embraces a wide variety offunctions. The special-purpose equipment that hasbeen developed to perform these functions isequally varied. In a general survey of train controltechnology it is not possible to describe all types ofautomatic and semiautomatic devices that are inuse. The following, therefore, is a brief catalog ofsome of the more important systems.

Train dispatching is concerned with the timing oftrain departures from terminals in accordance withthe schedule of operations. In conventional transitsystems this function is accomplished bypreprogrammed dispatching machines thatautomatically ring a bell or flash a light as a signalto the train operator that it is time to leave a ter-minal or intermediate waypoint. In some systems,the dispatch function may be assigned to a centraltrain control computer that transmits electric start-

31

ing signals to the train in accordance with a masterschedule stored in the computer memory.

Route assignment and control is a train super-visory function that is allied to the train protectionfunction of route interlocking. Route control is astrategic function, consisting of selecting routes fortrains and transmitting the orders to wayside points,where the orders are implemented tactically by in-terlocking equipment. In conventional transitsystems, route assignment and control is performedlocally, either manually or automatically, Withremotely controlled route interlocking, however, itbecomes operationally practical to place thestrategic and tactical management of routing in acomputer. The programing to accomplish this isrelatively simple and straightforward, and a com-puter is ideally suited to handle what is an essen-tially repetitious task with a limited number ofalternative courses of action. The safety aspects ofroute interlocking are assured not by central com-puter control, but locally by conventional interlock-ing equipment at the wayside,

Performance monitoring involves comparing theoverall movement of traffic with the schedule and

taking action to smooth out irregularities of trafficflow. In most transit systems this function is carriedout by central control personnel aided by automaticdisplay devices. One such device is a pen recorderthat marks a moving paper graph to record thepassage of trains past check points, Each spike onthe graph indicates the presence of a train, asdetected by the track circuits, at some time andplace along the route. A train supervisor, by check-ing this graph against the schedule, can monitor theprogress of all trains operating on line and detectdelays or queuing up of trains, (Figure 18, page 36

shows such a device, )

Another form of performance monitoring aid isthe model board (figure 11), which is a schematicrepresentation of the track plan of the transitsystem with indicator lights to denote track circuitoccupancy and, hence, the position of each train onthe line, This is the functional equivalent of thepengraph recorder, but in a more pictorial form ofdisplay. Another type of model board used in newertransit systems has, in addition to the master trackplan, small cathode ray tube displays that permit in-dividual supervisors to obtain more detailed or ex-panded views of selected track sectionsspecial-purpose presentations of data,

or to call up

.

FIGURE Il.—Model Board and Train Control Console

32

Pengraphs, model boards, and the like are notfully automatic supervisory devices. The humanoperator is still needed to interpret the display andto formulate orders to individual trains. In the mostadvanced systems routine performance monitoringis assigned to computers, which keep a continuouswatch on traffic movement and automaticallycalculate and transmit performance commands totrains. Man, in this circumstance, acts in a com-pletely different supervisory capacity. He does notmonitor and regulate traffic. Instead, he supervisesmachines which, in turn, monitor and regulatetraffic.

There are two general types of action that can betaken to smooth out irregularities in traffic flow.Both are accomplished in response to commandsfrom central control. One is to hold a train in a sta-tion for a time longer or shorter than the scheduleddwell time or, in extreme cases, to direct a train tobypass a station in order to close up a gap. The othermethod is to alter the speed of the train betweenstations. This latter method is called performancelevel modification and takes the form of a propor-tional reduction of train speed below the speed nor-mally allowed in each block. In systems supervisedby a central computer and with automatic trainoperation, performance level modification is ac-complished without human intervention. The re-quired reduction is calculated by the central com-puter and automatically transmitted to stations orother critical locations, where the signals are pickedup by carborne ATO equipment that modifies theresponse to the normal speed commands transmit-ted by the coded track circuits. These systems may

also include provisions for manual inputs and dis-plays at central control or on the train, but the nor-mal mode of operation is automatic.

A WALK

THROUGH A TRANSIT SYSTEM

To place ATC in perspective, it maybe helpful tomake a brief tour of the facilities of a transit system,pointing out the type and location of the equipmentthat carries out train control functions.

Station

The passenger’s first point of contact with a tran-sit system is the station. The most prominentfeatures-vending and fare collection facilities(possibly automated), escalators and elevators,heat ing and air condi t ioning, and platformamenities—have nothing to do with train control.There may also be public address systems and videoor audio surveillance equipment for fare collectionand platforms. These are not, strictly speaking, partof the train control facility even though they maybeconnected to the central control facility andmonitored by central supervisory personnel, Aboutthe only direct manifestations of ATC are the auto-mated train departure and destination signs orloudspeakers found in some transit systems. Thesepublic announcement devices are connected to theATC system and use information inputs derivedfrom track circuits and train identification equip-ment, There may be an ATC equipment room in thestation, but it is out of sight and locked. Its presenceis usually unknown to passengers.

FIGURE 12.—General View of Rail Rapid Transit Station

33

—-——

These are the impedance bonds that isolate thetrack into blocks. At the ends of the blocks, thereare small boxes, containing relays, with electricalconnections from the track circuits to the signalingapparatus,

FIGURE 14,—Track Circuit Wiring

Other signal equipment is contained in smallcases placed at intervals along the right-of-way.There are also telephones or other communicationequipment and antennas or transmitters used forprecision station stopping, train identification, orperformance level modification, In certain loca-tions, ATC apparatus and other trackside equip-ment may be housed in small sheds to protect theequipment from the weather and to facilitate main-tenance by wayside workers.

FIGURE 13.—Trip Stop

Wayside

An observant passenger might notice twowayside features that can be seen from the stationplatform. Looking down the tracks in the directionof train movement, there are wayside signal lightsthat change aspect from time to time. Often, justbeyond the downstream end of the platform andalongside the rail, there is a trip stop which can beseen to raise behind a train that has just left the sta-tion and later lower as the train recedes.

Moving out along the tracks, other wayside ele-ments can be found. The track circuits themselvesare not plainly visible since they are largely inwayside housings. However, at intervals there aresmall flat equipment cases situated between therails and connected to them by electrical wiring.

FIGURE 15.—Wayside Equipment Case (contains multiplex

equipment for transmitting information between trains andstation control rooms)

At junctions and crossovers there is switch ap-paratus, the most visible parts of which are theswitch points, frogs, levers, and motive equipment,This is the wayside equipment, known as a switchmachine, that performs the function of interlockingfor train protection,

34

FIGURE 16.—Track Apparatus at an Interlocking

By far the largest par t of the equipment , train but along the wayside and in central controlfacilities, and structures along the right-of-way— facilities.trackage, tunnels, bridges, the third rail, and powerdistribution equipment-are not related to train Central Controlcontrol. Nevertheless, the wayside is where thebulk of the ATC equipment in a transit system is lo- Supervisory control of the system may be exer-cated. The proportion varies as a function of the cised in a central control room equipped with modellevel of automation, but generally about 80 percent boa rds , communica t i on , equ ipmen t , sy s t emor more of all train control equipment is not on the monitoring apparatus, and individual supervisor’s

FIGURE 17.-Central Train Control Facility

35

FIGURE 18.—Two Views of a Central Control Facilitywith Electromechanical Equipment

left-a clock-driven paper tape device for dispatching trains

above-pengraph device for monitoring train movement

consoles. If the system has ATS, the computers andother data processing equipment are also located inthe central control building, which often houses ad-ministrative and training facilities as well.

Not all transit systems have a single centralizedcontrol facility. Some disperse control and supervi-sion to outlying towers, situated at major interlock-ing along the routes. Figure 19 is a photograph ofsuch a local control tower.

FIGURE 19.—Tower for Local Control of Interlocking

36

Vehicles

Most of the ATC equipment on transit vehicles iscarried in equipment cases under the body or in thetrain operator’s cab. About the only featuresdistinguishable from outside the train are a receivercoil mounted on the lead car to pick up coded trackcircuit signals (figure 20) and—for systems with op-tical scanners-small identification panels mountedon the side of each car.

FIGURE 21.—Train Operator’s Console for Systemwith ATO

The operator’s cab contains the displays and con-trols necessary to operate the train or to monitor thefunctions of ATC equipment. The amount andsophistication of this equipment varies greatly—ranging from very simple and utilitarian apparatusin manually operated systems to highly complexconsoles in the newest and most automatedsystems, The console typically includes propulsionand brake controls, a speedometer and commandspeed indicator, lighted placards indicating theoperating state of automatic elements, warninglights, pushbuttons or control knobs to make datainputs or to select various operating modes, a trainphone or radio for communicating with centralsupervisors, a passenger address microphone, and adeadman control to prevent the train from operat-ing in case the operator is inattentive or incapaci-tated.

Yards and Shops

A large part of the important activity of a transitsystem does not occur in revenue service on themain lines, but in the yards and shops, Thesefacilities, though seldom seen by the riding public,contribute greatly to the quality and level of servicethat the transit system offers.

The yards are usually located near terminals andconsist of a vast complex of tracks for storing vehi-cles and making up trains to be operated on thelines. Even in systems with the most advancedlevels of automation, train operation in yards isunder manual control. Train sorting and classifica-tion is also an essentially manual operation,although some systems have a limited amount ofautomatic switching in the yards, principally to andfrom revenue tracks.

Car shops and maintenance facilities are usuallylocated within the yard complex. The shops containfacilities for light and heavy maintenance, compo-nent repair, car washing, and checkout of vehiclesbefore they are dispatched back into service,

The maintenance facility may also include a test

track and special test equipment to qualify vehiclesand components for acceptance or to carry out trialsof equipment modifications.

37

——

FIGURE 22.—Aerial View of Rail Rapid Transit Yard and Maintenance Facility

38

39

——

LEVELS OF AUTOMATIONIt was suggested earlier that train control

automation can be viewed as a continuum. At oneextreme, all functions are performed by humanoperators; at the other, all are performed bymachines. The transit systems now in operation orunder development in this country lie at variouspoints between these extremes, with their relativepositions corresponding roughly to the age of thesystem. The older systems generally have the

LEVEL

Essentially Manual

Wayside Signal Protection

.

Carborne Train Protection

Automatic Train Operation

Automatic Train Supervision

Unmanned Operation

Full Automation

lowest levels of automation—primarily ATP withsome ATS. The newer systems have ATP and ATOand more extensive ATS, None are completelyautomated.

Historically, the conversion from manual toautomatic train control in rail rapid transit has beenincremental and has followed a more or less com-mon course for all systems. These major technologi-cal stops along the road to automation are outlinedbriefly below and summarized in table 1.

TABLE I.—Levels of Automation

CHARACTERISTICS

Train protection by rules and proceduresTrain operation manual (with or without the aid of

advisory wayside signals)Train supervision by towermen and/or central dis-

patched

Wayside block signals with trip stops for trainseparation and overspeed protection

Train operation manualSupervision manual with some automation of dis-

patching and route interlocking

Cab signals and equipment-enforced train protectionTrain operation manualSupervision as above

Automatic Train Protection as aboveTrain operation either completely automatic or with

manual door operation and train startingTrain supervision as above

ATP and ATO as aboveTrain supervision automatic

central computer control

ATP, ATO, ATS as aboveNo on-board operator

(or mostly so) under

System manned only by small number of centralcontrol personnel

ATP, ATO, ATS as above, with automatic, notmanual backups for each

Skeleton force at central controlYard operation automated

EXAMPLE

CTA(Ravenswood andEvanston Lines)

NYCTA

CTA

PATCO

BART

AIRTRANS

None

40

Essentially Manual

At this level, train protection, operation, andsupervision are carried out by train operators andtowermen or central supervisors with little or no aidfrom automatic equipment. Trains are protectedand operated either by rules and procedures aloneor with the aid of advisory wayside signals. Thereare no automatic stop-enforcing mechanisms eitheron the wayside or on board the train. Train dis-patching is carried out by personnel at terminals orat control towers along the routes, using either awritten schedule or timing devices that act asprompters to signal train departure. Route assign-ment and interlocking control are accomplished bymanually activated equipment that may have someautomatic safety features but are entirely controlledby human operators. Communications are by meansof visual signals (lights, hand signals, posted civilspeeds, etc. ) or by telephone from stations andtowers to central control.

Many of the older transit systems in this countrybegan operation at the manual level, but they havesince advanced to more automated forms of traincontrol. One of the last vestiges of a purely manualsystem is on the Ravenswood and Evanston lines ofthe Chicago Transit Authority, which as late as1975 operated without any automatic block signalprotection,

Wayside Train Protection

Wayside signals with trip stops form the basis forautomatic train protection, by assuring separationof following trains and preventing conflictingmoves at interlocking. Incorporation of timingdevices with the trip stops also provides equipment-enforced overspeed prevention. While train protec-tion thus becomes automatic, train operation is stillcompletely manual. Train supervision also remainsan essentially manual activity, although track cir-cuits and signals used primarily for train protectiondo permit some automation of route interlockingand dispatching—usually in the form of semi-automatic devices (i. e., manually activated butautomatically operating).

All t r a n s i t s y s t e m s i n t h e U n i t e d S t a t e s h a v e a t

least this level of automation. The most notable ex-

a m p l e o f a n e n t i r e s y s t e m w i t h e n f o r c e d w a y s i d e

s i g n a l i n g i s t h e N e w Y o r k C i t y T r a n s i t A u t h o r i t y .

P o r t i o n s o f t h e C h i c a g o , B o s t o n , a n d C l e v e l a n d

systems and all of the Philadelphia (SEPTA) systemalso employ this form of automatic train protection.

Carborne Train Protection

Cab signaling, using coded track circuits andautomatic carborne stopping and speed limit en-forcement, represents the same level of ATP aswayside signals with trip stops. To this extent, thislevel of automation is equivalent to the preceding.Generally, however, cab signaling is considered ahigher level of automation since it also providessome automatic aids to train operation—principallyautomatic and continuous display of speed informa-tion to assist the operator in running the train andstopping at stations. Other aspects of train operationare still essentially manual. Cab signaling does notnecessarily lead to any increase in the automationof supervisory function nor is it accompanied byany change in the communications systems.

This level and form of automation is generalIyregarded as the minimum for a new transit system,and most of the older transit systems either haveconverted or plan to convert to cab-signaled ATP.

Automatic Train Operation

The major advantage of cab signaling overwayside signaling is that bringing the speed com-mand on board the train also permits evolution toautomatic train operation. All of the informationneeded to operate the train automatically is eitherinherent in the cab signal system or readily availa-ble through modular additions. At this level, thehuman is removed from the speed control, stationstopping, door control, or starting loops--or anycombination of them. The human no longer func-tions as an operator but as an overseer of carbornecontrol systems.

Along with ATO, there is of ten (but notnecessarily) an increase in the level of automationof train supervisory functions. ATS functions thatare sometimes considered operationally desirable toimplement at the time ATO is installed includeautomatic dispatching, route assignment, and per-formance level modification.

The two newest transit systems in this country—Bay Area Rapid Transit and the Port AuthorityTransit Corporation —both have ATO. The newsystems under development in Washington, Atlan-ta, and Baltimore will also have it.

41

Automatic Train Supervision

Train supervision functions (except for dispatch-ing and route control) are among the last to be auto-mated. To be effective and operationally practical,ATS usually can be introduced only when there is ahigh level of automation in the areas of ATP andATO.23 Automatic train supervision also requires arather complex and sophisticated communicationnetwork, not only for voice messages but also forthe interchange of large quantities of data amongautomatic system elements on a real-time basis.The distinguishing feature of ATS, however, is theuse of a central computer (or computers) to processand handle data, make decisions, and formulate in-structions.

The Bay Area Rapid Transit system was the firstrail rapid transit system to make extensive use ofATS. The new Washington, Atlanta, and Baltimoresystems will also have highly automated trainsupervision based on computer control. While thereare some differences among them in the type andamount of control vested in ATS computers, thesefour stand apart from all other transit systems inthis country in the extent to which automationtechnology is applied to train supervision.

Unmanned Operation

At all the levels of automation described pre-viously, there is at least one operator on board eachtrain and some supervisory personnel in centralcontrol. While these people are not part of the nor-mal control loop, they do exercise important func-tions as overseers of automatic equipment andback-ups in case of failure or emergency; A moreadvanced form of automation is one where thetrains are unmanned, with all ATP and ATO func-tions performed by automatic devices. The few re-maining human operators in the system are atcentral control, but even these personnel may be

23Even with ATP and ATO, ATS is not truly necessary untilthe demands imposed by the complexity of the route structureand the required level of service outstrip the capacity for effec-tive real-time supervision by manual methods. ATS may alsobecome necessary when the load in peak periods approaches 100percent of system capacity.

reduced in number as more supervisory tasks areallocated to machines.

No rail rapid transit system in the United States,or anywhere in the world, is now operating at thislevel of automation. The technology to do this,however, is available; and it has been applied invarious people-mover systems, such as the Morgan-town Personnel Rapid Transit (PRT) and severalairport transportation systems. A notable exampleo f an unmanned a i rpo r t t r an s i t sy s t em i sAIRTRANS at the Dallas-Fort Worth Airport,where small unmanned transit vehicles circulate onfixed guideways over a complex of interconnectingroutes. The entire system is operated and super-vised from a central location by a few persons aidedby a train control computer,

Full Automation

Complete removal of man from control of transitsystem operation-even removing him from thecentral control point—is probably not technicallyfeasible or desirable, For safety and continuity ofoperation, it will always be necessary to have some-one to monitor the system and intervene to restoreoperations or assist passengers in an emergency.The number of such supervisors would be only ahandful, however, and it is doubtful that they couldever conduct normal operations manually as a back-up to automatic systems.

Such a “fully automatic” transit would requirean extremely sophisticated and costly ATC system,which would include ATP, ATO, and ATS for nor-mal modes of operation and—most important—automatic back-ups of these mechanisms for con-tingencies and emergency states, The communica-tion network would also have to be highly sophisti-cated, providing not only voluminous real-time in-terchange among automatic components but alsoextensive two-way voice links between passengersand the supervisory cadre. Another requirement ofsuch a system would be automatic operation,switching, and assembly of trains in yards. Thetechnology for automatic yard operation is availabletoday in rudimentary form in automated freightclassification yards, but it would need to be refinedextensively before application to a rail rapid transitsystem,

42

. . . .

Chapter 4

TRANSIT SYSTEM DESCRIPTIONS

There are eight operating rail rapid transitsystems in the United States and three more in theprocess of planning and construction. Rail rapid

. transit systems are also under consideration in othercities, but none has yet reached the point wherethere is a definite commitment to build a rail rapidtransit system in preference to some other mode ofurban mass transit. Visits were made to these.operating transit properties and planning agenciesduring the course of this study. A list of theorganizations and individuals interviewed is pre-sented in appendix F.

Five operating rail rapid transit systems wereselected for detailed examination. They represent awide range of characteristics and forms of train con-trol technology. They vary from old to new, simpleto complex, and essentially manual to highly auto-mated. This chapter provides a brief description ofthe five operating systems and the three currentlyunder development.

BAY AREA RAPID TRANSIT (BART)

System Characteristics

BART is the newest rail rapid transit system inthe United States, and the most highly automated, Italso serves the largest geographical area of anyoperating rail rapid transit system in the country.As shown in the vignette map above, the BARTroutes form an X-shaped pattern, whose dimen-sions are roughly 26 miles East-West and 30 milesNorth-South.

From the route map it is evident that BARTserves two major purposes: to connect the East Bays u b u r b a n c o m m u n i t i e s w i t h t h e O a k l a n dmetropolis and to link all of these with San Fran-cisco by means of the Transbay Tube under SanFrancisco Bay. The Oakland “Wye,” a junction andswitching complex at the eastern end of the

FIGURE 24.—BART Route Map

45

TABLE 2.—BART System Facts

STATIONS

VEHICLES

ROUTE MILES SurfaceElevatedSubway

NumberAvg. Spacing (mi.)

NumberWeight (tons)

Length (ft.)Capacity (psgrs.)1

Av. Age (yrs.)

CAR MILES (mill./yr.)

25

23

23

7 1

34

2.1

45028.5–29.5

70–75144

2

21.6

TRAIN LENGTH (cars) Max. 10 “Min. 2

SPEED (mph) Max, 80Av. 40

SCHEDULED MINIMUM HEADWAY (min.)26

MANNING No. in Train Crew 1()& M Employees/Car3 42 . 6

PASSENGERS Annual (mill.) 28,8

Av. Weekday (thou.) 125

TRAIN DEPARTURES PER DAY (each way) 280

MAIN LINE TRAIN CONTROL

Train Protection Automatic train separation and overspeed protection with advisory cab) signals, automatic rolling, and in-terlocking control.

Train Operation Automatic speed regulation, station stopping, and door operation

Train Supervision Centralized computer control with centralized manual control and local manual control available as back-upmodes

(1974/75 Data)

I Fll] I (;omplf:mf.nt of sf:tltf:(~ p~lssengf;rs p]lls Stan(]f;(?s in r~asonabl~ (;onfort; crtlsh ]oad is somewhat greater.~Will he retll~ceci to 2 miniltes when system is flllly operational.O&M (operations an{l ma intcnam:e) employees inclu[ie O&M sllperv isors, blit not station, a(imin istrative, engineering, planning

anti ma nag f’merf t pf~ rsonnel.~~;stimat(~(j stat)lf? year staffing.

Transbay Tube, is the engineering feature thatmakes it possible to provide through service, with-out changing trains, between any of the East Baylines and San Francisco.

The BART system consists of 71 miles of double-track routes. ” Approximately one-third of thesystem is underground, one-third on elevated struc-ture, and one-third on fenced surface right-of-waywith no grade crossings.

BART has a total of 34 stations (14 underground,13 elevated, 7 surface), with an average spacing ofslightly over 2 miles.

The BART fleet presently consists of 450 cars,which are of two types: A-cars, containing the

25 The San Francisco Muni line, a light rail system, runs

parallel to the BART line on 4 miles of underground trackbeneath Market Street in San Francisco. While the two sharestations, the Muni system is not part of BART and is operated bya separate transit agency.

operator’s cab and train control electronics, and B-Bears, which cannot operate independently inrevenue service. The non control end of A-cars andboth ends of B-cars are equipped with hostlingpanels to permit individual car movement in theyards and on storage tracks. The basic train make-up (consist) for revenue service is an A-car at eitherend and up to eight B-cars between. Ten-car trainsare run during peak periods, Four- to six-car con-sists are operated in the base period.

The maximum operating speed of trains is 80mph. The average line speed (including stationstops) is about 42 mph, At present, trains operate on6-minute headways through the Transbay Tube andon the San Francisco portion of the system. Head-ways are 12 minutes on the Concord and Fremontfeeder routes and on the through route from Rich-mond to Fremont. When BART reaches its full levelof service, headways will be reduced to 2 minutes inSan Francisco and 6 minutes elsewhere during peakperiods,

46

I n f i s c a l 1 9 7 4 – 7 5 , B A R T

2 8 , 8 m i l l i o n p a s s e n g e r s , f o r

carried an estimateda total of nearly 447

million passenger-miles, Thus, the average lengthof a passenger trip was 15.5 miles, and the averageduration 22 minutes. The average fare per ride wasapproximately 60¢.

ATC Features

Train control in the BART system is highly auto-mated and accomplishes three major functions: (1)overspeed protection, assurance of safe separationbetween trains, and route interlocking control, (2)train operation, including station stops and dooroperation, (3) train supervision, including dispatch-ing, schedule maintenance and adjustment. There isone operator on board the train, regardless of itslength. The normal responsibilities of the operatorare limited to surveillance of the track, monitoringof the train condition, and making passenger an-nouncements. The operator can override certainautomatic train operation functions, such as doorclosure, and can adjust some of the parameters ofautomatic operation, but the operator does not nor-mally intervene in train protection and operationprocesses.

The automated equipment which carries outtrain control functions is partly on board the train,partly at the wayside and in stations, and partly in acentral computer complex. Generally speaking,train protection and operation functions are ac-complished by wayside, station and carborne equip-

ment. Dispatching and schedule maintenance andadjustment are functions of the central computer,wayside equipment, and carborne equipment.

The role of the human operator in BART, eitheron the train or in central control, is intended to belargely supervisory in nature. The operator can alsoexercise certain override and back-up functions inthe event of equipment failure or unusual condi-tions not provided for in the computer programs,Thus, the train operator can always apply emergen-cy braking, keep the train in the station, prevent thedoors from closing, or modify the train performancemode to a more restricted level. The dispatcher atcentral control can manually set and cancel routes,hold trains at stations, order station run-throughs,adjust schedules, insert train identification in thecomputer schedule, and modify train perform-ance—although all of these train supervision func-tions are normally handled by the central computer.

Problems and Issues

The BART system has been the subject of in-tense controversy from the very beginning, Longbefore the first line opened for service in 1972, crit-ics alleged that the system was too costly and toocomplex, partly because of unnecessary sophistica-tion and technological innovation in the train con-trol system design. This complexity and reliance onunproven technology, critics contend, has alsoresulted in a system of lower inherent reliabilityand serviceability that costs more to operate and

FIGURE 25.—BART Train in Underground Station

47

FIGURE 26.—Interior of BART Car

gives poorer service than a system employing con-ventional technology. It is further contended thatthe ATC system is basically unsafe for two reasons.First, there are automated elements that could failand compromise the safety of train operation. Sec-ond, the human operator has been designed out ofthe system to the point where he has no effectivemeans of intervention in such circumstances, ex-cept to bring the train to an emergency stop andthus degrade the performance (and perhaps thesafety) of the system as a whole.

The defenders of BART rebut these charges bypointing out that the complexity is a necessary con-sequence of the high level of performance andsophistication required in the system engineeringspecifications. The design, they contend, was pur-posely innovative because it was necessary to breaknew ground in order to build a viable transportationsystem for a public that has had a long-standingpreference for the automobile. The safety of thesystem is defended in two ways: on theoreticalgrounds, it is asserted that BART has all the fail-safe provisions of the conventional system, but ac-complished in different ways that are not ade-quately appreciated by engineers of traditional traincontrol equipment. On practical grounds, it ispointed out that the BART safety record is com-parable to other transit systems, but operatingdifficulties and accidents in BART receive muchgreater attention because of the public controversysurrounding the system.

Fuel was added to the fire less than a month afterthe inauguration of service when a train ran off theend of the track at the Fremont Station. There wereno fatalities and only minor injuries, but the safetyof the ATC system was opened to serious question.Investigations of BART were undertaken by theCalifornia Senate, the California Legislat iveAnalyst, the California Public Utilities Commission,and the National Transportation Safety Board. Thecause of the accident was traced to a faulty crystaloscillator in the carborne electronics which, byoperating at the wrong frequency, generated toohigh a speed command. This design defect has sincebeen remedied by providing a redundant speed con-trol circuit; but the investigations exposed otherfundamental problems, especially in the train detec-tion system.

As a result, the California Public Utilities Com-mission has issued a series of rulings which willresult in additional tests and demonstrations beforeBART can be placed in full operation. The majorarea under scrutiny is the train detection system.Rail-to-rail shunting through the train axle andwheels, which decreases the signal in the track cir-cuits and thereby indicates the presence of a train,does not always occur to a sufficient degree in theBART system. Also, there are other factors that dis-turb the transmission of track circuit signals andsometimes cause the train detection system to givea false indication of track occupancy. To compen-sate for these faults and to assure positive detection

48

FIGURE 27.—BART Train Passing Through the Transbay Tube

of trains at all times, a logical back-up system hasbeen installed. This involves the use of specialminicomputers at the stations to monitor the out-puts of the primary track circuit detection systemand to clear trains for movement only if certainlogical conditions and criteria are met, These designmodifications are completed and tested but havenot yet been approved by the California PUC.Therefore, the BART system has not yet attainedfull operational status.

As BART has made the transition from designand development to operations, other problemshave emerged. Reliability of equipment, par-ticularly the cars, has been disturbingly low. Mostof the time as much as half of the car fleet is out of

service for repairs. Of the trains dispatched in themorning, only about two-thirds complete the daywithout a breakdown, This has been compoundedby problems of maintenance. Electronic compo-nents take somewhat longer to troubleshoot andrepair and other types of components, and a higherlevel of training and skill is required in mainte-nance technicians. The carborne equipment is noteasily accessible in some cases, requiring more timeto get at the failed component or making it neces-sary to remove one item in order to reach another.Spare parts are in short supply. Often the troublesreported in service are intermittent and cannot beconfirmed or located when the cars reach the yardor shop. The apparently healthy car is then restoredto service, only to fail again in a short time.

49

CHICAGO TRANSIT AUTHORITY (CTA)

System Characteristics

CTA is an integrated rail-bus transit system serv-ing the city of Chicago and 34 suburbs in CookCounty. It is the second largest public transit system

in North America,26 operating a fleet of 2,500 buses

26The largest combined bus-rail system in North America isthe New York City Transit Authority. Considering only the railportion of the system, Chicago is also second only to NYCTA.

FIGURE 28.-CTA Route Map

Table 3.—CTA System Facts

ROUTE MILES SurfaceElevatedsubway

STATIONS

VEHICLES

NumberAvg. Spacing (mi.)

NumberWeight (tons)

Length (ft.)Capacity (psgrs.)1

Av. Age (yrs.)

CAR MILES (mill./yr.)

41

39

10

9 0

142

0.6

1.094

2 0 – 2 4

4 8

75

16

48.9

TRAIN LENGTH (cars) Max.Min.

SPEED (mph) Max.Av.

SCHEDULED MINIMUM HEADWAY (min.)

MANNING No. in Train CrewO&M Employees/Car4

PASSENGERS Annual (mill.)Av. Weekday (thou.)

TRAIN DEPARTURES PER DAY (each way)

81

2 1 / 4

3 2

2.2

129.2512

1,450

MAIN LINE TRAIN CONTROL

Train Protection Mixture of cab signals with automatic overspeed protection and wayside signals with trip stops5

Train Operation Manual operation

Train Supervision Mixture of centralized and local manual control

(1974 Data)

1 Full (complement of seated passengers pllis standees in reasonable comfort; crush load is somewhat greater.~Newer cars are capable of 70 mph hut are governed to 55 mph.One-man train crew on the SkoA ie Swift Line and the Evanston Shuttle during off-peak hours,-r( )& M (()[)(,l.,l t Ions i] nl i m,, in I l~n(l n(:(~) [, mplo;,flf,+ i n(,l II( It> ()& M sllpf’rt’isnrs. I)llt not station. aI imin istra t i t’t’. [~n~ I no”rl” [1x, pi, in o’i n~

,In[{ milnilp,(’ml’nt personn(’t‘)14’ h(’n (; i I rn ‘n t I ;r pr{)g r,l mcl i I nst a I l{] t I on> ,] r(’ (;ompl (’ to i n t h(’ sp r i n~ of 1976.

and 1,100 rail rapid transit vehicles. The rail portionof the system consists of seven lines, of which allbut the Skokie Swift line pass through or circulatewithin the downtown area. Two of the six down-town lines are in subways, entering and leaving bytunnels under the Chicago River. The remainingfour are elevated lines that run on common trackson the Loop El. Access to the loop area is over twobascule bridges, which are raised several timesdaily during the navigation season to permit thepassage of ships. Thus, the throughput for over halfof the CTA system is determined by the volume oftraffic that can be accommodated on the tracks ofthe 75-year-old Loop El structure and its associatedmovable bridges.

CTA operates a total of 90 miles of routes (191.6

track-miles). Almost half (41 miles) are at grade.Elevated routes comprise 39 miles, and subwayroutes 10 miles, There are 142 stations (41 surface,

85 elevated, 16 subway), with an average spacing ofabout two-thirds of a mile,

CTA maintains a fleet of 1,094 cars, consisting offive types, All but four cars used on the SkokieSwift Line are 48 feet in length and of conventional

s t e e l cons t ruc t i on . The i r we igh t i s be tween 40 ,500

a n d 4 7 , 0 0 0 p o u n d s , d e p e n d i n g o n t h e t y p e , T h e

2 0 0 0 - , 2 2 0 0 - , a n d 6 0 0 0 - s e r i e s c a r s a r e o p e r a t e d a s

“ m a r r i e d p a i r s , ” c o n s i s t i n g o f a p e r m a n e n t l y

coupled A-car and B-car . The pairs can be operated

from either end as two-car trains, and they can bejoined with other pairs to form trains of up to eightcars in length. The fourth type of car (the 1 – 5 0

series) is designed to operate as a single and has anoperator’s cab at either end. The 1–50 series carscan also be joined to form trains. The fifth type is athree-compartment, articulated car, of which thereare only four, all assigned to the Skokie Swift line,These cars are about 89 feet long and weigh 93,000

51

72-6, x? () - 76 - j

— . —

pounds. The rolling stock is of varied age. The 6000-series cars are almost 25 years old; the 2200-serieswas acquired in 1969–70, The average age of thecars is about 16 years.

Trains of one to eight cars are run in peak periodson headways that range from 2 to 6 minutes for in-dividual lines. The maximum speed of trains is 50 to58 mph, depending on the type of equipment.Average speed is between 20 and 30 mph. Two fac-tors combine to keep the average speed relativelylow--close station spacing (0.64 mile, average) andthe nature of the right-of-way. Four lines operate,for at least some portion of their route, on the ele-vated tracks of the Loop El. This structure, whichdates from the turn of the century, has extremelysharp turns (90-ft. radius) that must be negotiated atlow speed. The Loop El is also a congested part ofthe system; the four lines using it operate on a com-posite headway of about 1 minute at peak periods.

Including originating passengers and transfersfrom bus routes, the CTA rail rapid transit systemcarried a total of 126.8 million passengers in 1973and an estimated 129.2 million in 1974. The averagelength of a passenger trip is about 7.9 miles or 16minutes (compared to 15.5 miles and 22 minutes inBART). The average passenger fare is roughly 28cents per ride.

ATC Features

The train control system in CTA has undergoneextensive change since the property was acquiredfrom the Chicago Rapid Transit Company in 1947.At that time, trains were operated almost com-pletely under manual control by the motormanusing visual observation and compliance with rulesto regulate speed, station stopping, and followingdistance behind other trains. Color-light waysideblock signals existed over about 10 percent of thetrackage, mainly on curves and in the subways.Wayside signals with trip stops for train protectionwere installed only in the State Street subway(about 10 track-miles). In all other areas, the motor-man had no display of information in the cab or atthe wayside, except signposts advising of speedlimits on curves or downgrades. The train crew con-sisted of a motorman, a conductor, and sufficientguards to man the doors, collect fares, and providepassenger information. Only a few cars had doorcontrols sufficiently sophisticated to permit a train-man to operate the door at the far end of a car, so

that trains required a crew of two to seven men, de-pending on length and type of cars,

Between 1947 and 1960, CTA installed waysidesignals with trip stops in the remaining portions ofthe subway lines and some of their extensions. Theelevated lines in the Loop, however, remained un-signaled; and train control was still essentially amanual operation accomplished by the motorman,with the assistance of towermen at interlocking.

In 1965, CTA began to install cab signaling, firston the Lake portion of what is now the West-Southline and then the new Dan Ryan and Kennedy ex-tensions, which were opened for service in 1969 and1970, respectively. By 1974 the conversion to cabsignaling was completed on the West-Northwestand North-South lines, The remaining lines—Skokie Swift, Ravenswood, and Evanston (includ-ing the Loop El)--are scheduled for conversion inearly 1976, At the completion of the project, about75 percent of the system will be cab signaled, andthe remainder will be protected by stop-enforcingwayside signals.

With the installation of cab signaling, CTA hasgone from the almost completely manual system toa semiautomated form of operation, Train separa-tion and overspeed protection are automatic. Trainoperation is manual, but with machine-aiding of themotorman by means of the cab display unit. Super-vision of trains (schedule maintenance, trafficmonitoring, and routing) are essentially manualoperations accomplished by dispatchers in centralcontrol or by towermen at interlocking, with someremote control and automatic interlocking.

Except for the Skokie Swift and off-peakEvanston shuttle trains, which are manned by asingle operator, all CTA trains have a two-mancrew, The motorman operates the train from thecab and controls all movement. The conductor, sta-tioned at least one car length to the rear of themotorman, controls the opening and closing ofdoors at stations and makes passenger informationannouncements. At certain stations, during off-peak hours when collection booths are closed, theconductor also receives fares.

Thus, the human operator (especially the motor-man) plays an indispensable role in the CTAsystem. Except for train protection and speed limitenforcement performed by wayside or cab signal-ing, the motorman controls the operation of thetrain, The skill with which propulsion and braking

52

FIGURE 30.—Interior View of CTA Train(Note conductor at rear of car.)

are handled determine the smoothness of the ride,the precision of station stops, the adherence toschedule, and the response to incursions on theright-of-way.

Problems and Issues

The basic problems facing CTA are typical of themature rail rapid transit systems in this country.The right-of-way, structures, rolling stock, and sig-nals are in need of modernization or replacement,

There is also a need to expand the service inresponse to the growth and extension of themetropolitan area, Paradoxically, however, thepatronage of CTA has been declining in recentyears. The ridership for the combined bus and railsystem in 1973 was off about 24 percent (about 188million passengers) from that of 1966, a drop ofroughly 3 percent per year.27 The figures for 1974show an upturn (30 million), which may indicate aswitch by the public away from the automobile as aresult of a growing concern with energy usage andconservation of resources. While the revenues fromtransit operations have generally declined, the costshave risen. This has created mounting operatingdeficits, which amounted to $22.1 million for CTAin 1973, despite nearly $37 million in emergencygrants from State, county, and municipal funds.CTA thus finds itself in a position where it must ex-pand and improve the system to meet public needs,but with a shortage of farebox resources to do so.

The conversion to cab signaling was motivatedby more than a desire to modernize the system andthereby attract more patrons. There was also a fun-damental concern with the safety of a system whichoffered only a very limited level of signal protec-tion. Operation of trains on rather close headwaysby means of visual reference and procedural separa-tion created safety problems. CTA has had an inten-

ZTInc]udes originating and transfer passengers.

53

sive safety training and awareness program since1954. While this has resulted in a steady and heart-ening decline of 40 percent in. the traffic andpassenger accident rate over 20 years, the problemsof collisions and derailments persisted, Between1964 and 1974 there were 35 collisions betweentrains which resulted in injuries and 48 derailments(only seven of which produced passenger injuries).This amounted to eight mishaps per year, or oneabout every 6 weeks.

Human operator error was determined to be acausal factor in every collision, Typically, themotorman either failed to observe a train ahead, didnot maintain the proper following distance, or mis-judged the stopping distance. In derailments, abouthalf of the accidents were also caused by human er-ror or improper operation (most commonly switch-ing mishaps or overspeed on curves), The installa-tion of a modern cab signaling system was seen byCTA as the way to prevent these types of accidents.On theoretical grounds, this would appear to be avery effective measure, but it is still too early todraw any firm conclusions from CTA operating ex-perience since the conversion to cab signals.

The cab signaling program has brought with itcertain new operational problems. The reliability ofthe new equipment, particularly during the transi-tional period, has been rather low. CTA engineers

h a v e found that the instal lat ion a n d d e b u g g i n gprocess takes several months: but, when completed,cab signals do not pose an inordinate maintenanceproblem from the point of view of equipmentreliability,

Another aspect of cab signal conversion whichrepresents a problem is in the area of human fac-tors. Installation, checkout, and servicing of theequipment calls for new skills in maintenance per-sonnel . CTA has encountered a shortage ofqualified signal maintainers and has had to under-take an extensive training (and retraining) programfor shop personnel, Train operators, too, have hadto be instructed in the use of the cab equipment, andthere is some anecdotal evidence that the process oflearning to run the train in this new mode of opera-tion is taking longer than expected.

The long-range program for CTA involves twomajor undertakings in the area of train control. Firstis the replacement of the antiquated Loop El with amodern subway system. A part of this project willbe installation of a cab signaling system for all un-derground lines in the downtown area. The secondproject will involve the incorporation of moreautomation in train supervisory and dispatchingfunctions. This includes installation of a modernmodel board in central control and computer aid forschedule maintenance and adjustment.

FIGURE 31.—Lake-Dan Ryan Train Entering the Loop

54

. —

MASSACHUSETTS BAY TRANSPORTATION AUTHORITY (MBTA)

A BROADWAY

FIGURE 32.—MBTA Route Map

System Characteristics

MBTA, serving the metropolitan area of Boston,is one of the oldest rail rapid transit systems in theUnited States. Service on the first line, now a part ofthe Orange Line, began in 1901. MBTA is an inte-grated rail and bus system, the rail portion consist-ing of three rapid transit lines (designated Red,

Orange, and Blue) and a trolley (light rail) lineknown as the Green Line (shown as a dashed line inthe route map). Only the three rail rapid transitlines are considered in this report.

The MBTA lines comprise 30 route miles, ofwhich a little over half (16 miles) are on protectedsurface right-of-way. Of the remainder, 10 miles of

55

TABLE 4.—MBTA System Facts

ROUTE MILES SurfaceElevatedsubway

STATIONS NumberAvg. Spacing (mi.)

VEHICLES NumberWeight (tons)

Length (ft. )Capacity (psgrs.) 1

Av. Age (yrs.)

CAR MILES (mill. /yr.)

164

10

3 0

430.7

354

2 2 – 3 5

4 8 – 7 0

1 2 5 – 2 5 0

18

10.3

TRAIN LENGTH (cars) Max.Min.

SPEED (mph) Max.Av.

SCHEDULED MINIMUM HEADWAY (min.)

MANNING No. in Train CrewO&M Employees/Cars

PASSENGERS Annual (mill.)Av. Weekday (thou.)

TRAIN DEPARTURES PER DAY (each way)

4

2

250320

2 1 / 2

42 – 3

3.0

8 5

283

590

MAIN LINE TRAIN CONTROL

Train Protection Mixture of cab signals with automatic overspeed protection and wayside signals with trip stops

Train Operation Mixture of manual operation and automatic speed regulation

Train Supervision Mixture of centralized and local manual control

(1974 Data)

IFIIII (complement of seated passenge~ plUS stan(]ees in reasonah]e comfort; crllsh load is somewhat greater.

~NwfJr (:,lrs (In th(’ R[’~1 I.inv i)r(’ (:~]pill)lo {If 70 mph I)llt ilr(’ gov~’rn[’l i to 50 mphIAv(’rilg(’ sp~J{II 1 of new (:i] rs on th[’ RPI I 1. i n[~ is iIl)OII t 30 m p h ..lTra in crew ~onslsts of motorman and one train gllar{] (condllctor) for each pair of (;ars.~O& M (operat ions an(] main tenan~e) (~rnplov(les incl I I(]() (l& M supervisors, but not station, a{lministrative, engine~ring, planning

and management personnel.

route are in subways, and 4 are on elevated struc-ture. All trackage in the central business area ofBoston is underground. MBTA has 43 stations (20subway, 17 surface, 6 elevated), with an averagespacing of about 0.7 mile.

A distinguishing feature of the system is the ageand diversity of the rolling stock. Five differenttypes of cars are in operation. The cars on the BlueLinte are oldest. consisting of 40 cars dating from1923 and 48 from 1953. They weight 44,000 and

46,000 pounds, respectively, and are 48 and 49 feetin length. Orange Line cars are 17 years old, weigh58,000 pounds, and have an overall length of 55 feet.The Red Line has the newest equipment--90 so-called “Bluebirds” acquired in 1963 and 76 “Silver-birds” acquired in 1970. Both types are 70 feet long.The Silverbirds weigh 64,000 pounds, and the olderBluebirds 70,000 pounds. All cars are operated as

married pairs in consists of two or four. Some of theRed Line Silverbirds are capable of single-car opera-tion, but they are not so used at the prsent time.

Because there is no connecting trackage andcommon yards and because of varying platformheights and car widths, cars cannot be exchangedbetween lines, In effect, MBTA operates as asystem with three separate parts, linked only bypassenger transfer stations where routes intersect.One consequence of this arrangement is a fleet witha relatively high proportion of reserve cars—about150 in a fleet of 354, or 43 percent.

Another distinguishing feature of MBTA is thecomposition of the train crew which, in addition tothe motorman, is made up of one train guard (con-ductor) for each pair of cars. The rush hour consistof four cars thus requires a crew of three. The train

56

guards are stationed either on a platform betweeneach pair of cars or inside at the rear of each pair ofcars and are responsible for door operation. Theorigin of this manning formula is obscure, but it isreputed to be a safety measure for emergencies orbreakdowns, where the train guards could helpevacuate passengers. It may also be a carryoverfrom the time when sophisticated door operatingequipment was not available, and a pair of cars wasall that one person could handle. Whatever theorigin, this manning formula is now a part of thecontract with the labor union and has not beenchanged even though all MBTA cars are equippedwith doors that can be operated by one man regard-less of train length.

FIGURE 34.—Red Line Train Arriving at Wollaston Station

Depending on the type of car, the maximumdesign train speed is between 30–70 mph—thenewer equipment having the greater top speed,However, because of close station spacing andMassachusetts Department of Public Utilities safety

rules, train speed is governed to 30 mph on the BlueLine, 35 mph on the Orange Line, and 50 mph on theRed Line. Average line speed (including stationstops) is between 20 and 28 mph. Trains are oper-ated on headways of 21/2 to 31/2 minutes in peakperiods and 41/2 to 9 minutes in the base period.

In 1974, MBTA carried a total of 85 millionpassengers. Average weekday patronage was ap-proximately 283,000, including bus and light railtransfer passengers. The typical passenger trip isabout 3.1 miles in length and consumes a little lessthan 8 minutes.

ATC Features

MBTA has only a minimal level of train controlautomation. Most of the system (all but theAndrew-Quincy Center branch of the Red Line) haswayside signals and trip stops for train in separationand automatic interlocking contro1 but no other

ATC features. Since 1971, the Andrew-QuincyCenter (or South Shore) branch of the Red Line has

b e e n e q u i p p e d f o r c a b s i g n a l s . H o w e v e r , t h eMassachusetts Department of Public Utilities hasnot yet authorized cab signal operation because ofquestions as to the safety of the installation, As aninterim measure. Red Line trains are run on a“manual block” system with one-station separationbetween trains. Under this procedure, a followingtrain may not leave a station until a radio messagehas been received from a dispatcher that the leadingtrain has departed,

Train operation (speed regulation, station stops,and door control) is manual, except for Silverbirdcars, which are equipped with automatic speedregulation. There is some machine-aiding of themotorman in running the train, in the form of slip-slide control (for Silverbirds only).

Train supervision is essentially manual, exceptfor automatic train dispatching devices, Trainprogress is monitored by personnel at central con-trol by means of three separate train boards (one foreach line), activated by track circuits. Contact withindividual trains and with wayside and station per-sonnel is maintained from central control by voiceradio. Except for a few locations equipped withautomatic interlocking to control train turnaroundat terminals, all route assignment functions are per-formed manually.

57

Problems and Issues

MBTA is an old system in the process of modern-ization and transition. Rolling stock on the Orangeand Blue Lines is approaching the end of its servicelife and will be replaced with the help of a recentlyreceived $70 million grant from UMTA. Track,way, and structures in older parts of the system arebeing refurbished, and extensions of the lines areunder construction or in the planning stage.28 T h epower generation and distribution system29 is anti-quated and no longer adequate to meet demand. Along-range program of replacement is underway.

Like other parts of MBTA, the signal and traincontrol system is undergoing modernization. Here,the situation is much like that in CTA a few yearsago at the start of their cab signal conversionprogram, There is wayside signal and trip stop pro-tection on most lines and the beginnings of a con-version to cab signaling on two extensions (the RedLine Quincy branch and the Orange Line Wollastonextension). The remainder of the Red Line isscheduled for conversion to cab signaling, and thenew cars for the Orange and Blue Lines will beequipped with cab signal equipment to permiteventual conversion of these lines too,

The Red Line cab signal installation has hadseveral problems, The Massachusetts Departmentof Public Utilities has not yet certified the safety ofthe installation, DPU concern centers in two areas:the reliability of the equipment and the possibilityof incorrect speed commands. Pending DPU ap-proval, the Red Line has been operating under amanual block system (in effect, without cab signals)since 1971,

The operational experience with cab signals hasbeen disappointing. In addition to problems of

ZBThe We]iington extension of the Orange Line opened forservice in September I!li’!i

z9MBTA, unlike other transit systems, still generates much ofits propulsion power (25 Hz a.c.). New lines and most stations,however, run on 60 Hz a.c. power purchased from local utilitycompanies.

r e l i a b i l i t y , t h e r e h a v e b e e n m a i n t e n a n c edifficulties. Shop facilities have not been ade-quate. 30 Spare parts are in short supply. There hasbeen insufficient funding for maintenance work,with the result that not enough repairmen can behired. Cab signal equipment tends to need mainte-nance more often and to require more maintenancetime than other kinds of transit equipment. MBTAmaintenance supervisors estimate that a major partof the maintenance effort is devoted to repairingbreakdowns, with the result that preventive main-tenance and overhaul must be somewhat slighted.

A complicating factor in the maintenance situa-tion is the shortage of qualified maintenance per-sonnel, Union rules permit transportation depart-ment employees (motormen and train guards) withseniority to bid for openings in car shop jobs with-out regard for work skills and experience. Thelimited funding available for maintenance does notallow a complete formal training program for suchpersonnel, who must receive much of their trainingon the job by informal methods. This has notproven to be an effective way to develop the skillsneeded for maintenance of sophisticated electronicequipment.

The problems of MBTA are typical of a system intransition to a new form of technology. Installationand checkout of new equipment disrupts theestablished pattern of operation and maintenance.The new equipment must be integrated with the ex-isting system. Debugging is a troublesome process.Learning to make effective use of the equipmenttakes time and places demands on the labor force toadapt to new procedures and techniques. The entiresystem must find a new equilibrium. MBTA, likeother older transit systems, is finding that the pro-cess of incorporating new technology is not alwayssmooth and trouble-free.

~OMBTA is currently building three modern rail transit main-

tenance facilities, the first two of which (for the Red Line andthe Orange Line extension) were dedicated in 1975,

58

——

I

Tower C, at one time the busiestcontrol and interlocking tower inthe MBTA system, now replacedby a modern automated facility.

Remode ledArlington St.

Station.

Construction of thenew Community CollegeStation on the OrangeLine Extension.(Overhead is InterstateHighway I-93.)

FIGURE 35.—The Old and The New

59

NEW YORK CITY TRANSIT AUTHORITY (NYCTA)

System Characteristics half of the total rail rapid transit track-miles in thecountry. On an average weekday NYCTA carries

NYCTA is the largest and most complex rail more passengers than the total population ofrapid transit system in the United States. NYCTA Chicago. Of the roughly 2 billion rail rapid transithas more route-miles than BART, CTA, MBTA, passengers in the United States each year, half areand PATCO combined; and it has approximately NYCTA patrons. NYCTA has almost 29,500

60

TABLE 5.—NYCTA System Facts

ROUTE MILES SurfaceElevatedSubway

STATIONS

VEHICLES

NumberAvg. Spacing (mi.)

NumberWeight (tons)

Length (ft.)Capacity (psgrs.)2

Av. Age (yrs.)

CAR MILES (mil1./yr.)

23

72

137

2 3 2

4630.5

16,681

34–4351–75

136–20417

320.6

TRAIN LENGTH (cars) Max.Min.

SPEED (mph) Max.Av.

SCHEDULED MINIMUM HEADWAY (min.)

MANNING No. in Train CrewO&M Employees/Car4

PASSENGERS Annual (mill.)Av. Weekday (thou.)

TRAIN DEPARTURES PER DAY (each way)

112

50320

11/2

23.1

1,0!363,740

8,000

MAIN LINE TRAIN CONTROL

Train Protection Wayside signals with trip stops

Train Operation Manual operations

Train Supervision Mixture of centralized and local manual control

I Does not incl(l(]e

zFu1l complement

(1974/75 Data)

754 new R–46 cars now being delivered.of seated passengers plus standees in reasonable comfort; crush load is somewhat greater.

~Loca 1 service; express service averages about 28 mph.40&M (operations and rnaintenanc~) employees include O&M supervisors, but not station, administrative, engineering. planning

and management personnel,‘) The npwer R —44 and R —46 series cars are eq II ipped for automatic speed regulation and programed station stopping in anticipation

of use on planned or new lines and extensions.

employees, 31 not counting the 5,100 transit policewho constitute the eighth largest police force in theUnited States. The annual operating budget forNYCTA in 1974–75 ($951 million) is equivalent to10 percent of that of the entire U.S. Department ofTransportation for FY 1975 and only slightly lessthan the DOT funds budgeted for all of mass transitand railroads in the same period ($965 million).

The complexity and density of the NYCTA net-work can be appreciated by comparing theschematic route map above with those of othersystems. The geographic area served by NYCTA isroughly 15 x 20 miles, which is only slightly largerthan the CTA area but less than half that covered

31 NYCTA also employs about 8,6oo in bus operations,ing a total workforce of 38,066 (43,167 including police).

mak -

by BART. Within this area, however, NYCTA oper-ates 29 routes (26 regular, 3 shuttle) as compared to7 in CTA and 4 in BART. Expressed as the ratio ofroute-miles to area served, NYCTA has 0.77 milesof transit route per square mile; CTA and MBTAhave 0.36; and BART has 0.09. In other words, theNYCTA network is about twice as dense as CTAand MBTA and eight times denser than BART.Density alone, however, does not account for thewhole difference between NYCTA and othersystems since the complexity of the system in-creases exponentially as a function of the numberof lines on common tracks. In NYCTAthe lines in Manhattan and Brooklynwith at least one and as many as three

virtually allshare trackother lines.

The NYCTA system is made up of two operatingdivisions—Division A (the former IRT lines) and

61

Division B (the former BMT and IND lines)32—comprising 232 miles of route. Over half of theroute-miles are in subways (127 miles). There are72 miles of elevated route and 23 miles on protectedsurface right-of-way. NYCTA has 463 stations (265subway, 160 elevated, 38 surface), with an averagespacing of 0.5 mile.33

The fleet consists of 28 different types of cars,ranging in age from R–1 series (1930) to the R–46series acquired in 1975. The newest equipment (300R–44 cars and 754 R–46 cars) is 75 feet in lengthand weighs 80,000 and 86,000 pounds, respectively.Older equipment (the R–38 to R–42 series) is 60feet long and weighs 68,000 to 74,000 pounds. Thereare also about 1,600 51-foot cars acquired in1946–58. The total fleet now numbers 6,681 andwill grow somewhat when - delivery of the R–46series is completed and older equipment is phasedout.

Platform length and operating practice governthe size of the peak period consist, which is eight75-feet cars, ten 60-feet cars, or eleven 51-feet cars.The maximum operating speed of trains is 50 mph.The average line speed (including station stops) is18.5 mph for local service and about 28 mph for ex-press trains. Minimum peak period headway on anindividual line is scheduled at 2 minutes, but thesignal system is designed for 90-second headways.The composite headway at some interlocking maybe as short as 50–60 seconds.

In fiscal year 1973–74, NYCTA carried 1,096million passengers for a total of 5,480 millionpassenger-miles. Average weekday ridership wasabout 3.7 million. Only slightly more than half ofthese riders (53 percent) were carried in the rushhours. 34 This suggests a unique pattern of ridershipfor NYCTA in comparison with other U.S. systems,New Yorkers tend to use the NYCTA throughoutthe day (not just for trips to and from work) and forshort trips. The average trip length is estimated tobe slightly over 5 miles and to take about 17minutes. 35

szIRT—Interborough Rapid Transit , BMT—BrooklynManhattan Transit, IND—Independent.

ssThis is sWcing between local stations. The spacing of ex-press stations is greater, on the order of 1 mile,

sqIn other systems, peak-period ridership customarily ac-counts for about two-thirds of the daily traffic.

sSTrip len@ is also partially a function of the compactness

of the boroughs of Manhattan and Brooklyn where most trips oc-cur.

ATC Features

NYCTA has a relatively low level of automation.Train protection (train separation and interlockingcontrol) is accomplished automatically by waysidesignals with trip stops to prevent block violation.Some portions of the system (principally curves andgrades in the subways) also have time signals andtrip stops for overspeed protection, but elsewherethis function is accomplished manually by themotorman using operating rules and posted civilspeed limits.

Train operation is manual. The crew is two(motorman and conductor), regardless of trainlength. The train is under the control of the motor-man who regulates speed by estimation. (There isno speedometer in the cab except for the new R–44and R–46 cars,) Station stopping and door controlare manual operations-the former by the motor-man, the latter by the conductor.

Except for automatic train dispatching equip-ment, automatic train identity systems, and someautomatic interlocking, t rain supervision ismanual, Scheduling, route assignment (except atau toma t i c i n t e r l ock ing ) , and pe r fo rmancemonitoring are performed by supervisory personnelat central control and by towerman at remote loca-tions. Train supervision is somewhat more de-centralized in NYCTA than in other systems, pri-marily because the size and complexity of thesystem make central control by manual means im-practical. ‘Automated train identification equipmentis used in some locations, but for most of the systemthis function is performed by manual methods.Computer-assisted maintenance scheduling andrecord keeping is employed. Equipment forau toma t i c r eco rded pa s senge r i n fo rma t ionannouncements is installed at some stations, pri-marily major transfer points.

Problems and Issues

NYCTA has embarked upon an ambit iousprogram of modernization and expansion, Morethan 1,800 new cars have been delivered or are onorder. New lines to ease the congestion in heavilytraveled corridors are in the planning stage. Thesenew lines, notably the proposed Second Avenueline, will have cab-signaled ATP and ATO. It is alsoplanned to upgrade train control on existing linesover a 20-year period by converting to cab-signaledATP, Another part of this modernization program,already in progress, is installation of a centralized

62

FIGURE 37.—IND F Train on Elevated Line in Brooklyn

R–16 BMT-IND (1953)

R–36 IRT (1962) R–44 BMT-IND (1970)

FIGURE 38.—Examples of NYCTA Transit Cars

63

communication center for train supervision atNYCTA headquarters in Brooklyn. A new two-waytrain radio and police communications system hasrecently been completed.

Continuing, and worsening, deficits in transitoperations have recently been forced a cutback inthe program. Funds intended for system improve-ment have had to be siphoned off to meet operatingexpenses. The financial crisis of New York City as awhole has also had an impact on NYCTA, forcingeven further curtailments in the planned new tran-sit lines and procurement of replacement equip-ment.

The new R–44 and R–46 ser ies cars areequipped with cab signal units; but since the associ-ated track and wayside equipment has not yet beeninstalled, trains are run with cab signals deacti-vated, relying on wayside signal and trip stop pro-tection. The maintenance and reliability problemsthat have been encountered with the R–44 cars andwith the recently delivered R–46 cars are thus notATC problems, and there is no way of estimatingwhat influence the ATP and ATO equipment ofthese cars may have on car availability.

The gravest maintenance problem for theNYCTA has nothing to do with ATC as such, butdoes influence the ability of the shop force to keeptrain control equipment running. The NYCTA hasbeen stricken with an epidemic of vandalism. Themost obvious form is graffiti, which completelycovers the outside and inside of cars. Officials esti-mate that 95 percent of the cars are defaced on theoutside and 80 percent on the inside. There is alsoextensive breakage of windows, safety equipment,train radios, and motorman consoles. The vandal-ism even extends to yards and track equipment.The Flushing line averages 40 broken windows aday, and 70 or more trains are vandalized (and oftenrendered unserviceable) on the BMT each week,The funds and maintenance force that must becommitted to coping with the damage are of suchmagnitude that other forms of corrective and pre-ventive maintenance suffer.

64

PORT AUTHORITY TRANSIT CORPORATION (PATCO)

System Characteristics

PATCO, also known as theSpeed Line, consists of a single

FIGURE 40.—PATCO Route Map

Lindenwold Hi-route connecting

seven southern New Jersey suburban communitieswith the city of Philadelphia, PATCO is a hybridsystem, resembling a commuter railroad in subur-ban New Jersey and a subway transit system ind o w n t o w n C a m d e n a n d P h i l a d e l p h i a , T h eCamden-Lindenwold segment of the line wasopened for operation in January 1969; through serv-ice to Philadelphia over the Benjamin FranklinBridge began a month later. The line is owned by aNew Jersey -Pennsylvania bi-State agency, theDelaware River Port Authority (DRPA).

Like BART, PATCO was planned and built as analternative to another automobile bridge or tunnelto link the growing suburbs and a central business

area separated by a body of water.36

accumulated in its 6-year historyPATCO has been successful inpatronage of the automobile driver.

The evidencesuggests thatwinning theSurveys have

shown that about 40 percent of PATCO patrons areformer motorists. It has also been established thatPATCO now carries about 30 percent of all dailycommuter trips between South Jersey and center-city Philadelphia. A side benefit is that PATCO hasserved to reduce traffic congestion on parallel high-way arteries. For instance, the average rush hourspeed on White Horse Pike (running alongside thePATCO line) increased by 30 percent from 1960 to

sBUn]ike BART, however, PAT(20 did not involve building aseparate water crossing, PATCO trains run on right-of-way ofthe former Camden-Philadelphia Bridge line on the BenjaminFranklin Bridge.

65

TABLE 6.—PATCO System Facts

ROUTE MILES SurfaceElevatedSubway

STATIONS

VEHICLES

NumberAvg. Spacing (mi.)

NumberWeight (tons)

Length (ft.)Capacity (psgrs.)1

Av. Age (yrs.)

CAR MILES (mill./yr.)

914

14

12

1.2

753968

1206

4.3

TRAIN LENGTH (cars) Max.Min.

SPEED (mph) Max.Av.

SCHEDULED MINIMUM HEADWAY (min.)

MANNING No, in Train CrewO&M Employees/Car2

PASSENGERS Annual (mill.)Av. Weekday (thou.)

TRAIN DEPARTURES PER DAY (each way)

61

75

4 0

2

12.7

11.24 0

182

MAIN LINE TRAIN CONTROL

Train Protection Cab signals with automatic train separation, overspeed protection, and interlocking control

Train Operation Automatic speed regulation and programed station stopping

Train Supervision Centralized manual control

(1974 Data)

I Fll]] ~omplement of seate(i pass~ng~rs plllS stan(~ees in reasonal)le (;omfort: crllsh load is somewhat greater.

zO&M (operations and maintenance) employees include O&M supervisors, hut not station, administrative. engineering, planningand management personnel.

1970, primarily as a result of the start-up of railrapid transit service.

The PATCO line is approximately 14 miles long(9 miles on surface right-of-way or in cuts, 1 mile onelevated structure, and 4 miles of subways inCamden and Philadelphia). There are 12 stations (6elevated or surface and 6 subway), with an averagespacing of 1.2 miles.

The car fleet is made up of 75 vehicles—25 mar-ried pairs and 25 singles. The married pairs aresemipermanently coupled A-cars and B-cars, con-taining one set of train control equipment per pair,and may be operated from either end. The singlesare double-ended cars, capable of independentoperation or of running in trains with other singlesor married pairs. The cars all weigh about 78,000pounds and are 67.5 feet in length. Capacity, withstandees, is about 120 passengers in the A-cars or B-

Bears and slightly less in the singles, because of thetwo operator cabs. Six-car trains are run in peakperiods, two-car trains in base periods, and singlecars nights and Sundays.

The cars are designed to run at 75 mph, a speedwhich is regularly attained on the suburban por-tions of the line. Maximum operating speed on thebridge and in tunnels is considerably lower (15–40mph) because of grades and curvature. The averagespeed for an entire run, including station stops, isabout 38 mph. Trains operate on 2-minute head-ways in peak periods.

In 1974, PATCO carried approximately 11.2million passengers-over 40,000 on an averageweekday, Total passenger-miles amounted toslightly over 95 million, The average trip, therefore,was 8.5 miles in length and took about 131 / 2

minutes. The average fare per passenger was 57cents.

66

ATC Features

The PATCO train control system is a blend ofmanual and automatic operation. The designphilosophy reflects two basic principles. First, thedesign of the system made use of technology that, atthe time, represented the best of available, provenequipment. Technological innovation (and risk) atthe component and subsystem level was held to aminimum. The combination and integration of ele-ments, however, resulted in a system of highly ad-vanced character. Second, the human operator wasto be fully integrated into the system, such that hecould act as a back-up to automated equipment andas the means of enhancing system performance inresponse to unusual conditions.

All train protection (ATP) functions are auto-mated, accomplished by a mixture of carborne,wayside, and track equipment. Train operation(ATO) is also automatic, with two important excep-tions. The single on-board operator (who is theequivalent of a motorman and a conductor) controlsdoor opening and closing. The operator also con-trols train departure by pushing a start button on thecab console. Providing the doors are closed, thismanual action initiates an automatic sequence ofevents in which the train accelerates (withautomatic jerk limiting and slip-slide control), runsto the next station, decelerates, and brakes to a stop.Speed throughout the run is controlled to within +/-2

mph of command speed, and station stopping iswith an accuracy of +/-50 feet.

Although train operation is normally automatic,it is also possible to operate under varying degreesof manual control (within the constraints of over-speed protection). This is often used in bad weatherwhen the rails are slippery, especially on grades.The operator can order the train to bypass a station,without otherwise interfering with the automaticcontrol process. The train can also be run in a com-pletely manual mode (except for ATP). It is a pro-cedural rule of PATCO that each train operatormust run the train manually for an entire trip once aday in order to retain his operating skills. Thus,train control in PATCO can be characterized as anautomatic system under supervision of an on-boardoperator who has the capability for manual inter-vention to compensate for malfunctions and to aug-ment system performance.

In contrast, train supervision (ATS) is largelymanual. PATCO uses dispatchers at a central traincontrol board to oversee train movements. orderschedule adjustments, and monitor overall systemperformance, Routing (switch control at interlock-ing) is automatic, but it can be overridden bycentral control. Communication with the train is bymeans of train phone, which uses the third rail asthe conductor. Police, wayside maintenance per-sonnel, and the Lindenwold car shop are linkedwith central control by a radio network.

FIGURE 41.—PATCO Train in Lindenwold Yard

67

.—— —-— — — --

FIG(JRE 42,—Outbound on Ben Franklin Bridge

PATCO stations are entirely unattended, faresbeing collected by an automatic vending and gatesystem under closed-circuit television surveillance.One or two employees at central control overseestation activities by TV, make public address an-nouncements, and handle calls for assistance frompatrons by direct-line telephones at the fare gates.

Problems and Issues

The PATCO train control system has beensingularly trouble-free, The engineers of the systemattribute this to the design philosophy that madeuse of only proven elements and conventional tech-nology. However, it is also true that the PATCOsystem is relatively simple, consisting of a singleline without merging points and complex interlock-ing. The PATCO approach was not so ambitious asthat of BART, to which it is often compared. Whileit can be said that PATCO accomplished its objec-tives more fully, it should also be noted that lesswas attempted. Still, the PATCO system is an ad-mirable transit system engineering achievement,and it is widely publicized as an example of prudentand effective use of automation.

There appears to be no recurr ng reliability andmaintenance problems associated with the ATCequipment in PATCO, Certain deficiencies ofdesign and manufacturing quality control came tolight during the initial year of operation, faulty wir-ing connections and termina1s being the most prev-alen t. PATCO maintenance Supervisors considerthese to be no more than the usual start-up anddebugging difficulties, even though it did takealmost a year to wring the system out. In general,car availability has been excellent throughout the 6years of operation, The number of cars needed toprovide scheduled peak-hour service has beenavailable 99.2 percent of the time or more each year,although this requires a two-shift maintenance ac-tivity that is not common in the transit industry,ATC equipment has not contributed a dlispropor-tionate share to the overall pattern of equipmentfailures and maintenance time.

In the initial planning of PATCO, it was pro-posed to build a three-branch system in New Jerseywith a common trunk line over the bridge intoPhiladelphia. This plan was dropped in favor of thesingle-line system that was eventually built, Plan-ning is now underway to build the two additionalbranches (to Mount Laurel and Glassboro) and toextend the existing Lindenwold line to Waterford-

FIGURE 43.—PATCO Train OperatorMonitoring ATC Equipment

68

Berlin. This will result in a three-pronged routeplan, very much like the BART system but some-what smaller in scale. The junction of the threebranches, equivalent to the Oakland Wye in BART,is a train control engineering problem of concern toPATCO, Experience with the existing system hasshown that the PATCO ATC system is adequate fora single route. However, the level of automation(especially in the area of ATS) may not be suffi-cient to handle three routes merging and running ona single line over the Benjamin Franklin Bridge. Inorder to maintain the regularity and level of servicenow offered, it may be necessary to install moresophisticated and highly automated equipment tocontrol interlocking and supervise traffic move-ment.

SYSTEMS UNDER DEVELOPMENT

There are three rail rapid transit systems nowunder construction—WMATA (Washington, D.C.),MARTA (Atlanta), and MTA (Baltimore). Of these,WMATA is nearest completion; the first 4.6-milesegment is scheduled to open with limited revenueservice (7 a.m. to 7 p.m.) in the spring of 1976.Ground breaking for MARTA took place in Febru-ary 1975, and initial service is planned for 1978–79,

The Baltimore system is in the advanced planningstage and scheduled for completion in 1981 –82.

All three systems will employ advanced traincontrol technology, at levels of automation in therange between the PATCO and BART systems. Ta-ble 7 is a summary of the ATC features planned or

TABLE 7. Automated Features of Three Transit Systems Under Development

ATC FUNCTIONS

ATP

Train in SeparationOverspeed ProtectionRoute Interlocking

A TO

Velocity Regulation

P r o g r a m m e d S t o p p i n g

Door Control and Train Start-ing

ATS

Dispatching and Monitoring

Performance Level Control

COMMUNICATIONS

Operator- Passengers

Central Control–Passengers

operator-Central Control

WMATA

Fully automaticFully automaticFully automatic

Fully automatic, with alterna-tive of manual operation

Fully automatic, with alterna-tive of manual operation

Fully automatic, controlled byloca l t imer sub jec t t omanual override

Console and display boardsupported by computer

Four levels of runtween stations, w

time be-ith sepa-

rate control of accelerationrate, dwell time, and skip-Stop

One-way PA and noise moni-tor system

One-way PA

Two-way radio phone

MARTA

Fully automaticFully automaticFully automatic

Fully automatic

Fully automatic

Fully automatic

Aided, hut not directly con-trolled, by computer

Computer modification ofspeed, acceleration, anddwell time, with manualoverride

one-way PA

One-way via train PA

Two-way radio phone

MTA

Fully automaticFully automaticFully automatic

Fully automatic

F u l l y a u t o m a t i c

M a n u a l

Centralized traffic controlm a c h i n e a n d a u t o m a t i c

dispatching units

Six levels of speed.train opf’rater into visual signals at stations

set in byresponse

One-way PA

One-way via train PA

Two-way radio phone

69

FIGURE 44.—WMATA Route Map

70

proposed for each. Note that only the WMATAtrain control system is a firm design at this point;MARTA and MTA are tentative and subject tomodification as the system evolves.

W a s h i n g t o n M e t r o p o l i t a n A r e a

T r a n s i t A u t h o r i t y ( W M A T A )

The WMATA Metro System is being built as aseven-phase project, with the last phase scheduledfor completion in 1982. 37 At that time, WMATAwill consist of 98 route-miles. serving 86 stations.There will be 47 route-miles underground, 42 milesat surface, and 9 miles on elevated structure. TheWMATA system will serve the largest geographicarea of any rail rapid transit system in the country(30 miles N-S and 35 miles E-W). However, thedensity of the network (route miles per square mile)will be rather low—about 0.09. which is the same asBART.

The WMATA fleet will be made up of 556 cars,75 feet in length and weighing 72,000 pounds. Carcapacity will be 175 (81 seated and 94 standees).The cars are designed to operate as semiperma-nently coupled A and B units (married pairs) to b erun in consists of two to eight.

The train control system wil l have ful lyautomatic train protection (ATP), including separa-tion assurance, overspeed prevention, and route in-terlocking. The normal mode of train operation willbe automatic (ATO), under the supervision of anon-board operator. Door closure, train starting,velocity regulation, programed station stopping, anddoor opening will be automated functions. Trainoperation will, therefore, be similar to the ATOsystem of BART, except that station dwell time willbe under control of a local timing device inWMATA instead of a BART-like central computer.Unlike the BART system, however, the WMATAtrain operator will have several methods for inter-vening in the automatic operating process either toaugment system performance or compensate forpartial failures. In this regard, the WMATA trainoperation system will be similar to PATCO. Trainsupervision (ATS) will be computer assisted andwill permit either manual or automatic adjustmentof performance level, station stopping, and dwelltime. In general, the WMATA approach to ATC hasbeen to employ proven, existing hardware and ad-vanced, but not revolutionary, technology.

Metropolitan Atlanta Rapid TransitAuthority (MARTA)

MARTA has recently begun construction of a 70-mile system of rail rapid transit integrated withh i g h - s p e e d b u s w a y s , s e r v i n g t h e A t l a n t ametropolitan area in De Kalb and Fulton Counties.The rail portion of the system will consist of ap-proximately 50 route-miles, radiating from down-town Atlanta. The first segment (13.7 miles) is ex-pected to be finished by 1980.

The MARTA fleet will have 200 cars, operatingas married pairs in trains of up to eight. Speeds of upto 75 mph on 2-minute headways are proposed in-itially, with eventual reduction to 90-second head-ways in heavy demand corridors. The train willhave one operator, who will monitor automatictrain control equipment and provide limited manualback-up.

The train control system to be used in MARTA isstill in the early stage of definition; a general func-tional design has been developed, but detailedengineering specifications had not been issued atthe time this report was prepared. With regard toATP and ATO, the MARTA system will be verymuch like BART.38 Train protection and operationwill be fully automatic, the on-board operator serv-ing as a performance monitor. The operator willalso be able to impose modifications of train opera-tion functions. It is envisaged that the operator willact as a back-up to ATO equipment for emergencyand degraded states of operation, but without thecapability of running the train at full performancelevels.

The supervisory functions carried out by centralcontrol will be aided extensively by a computer butwill not be under direct computer control. A uniquefeature of the ATS system design is that it will beimplemented in two stages. The first stage will pro-vide for semiautomatic operation--computer-ex-ecuted routing, dispatching, and monitoring inresponse to manual inputs and override by centralpersonnel. The second stage will provide forautomation of the routing and dispatching functionsand will incorporate an Automatic Line Supervision(ALS) system for computer-controlled trafficregulation (dwell, performance level, schedule ad-justment, reverse running, and stat ion run-through). The implementation strategy is to use the

‘lTThe first 4.8-mile segment was opened for service onMarch 27, 1976,

3cMARTA has engaged the same general engineering Consul-

tant, Parsons Brinkerhoff-Tudor-Bechtel, who designed theBART system.

71

FIGURE 45.—MARTA Route Map

72

Baltimore Region Rapid Transit System—Phase 1

first, semiautomated stage as a baseline to get thesystem into operation and debugged and then toupgrade the central ATS complex to full automa-tion when traffic demand increases. However, thefirst stage will be retained in an operable state, as abackup to automated central control for emergen-cies and nonnormal modes.

Mass Transit Administration ofMaryland (MTA)

MTA in Baltimore is proposing to build a 28-milerail rapid transit system extending from the north-west area of the city through downtown and ter-minating south of the Baltimore-Washington Inter-national Airport. So far, Federal grants have beenadvanced for only the northern half of the system;f u n d i n g f o r t h e r e m a i n d e r i s i n q u e s t i o n .Groundbreaking for construction of the northwestline was held in the fall of 1974.39 Revenue serviceis scheduled to begin in 1981.

The ATC sys t em fo r Ba l t imore ha s no tprogressed much beyond the preliminary designstage, The design concept calls for an automated

~gconcern over the cost of the proposed system led to asuspension of construction activity in the fall of 1975, pending afull review of costs and available sources of funding.

system similar to BART in technology but withmore direct involvement in train operation by anon-board attendant. ATP will be fully automatic, asin WMATA and MARTA, Train operation (ATO)will be automatic under normal conditions, exceptfor door control and train starting, which will bemanually initiated (like PATCO). There will alsobe provision for train operation at full performancelevels in a semiautomated cab signal mode, A novelfeature of the proposed ATC system is that the on-board operator will be able to set the train speedprofile to any of six levels in response to commandsfrom centraI control transmitted by visual signals atthe stations. Train supervision (ATS) will incorpor-ate several automated features, but the generallevel of automation of central control facilities willb e s o m e w h a t l o w e r t h a n t h a t o f W M A T A o r

M A R T A ,

A noteworthy aspect of the Baltimore systemdesign is the requirement that it be compatible withWMATA, thus making it feasible to link up the twosystems at some future t ime if demand andmetropolitan area growth patterns so dictate, Atthis time, however, there is some question in theminds of the designers as to whether compatibilityshould be limited to physical characteristics (suchas clearances, platform height, car size, and tractionvoltage) or whether it should also include the sig-naling and train control system.

74

Chapter 5

OPERATIONAL EXPERIENCE

i1

INTRODUCTION

The advocates of automatic train control ad-vance three general arguments to support theircase-safety, performance, and cost.

An automated system, they contend, has a higherlevel of safety than one in which the basic control-ling element is the human operator, Automaticd e v i c e s f u n c t i o n w i t h a c o n s i s t e n c y a n drepeatability that man simply cannot match. In awell-designed automatic system. hazardous events

are precluded by the engineering of the system; andif an automated device should fail, there are otherdesign features to assure that the system will revertto a condition known to be safe (the “fail-safe”principle). In short, because the behavior ofmachines is predictable, contingencies can be fore-seen and compensated for in the design. The humanoperator, by contrast, is not as predictable. Man isprone to errors of judgment, inattention, fatigue,and other frailties. Furthermore, the human opera-tor takes longer to process information and to re-spond, with the danger that he may not do so cor-rectly. And so, the argument runs, the automateddevice should be preferred over the human becauseit leads to a system of greater inherent safety.

The second argument is that an automated traincontrol system leads to superior performance. Here,the argument rests on the superiority of machineover human capabilities, Automated devices workrapidly, with greater precision, and in a manner al-ways consistent with the objectives of the system.In the case of computers, they have a recognized ad-vantage over man in their ability to process, store,and retrieve large amounts of information and toapply this information in the solution of complexproblems. Thus, an automated train control systemcan move traffic at higher speeds and on closerheadways; and-equally important—it can makerapid compensations and adjustments in responseto changing conditions.

Automated train control systems are alsoasserted to be less expensive than manual systemsin the long run. The initial capital costs of an auto-mated system are admittedly higher, s implybecause there is more equipment to design andbuild, It is claimed, however, that these costs aremore than offset by the reduced operating expensesof an automated system. Automated systems arecheaper to run because they have fewer operators,and it is labor costs that represent the bulk of

operating expense, Automation can also produceother savings. An automated system is claimed tobe more economical in its energy use because theequipment is operated at optimal speeds and ac-celeration-decelcration profiles. This leads to a Sec-ond form of economy. less wear and tear on theequipment due to improper operation. Finally, theoptimum mode of operation brought about byautomation supposedly leads to a more efficientsystem, making it possible to provide the sameamount of passenger service with less rolling stock.

All of these assertions about the safety, perform-ance, and cost advantages of automated systems aresubject to question. The purpose here, however, isnot to enter into debate. Instead, the arguments ad-vanced for automation will be treated as hy-potheses, to be tested by the empirical evidence andoperating experience of transit systems wherevarious automated control features are in use. Theaim is to look at the operational record to see ifthere are differences among transit systems whichare attributable to the level of automation. The dis-cussion is presented as a series of propositions orissues, grouped under the general headings ofsafety, performance, and cost. As a corollary, an ex-amination is also made of the role and effectivenessof man in systems with different levels of automa-tion.

SAFETY

Safety has two aspects. There is the immediatequestion of passenger accidents and injuries whichmay be attributable to some aspect of automatedtrain control. There is also the question of the in-herent safety of the system, i.e., the extent to whichthe design of the system helps prevent accidents.The first question has to do with the narrower,historical concern of whether accidents have oc-curred, while the second deals with the larger topicof safeguards incorporated in the design againstpossible future accidents.

Allied to these questions is the matter ofpassenger security. Automated systems, with fewertransit property employees on board the trains andin the stations, might be assumed to offer thepassenger less protection from assault, robbery, andother criminal actions. This point needs to be ex-

amined first because of its implications for publicsafety and, second, because of its influence on thedecision to replace humans with automated devicesin other, future, transit systems.

77

ISSUE O-1: TRAIN PROTECTION

Are automatic train protection (ATP) devicesmore effect ive, and inherently safer , thanmanual train protection methods?

The experience of the transit industry indi-cates clearly that ATP provides a surer methodof train protection, and all new systems nowunde r deve lopmen t w i l l emp loy ATP inpreference to manual means.

Train protection involves three basic controlfunctions: train separation, overspeed protection,and route interlocking. In a manual system, thesefunctions are performed by the train operator whomaintains visual observation of the track ahead andruns the train in conformance with establishedrules and procedures. When these functions areautomated, there are mechanical devices andelectrical circuits at the wayside and on the train it-self to assure that proper following distance ismaintained (train separation), that train speed doesnot exceed that required for safe stopping ornegotiating curves (overspeed protection), and that

conflicting moves along the lines or throughswitches are prevented (route interlocking).

The degree of automation and sophistication ofcontrol varies from system to system. In thesimplest form, ATP is accomplished by automaticwayside block signals and mechanical trip stopsthat activate the emergency brakes for any train en-tering a block illegally or exceeding the allowedspeed. At higher levels of automation, train move-ment is regulated continuously to maintain safespeed, following distance, and routing.

Train control engineers and transit propertiesuniversally consider ATP to be the first and basicmethod of preventing collisions and derailments.The newer systems built and those now under con-struction all incorporate fully automatic train pro-tection mechanisms, Older properties (such asNYCTA, CTA, and MBTA) have long had waysidesignals with trip stops to provide ATP, but they areinstalling fully automated cab signal equipment asthey build new lines or modernize the existinglines. Table 8 is a summary of ATP provisions inexisting and planned transit systems.

The operating experience of existing transitsystems with automatic train protection devices at-

TABLE 8.—Train Protection Methods in Existing and Planned Transit Systems

TRANSIT SYSTEM TRAIN SEPARATION OVERSPEED PROTECTION

Existing Systems:BART (San Francisco)

CTA (Chicago)

CTS (Cleveland)

Dallas-Ft. Worth Airport

MBTA (Boston)

NYCTA (New York)

PATCO (Lindenwold Line)

Seattle-Tacoma Airport

In Planning/Construction:

MARTA (Atlanta)

MTA (Baltimore)

WMATA (Washington, DC.)

Automatic, with advisory cab signals

Mixture, converting to cab signalsl

Airport Ext. automatic trip stops on rest

Automatic

Red Line Ext. automatic, trip stops onrest

Wayside signals with trip stops2

Automatic, with advisory cab signals

Automatic

Automatic, with advisory cab signals

Automatic, with advisory cab signals

Automatic, with advisory cab signals

Automatic, with advisory cab signals

Mixture of manual, trip stops withtimers, and cab signals

Airport Ext. automatic trip stops with

timers on rest

Automatic

Red Line Ext. automatic, manual on rest

Trip stops with timers

Automatic, with advisory cab signals

Automatic

Automatic, with advisory cab signals

Automatic, with advisory cab signals

Automatic, with advisory cab signals

lpresent system is a mixture of n. signals, wayside signals with trip stops, and Cab signals with automatic stop enforcement.Zconversion to cab signals is planned for new lines and extensions.

78

tests to the general effectiveness and reliability ofsuch equipment. PATCO, AIRTRANS, and SEA-TAC have never had a collision or derailment inpassenger service attributable to malfunction ofATP equipment, BART has had one ATP accident,In 1972, shortly after inauguration of service, a trainran off the end of the track at the Fremont Station.The cause of the accident was traced to a faultycrystal oscillator in the carborne speed controlelectronics, causing the train to speed up when itshould have slowed to enter the station. A redun-dant speed control circuit has been added to preventrecurrence of such a mishap and there have been noother accidents related to ATP in the succeedingthree years of passenger service.40

The most frequent types of accidents in a manualtrain protection system are the result of one trainfollowing another too closely, misjudging stoppingdistance, exceeding safe speed on curves, or enter-ing improperly alined switches. All are products ofhuman error. ATP is specifically designed to pre-vent these types of accidents by interposingautomatic safeguards to keep trains properly spacedand running at a safe speed on the correct route,regardless of human error or inattention. The safetyrecord of rail rapid transit owes much to the effec-tiveness of such automatic protective devices whichapply the fail-safe principle to assure that the train

qOThe collision between a BART test train and a maintenancevehicle in January 1975 occurred at night on a weekend, whenthe system was shut down. The cause was found to be human er-ror and improper opera ting procedure by the maintenance vehi-cle driver and the train supervisor in central control.

will maintain a known safe conditionan automated element malfunctions.

in the event

The operating experience of the Chicago TransitAuthority over the past 10 years offers an instruc-tive example of the safety advantages of automaticover manual train protection methods. The case ofCTA is singled out because it is typical of theoperating experience that has led existing transitsystems to conclude that ATP is a necessity.

CTA can be characterized as a mixed system.Ten years ago CTA had wayside signals with tripstops on some lines or parts of lines and no signalprotection on the remainder. In the unsignaled por-tion of the system the safety of train operation de-pended solely on the alertness of the motorman andcompliance with operating rules designed to pre-vent collisions and derailments, As the new DanRyan and Kennedy extensions were built, theywere equipped with cab signals and automatic over-speed protection, In some cases, however, thesenew lines merged with older portions of the systemhaving either no signals or wayside signals with tripstops. Beginning in 1965, CTA undertook a modern-ization program, part of which involved installationof cab signaling to protect segments of trackage for-merly not signaled. This work is now nearing com-pletion, but in late 1975 the system remained a mix-ture of wayside and cab signals, with a few sectionsstill not signaled at all. A train operator on theNorth-South line or the West-Northwest line, forexample, runs the train under all three forms oftrain protection at one time or another during thecourse of a single trip.

FIGURE 47.—Interlocking Control Tower for Train Protection

79

The1965 to

record of collisions and derailments from1974 illustrates the consequences of operat-

ing under incomplete signal protection or bymanual and procedural methods alone. There were35 collisions and 52 derailments in this period, a naverage of about one accident every 6 weeks. Mostw e r e minor accidents , but there w e r e t w ofatalities—both in a 1966 derailment produced byequipment falling off the train. An analysis of acci-dent causes (Table 9) shows that human error was acontributing factor in every collision and in almosttwo-thirds of the derailments.41 Collisions typicallyresulted from the train operator misjudging stop-ping distance or following too closely. Derailmentswere most often caused by overspeed on curves o r

by the operator entering an improperly alinedswitch while proceeding on hand signals.

The record also shows that cab-signaled ATPwas a contributing factor in only one accident. Inthis case, the motorman was operating in cab-signalterritory for the first time on the first day of opera-tion of a newly extended line. The cab signalingunit had not cut in as the train passed from unsig-naled into cab-signaled territory, and not noticingthis, the motorman failed to operate accordingly.42

The train rounded a curve in the subway and hit astanding train ahead because the motorman wasunable to stop in time. CTA determined the causeof the accident to be a combination of cab signalequipment failure and human error. CTA has takenmeasures to prevent recurrence by tighter instruc-tions, modification of procedures for entering cabsignal territory, and more conservative turn-on and

qlApart from human error, the greatest contributing cause inderailments was car defects (16 of 52 cases).

4ZCTA procedures prescribe that, in this circumstance, themotorman should continue to operate under manual rules and beprepared to stop within line-of-sight distance,

testing procedures when initiating service with newcab signal equipment.

Two points emerge from this analysis. First, ATPis superior to manual methods of train protectionbecause it safeguards against most types of humanerror, which cause the majority of collisions andderailments. 43 Second, a mixture of signaled andunsignaled lines requires two distinctly different(and perhaps incompatible) modes of response fromthe train operator, with the attendant risk of confu-sion between the two at a critical moment.44 Boththese points were recognized by CTA, which citedprevention of accidents resulting from human errorand attainment of a uniform level of signal protec-tion for the whole system as prime reasons for un-dertaking the cab signal conversion program.

q3No automatic system is foolproof. After the collision oftrains in the Mexico City transit system on October ZO, 1975, inwhich 27 people were killed, the investigation disclosed that thetrain operator (in violation of established rules) had discon-nected ATP equipment that would normally have stopped thetrain.

qAIn a different way, the recent collision in Boston illustratesthe risk associated with mixing manual and automatic methodsof train protection, On August 1, 1975, in the tunnel between theCharles Street and Park Street stations, an MBTA Red Line trainwas struck from the rear by a following train. About 2 minuteslater, the second train was hit by a third entering the tunnel.There were no fatalities, but 130 were reported injured. Thispart of the Red Line is protected by wayside signals and tripstops. However, about an hour before the accident, a trip stophad malfunctioned; and trains were being moved past the tripstop under manual rules requiring that the motorman proceedslowly and be prepared to stop within line-of-sight distance. Theexact cause of the accident has not been officially determined,but it seems to have resulted not from a failure of the ATPsystem but from a lapse in the manual back-up procedure, Thissuggests that a transit system becomes vulnerable to human er-ror at a time when the normal automatic protection methods areinoperative and train operators must revert to unaccustomedmanual and procedural methods.

TABLE 9.—Analysis of CTA Accident Record, 1965–74

CONTRIBUTING FACTORS*

Car Track Wayside Cab HumanTYPE OF ACCIDENT TOTAL Defect Defect Weather Signals Signals Error Vandals

Collision 35 5 0 2 1 1 35 3Derailment 52 16 4 0 6 0 31 1

● Some accidents had more than one contributing factor.

80

ISSUE O–2: TRAIN OPERATION

Does automatic train operation (ATO) havean influence on safety, as measured by the typeand number of passenger injuries?

Based on analysis of the records of four repre-sentative transit systems, there is no differencein t he i n ju ry r a t e s be tween manua l andautomatic modes of train operation. Passengerinexperience is more of a causal factor than themode of operation.

There are two types of passenger accidents thatmight be influenced by automatic train operation—falls on board due to train motion and door closureaccidents. If either automatic or manual train opera-tion resulted in a characteristically smoother ride,the frequency of passenger falls and injuries due tolurching of the train during starts, stops, and run-ning on curves would be expected to be lower.Automatic door operation might be expected to pro-duce more instances of passengers being struck orcaught by closing doors because there is no train at-tendant to regulate door operation for the tardy orunwary passenger,

An analysis of accident records for four repre-sentative transit systems (NYCTA, CTA, PATCO,and BART) does not substantiate either of these hy-potheses. The frequency of train motion accidentsin the NYCTA and CTA systems, where trains arerun manually by a motorman, is essentially thesame as in PATCO and BART, where train opera-tion is automatic under the supervision of an opera-tor in the control cab. Similarly, the rate of door

closure accidents does not differ regardless ofwhether doors are operated manually (either by aconductor or train operator) or automatically. (Seetable 10.)

A word of caution must be given regarding tran-sit passenger injury statistics. There are no commondefinitions of injury (or its causes) employed by thefour systems considered here or by the transit in-dustry as a whole. For this reason, the injury ratesfor various kinds of accidents are not precisely com-parable from system to system and should be takenonly as general indications of the safety record. Itshould also be noted that the figures given are forclaimed injuries, i.e., passenger reports of injury atthe time of the accident without regard to severityor substantiation by medical examination. Thenumber of actual injuries (e.g., those requiringmedical treatment or those that lead to a later claimfor compensation) is considerably lower, perhapsby as much as half.

It must also be emphasized that passenger in-juries due to any aspect of train operation areevents of extremely low frequency—literally aboutone in a million. By far, the greater proportion of in-juries to transit system patrons (60–80 percent of allaccidents) occurs in stations. Falls on stairways, forexample, typically account for more injuries thanall types of train accidents combined. Table 11, asummary of passenger accident statistics in foursystems, illustrates the general nature of the railrapid transit safety record.

With regard to fatalities, rail rapid transit is oneof the safest of all modes of transportation. In 1973,15 people lost their lives in rail rapid transit acci-

TABLE 10.—Passenger Injuries Due to Train Operation

TRANSIT TRAIN MOTION DOOR CLOSURESYSTEM Mode of Operation Rate 1 Mode of Operation Rate 1

BART (1974) Alltomatic 21 . 0 A u t o m a t i c 1.6

. CTA (1973) Manual 0.7 Conductor 1.3NYCTA (1973/74) Manual 0.4 Conductor 0.4PATCO (1973) Automatic 0.6 Train Operator 1.4

1 Reported in jllries per million passengers.‘Z The BART figllre is for all on-board accidents, which include falls dlle to train motion and other types of mishaps. The rate of acci -

IIents dlle to train motion alone is therefore somewhat lower, probably about the same as the other systems.

81

FIGURE 48.-CTA Passengers Alighting at Belmont Station

TABLE Il.—Passenger Accident Summary

TYPE OF BART (1972-74) CTA (1969-73) PATCO (1969-73) NYCTA (1973–74)INJURY Rate1 Percent Rate 1 Percent Rate1 Percent Rate1 Percent

STATIONS 3.4 61Falls on Stairs. . . . . . . . . . . . . 24.1 61 3.3 24 3.1 21 NA 2 –

Gates/Turnstiles. . . . . . . . . . . NA 2 – 0.2 1 2.0 14 NA 2 –All Other . . . . . . . . . . . . . . . . . 8.5 21 7.2 52 4.4 30 NA 2 –

TRAINSBoarding/Alighting. . . . . . . . . 1.3 3 0.7 5 0.7 5 0.4 7Doors. . . . . . . . . . . . . . . . . . . . . 3.8 10 1.2 9 1.8 12 0.4 7Train Motion. . . . . . . . . . . . . . 1.4 4 0.8 6 0.9 6 0.4 7All Other. . . . . . . . . . . . . . . . . 0.3 1 0.4 3 1.6 11 1.0 18

TOTAL. . . . . . . . . . . . . . . . . 39.4 13.8 14.5 5.6

IRepOrted injuries per 1 million passengers.ZNot avai]able.

d e n t s4 5- - a r a t e o f 0 . 0 0 7 5 f a t a l i t i e s p e r m i l l i o n

p a s s e n g e r s . F a t a l i t y d a t a f o r o t h e r m o d e s o f

t r a n s p o r t a t i o n d u r i n g 1 9 7 3 a r e s h o w n i n t a b l e 1 2 .

R a i l r a p i d t r a n s i t r a n k s a m o n g t h e s a f e s t o f

t r a n s p o r t a t i o n m o d e s i n t e r m s o f f a t a l i t y r a t e , a s

well as in absolute numbers. In the period 1970–72,

the rate was 0.83 deaths per bi l l ion passenger-miles

in rail rapid transit , 46 a s c o m p a r e d t o 0 . 6 9 i n t r a n s i t

qsThere were also 94 deaths due to suicide jumps from sta-tion platforms or trespassing on the right-of-way.

qeExcludes suicides and trespasser deaths.

8 2

buses, 1.03 in scheduled air carriers, 2.6 in passengerrailroads, 20,8 in private motor vehicles, and 21.1 inelevators (Battelle, 1975), To set the rail rapid tran-sit fatality rate in additional perspective, the figureof 0.83 per billion passenger-miles is the equivalentof a six-car train, carrying a total of 900 passengers,traveling over 53 times around the earth before adeath occurs.

Of the passenger deaths in rail rapid transit,about 80 percent are the result of falling whilewalking between cars on a train in motion. The re-

TABLE 12.—Fatalities in the United States by TransportationMode During 1973

TRANSPORTATION MODE NUMBER OF DEATHS

Private AutoTrucksMotorcycle/Motor BikeMarine, recreationalMarine, commercialAviation, privateAviation, commercialGrade CrossingAll RailroadsTaxicabsBusesPipelineRail Rapid Transit, passengersRail Rapid Transit, suicides and trespassers

33,5005,7003,1301,754

3201,340

2271,215

698170170

701594

SOURCES: New York City Council on Economic Environ-ment, 1974; and National Safety Council, 1974.

mainder are produced by a variety of causes, no oneof which accounts for a significant proportion.Thus, train control (either manual or automated) isa contributory factor in only a tiny fraction of railrapid transitdeath in the

fatalities—probably notapproximately 2 billion

more than onepeople carried

2

1-

0

—.

each year, During the 5-year period studied for CTAand PATCO and during the 3 years of BART opera-tion, there have been no passenger deaths on trainsor station platforms as a result of transit operations.In NYCTA between July 1969 and October 1973,there were five deaths related to train operation(three caught in doors and two killed in a collision).

Examination of the accident records for newertransit systems reveals that the patrons’ experiencewith rail rapid transit seems to be more of a con-tributing factor than the difference between manualand automated modes of operation. Accident ratesin the first year of operation of a new transit systemto be three or four times higher than for older andestablished systems or for the same system after thepublic has gained riding experience. Figure 49shows the history of train motion accidents for thePATCO Lindenwold Line and the Dan Ryan exten-sion of the CTA West-South Line, both opened forservice in 1969. Comparable data for the first 3 yearsof BART operation are also shown. PATCO andBART have automated train operation. Trains areoperated manually with cab signals on the DanRyan Line. For comparison, the train motion acci-dentyear

rates for CTA as a whole are shown for the 5-period 1969–73. Here, the rate for a presuma-

83

72-683 [) - 76 - 7

—

bly experienced riding public is steady between 0.6and 1.0 per million passengers, a range which in-cludes the latest figures for PATCO and BART.

A similar learning phenomenon appears in thepattern of door closure accident rates. The rate inBART for the first year (1972–73) was 5.5 permillion, but it declined to 4.3 and then 1.6 in thenext year and a half. In PATCO, the decline wasfrom 2.7 to 1.4 over a 4-year period (1970–73). InCTA as a whole, it fluctuated in the narrow rangeof 1.0 to 1.4. Since car door operation is automatic inBART and manual in PATCO and CTA, automa-tion does not appear to have anything to do with theaccident rate. All three systems seem to be ap-proaching, or to have reached, a common floor ofabout 1.0 to 1,5 per million passengers.

ISSUE O-3: DESIGN SAFETY

With respect to design and engineering, areATC systems safe?

On theoretical grounds, ATC is at least as safeas manual control, and probably safer. However,there is insufficient evidence from actual transitoperations (except in the area of ATP) to evalu-ate safety empirically. There is also somedifference of opinion in the transit industry onhow to assure the safety of a design.

The rail rapid transit industry is extremely con-scious of safety, which is customarily defined as“freedom from fatalities or injuries resulting fromsystem operation. ” Safety-consciousness isreflected not only in the approach to transit opera-tions but also in the design and engineering of track,wayside equipment, and rolling stock. All compo-nents judged to be critical to safety (“vital” compo-nents, in transit engineering parlance) are designedaccording to the fail-safe principle. Stated simply,fail-safe is “a characteristic of a system which en-sures that any malfunction affecting safety willcause the system to revert to a state that is generallyknown to be safe"47 (NTSB, 1973).

qTThe exact interpretation of the fai l-safe principle isdifficult under some conditions, especially where it may lead tostoppage of a train in hazardous circumstances, e.g., a tunnelfire. A discussion of this point is presented later, beginning on~)il~l’ 86,

The fail-safe principle appears to be applied asrigorously to the design of ATC as to other transitsystem components, and probably even more sobecause of the concern engendered by removing thehuman operator from direct involvement in traincontrol functions. Therefore, at the design level atleast, there is no reason to conclude that automatedtrain control systems are not as safe as manualsystems. They may even be safer because possiblehazards due to human error and variability havebeen eliminated by substitution of machine compo-nents.

But has this substitution merely replaced oneform of hazard with another, perhaps to the generaldetriment of system safety? This question goes tothe heart of the automation issue, but it is largelyunanswerable at present for two reasons. First,there is very little empirical evidence from auto-mated systems by which to judge safety historically,except for the case of ATP.48 Second, there are nogenerally acceptable criteria by which to evaluatesafety from a theoretical viewpoint, especiallywhen comparing dissimilar systems.

At present, there are only two operational railrapid transit systems in the United States with asubstantial degree of automation for functions otherthan ATP. PATCO, opened in 1969, has ATP andATO, However, PATCO is a system consisting ofonly one line, and therefore neither representativeof a large urban mass transit system nor a true testof automation technology. On the other hand, thesafety features of PATCO are impressive, reflectingboth safety-consciousness in design and awarenessof the realities of rapid transit operation. The safetyrecord attained by PATCO is excellent and atteststo the basic safety of ATO, at least at that level ofautomation and in a system of that complexity.

The San Francisco BART system is more highlyautomated than PATCO, incorporating ATS as wellas ATP and ATO, but the system is relatively newand still undergoing start-up problems. Testing andevaluation prior to full operational certification arestill being conducted by the California PublicUtilities Commission, However , even before

qcThe traditional view of transit engineers is that the safetyof a transit system is wholly assured by train protection func-tions and that ATO and ATS play no part in safety. This is cor-rect if safety is defined simply as the prevention of collisionsand derailments. However, if safety is defined more broadly asthe freedom from injuries or fatalities resulting from systemoperation (the view taken here), then the safety of ATO andATS equipment becomes highly germane.

84

*revenue operations began in late 1972, BART wasthe subject of intense public controversy over thesafety of ATC design, and the debate continueseven now. The concern over ATC in the transit in-dustry and in State and Federal Government bodiesseems to have been engendered by the BART ex-perience. Nevertheless , i t appears that thedifficulties besetting BART result more fromspecific engineering defects and managementproblems than from any inherent shortcoming ofATC technology itself.

The application of automation technology in railrapid transit is not, of course, limited to PATCO andBART. There are individual lines within largersystems (e.g., the CTA Dan Ryan extension and theQuincy extension of the MBTA Red Line), but theextent of automation is not so great as in PATCOand BART, consisting only of ATP and machine-aided train operation by means of cab signaling.Also, the results in CTA and MBTA are hard to dis-tinguish because of the merger of the cab-signaledportions into lines with other forms of signaling andtrain control.

Outside of rail rapid transit there are some nineautomated guideway transit (AGT) systems49 in theU. S., such as the Dallas/Fort Worth (AIRTRANS)and the Seattle-Tacoma (SEA-TAC) airportsystems, operating without a human controller onboard. The adequacy of ATC with respect to designsafety has been generally established in thesesystems, which employ a technology derived fromrail rapid transit. However, there is some questionwhether this experience is transferable back to theparent rail rapid transit technology. Speeds aregenerally lower in AGT; vehicles are smaller; andthe lines are fewer, with less complex interlocking.

Thus, the pool of operational experience withATC in rail rapid transit is rather small, consistingof 6 years of relevant data from a simple one-linesystem (PATCO) and 3 years from a complexsystem (BART), There is also fragmentary evidencefrom the CTA Dan Ryan and MBTA Red lines,where the level of automation is lower and notcharacteristic of the system as a whole, The datafrom AGT may or may not be applicable to rail

qgAutomated Guide way Transit (ACT) is a general designa-tion for transportation systems operating relatively small, un-manned vehicles+ ither singly or in trains-on fixed guide-ways along an exculsive right-of-way, See the OTA report,Automated Guidewoy Transit (Report No, OTA–T–8), June1975, for an assessment of this type of transit technology.

FIGURE 50-Unmanned Train at Seattle-Tacoma Airport

rapidscale

transit because of certain basic differences ofand complexity.

The opinion of transit system managers withregard to the safety of ATC is significant. A recentsurvey of transit system safety problems, conductedunder UMTA sponsorship, did not identify ATC asan area of concern. Priority action was recom-mended for several safety problems; but train con-trol systems and automation were not mentioned,even though these topics were listed in the surveyform circulated among transit system operatingauthorities (Transit Development Corporation,1975).

85

FIGURE 51.—Fully Automated AIRTRANS Train

The matter of available data on operating ex-perience aside, there remain more fundamentalquestions of methodology and criteria. How issafety to be measured, either empirically andtheoretically? How safe is safe enough? What ismeant by safety? Is ATC Safety equitable with thetrain protection function, or are there safety im-plications in ATO and ATS? Not all these questionshave answers generally accepted by experts in thefield of safety and train control engineering.

A study of ATC safety conducted by the DOTTransportation System Center (1974) reached theconclusion that it is “literally impossible to achievefail-safe design in a large complex control systemhaving many interacting elements and functions.”No matter how carefully designed and tested asystem may be, there will always be certain com-binations of component failure or operational con-ditions that cannot be wholly compensated for. Theprobability of such events, although infinitesimallysmall (1 X 10–6 or less), represent potential safetyhazards that must be dealt with. In other words, no

system as large and complex as a rail rapid transitsystem can be made perfectly safe. Some risk mustalways be taken.

And so, on theoretical grounds, the question ofATC safety reduces to a matter of probabilities andacceptable levels of confidence, At the present time,there is some disagreement within the transit indus-try and among Federal and State regulatory agen-cies as to how these probabilities are to be estimatedor what measure of risk is tolerable.so

The traditional design approach followed in thetransit industry for ATP has been the fail-safe con-cept, where the essential concern is the immediateor short-range response to protect the system fromthe consequences of component or human failure.Customarily, this protection is achieved by initiat-ing a shutdown or reversion to a lower level of per-formance (e.g., decreased speed, greater headways,

sOTransit system professionals have also taken issue with thegeneral approach and some of the conclusions of the TSC study.

86

—— —

longer station dwell time). The difficulty with thisapproach is that most modern transit systems oper-ate on very short headways. Thus, if a failure oc-curs, it is not simply a matter cf stopping one train.The effect reverberates through the entire system,or a large part of it, requiring that many other trainsbe stopped or slowed until the failure can be cor-rected, Such sudden and unexpected changes in theoperating mode of the system can produce a risksituation that pervades far beyond the point offailure and persists long after the failure has beencorrected. Thus, application of the fail-safe princi-ple may produce a response which is safe for theimmediate and local circumstances but which alsoproduces longer-term and more far-reaching conse-quences for the general safety of the system.51

(NTSB, 1973)

As a supplement to the fail-safe approach, NTSBhas advanced the concept of total system safety.The first step of this approach is to select systemgoals, e.g., prevention of collisions and derailments.The system is then analyzed with respect to thesegoals to determine where the system could fail andallow a collision or derailment to happen. Theanalysis permits construction of a “fault tree, ”which includes not just single component failures,but also multiple failures and environmental in-teractions, making it possible to identify those partsof the system which are critical to safety and totrace out the paths where failure must be preventedfrom compromising any of the system safety goals.This, in turn, shows the designer the parts of thesystem which must be provided with redundantcomponents, functionally equivalent mechanisms,self-checking circuits, or inhibitory dev ices .Through application of statistical techniques, it isalso possible to evaluate the likelihood of failuresand adverse circumstances and thereby place theassessment of risks on a quantitative basis. (NTSB,1973; Battelle, 1975)

The approach suggested by NTSB recognizes thatthe safety of the system as a whole is not equivalentto the safety of its parts and offers an alternativemethod to assess interactive and combinatorialeffects of component failure. The NTSB approachalso offers a way to identify hazards on a system-wide basis and to make explicit the level of risk im-posed by each. However, both the methodology and

Slsome memhers of the transit industry have disputed theseconclusions on the grounds that NTSB has misinterpreted thefail-safe principle and that the concept of pervading risk is ap-propriate to aviation hot not to a transit system.

validity of this approach have been challenged bytransit system engineers. Some maintain that thefail-safe principle-correctly applied—is adequateand proven by experience and that there is no needfor recourse to a total system safety concept, Otherscontend that the NTSB approach offers nothingnew and that it is only a restatement of the safetyanalysis methodology customarily applied as part ofthe fail-safe approach.

In summary, the safety of ATC design (exceptfor ATP) has not been conclusively determined.With respect to the theoretical safety of ATC, ade-quate precautions appear to be taken in the designprocess to assure that automated devices result in alevel of safety at least as high as that conventionallyattainable with manual means of train control. Theabsolute safety of ATC devices cannot be ascer-tained by any safety methodology, criteria, ordesign philosophy currently employed in the transitindustry. Empirically based judgments about thesafety of ATO and ATS can be only tentative atpresent because data are limited to a few systemsfor only relatively short periods. With respect toATP, the avai lable evidence indicates thatautomatic methods are safer than manual train pro-tection. In practical terms, accidents due to defectsof train control (either manual or automatic) areevents of very low probability-estimated here tobe on the order of one injury per million passengersand one fatality per billion passengers, rates whichare among the lowest of all modes of transportation.

ISSUE O-4: PASSENGER SECURITY

Does reduction in the number of on-board per-sonnel, brought about through ATC, have an ad-verse effect on passenger security from crime?

There is no evidence to suggest that passengersecurity on trains is affected by reducing the sizeof the train crew.

The security of passengers from criminal acts instations and on trains is a matter of serious concernto rail rapid transit operating authorities. It has beenconjectured that automation, because it tends toreduce the number of transit property employees ontrains and in stations, might have an adverse effecton passenger security. Passengers, especially onlong trains with a crew of only one, might be morevulnerable to assault, robbery, and other criminal

87

acts because the only transit employeerender assistance is located at the fronttrain, often in an isolated compartment,

who couldend of thegiving full

time and attention to train operation or supervisionof ATC equipment.

This line of reasoning has been advancd pri-marily as an argument against reducing the numberof on-board employees as a result of automatingtrain control functions. The argument also bears in-directly on the justification for ATC itself, If per-sonnel in addition to the train operator (the so-called second and third men) are to be kept on boardanyway for security purposes, then they couldassist in train operation by performing manuallysuch functions as door operation, train announce-ments, and equipment monitoring.

The managers of operating transit systems tendto the belief that personnel on board the train havea favorable influence on security, both in protectingpassengers from robbery and assault and in deter-ring vandalism to the train itself. Agencies planningnew systems generally hold the same view, andthose planning to have only one or no on-board at-tendant intend to compensate by having more sta-tion personnel and roving security employees,

The operating transit systems have greatly vary-ing approaches to passenger protection and trainpolicing. NYCTA maintains a very large transitpolice force (5,100, the eighth largest police force inthe country), with patrolmen posted in stations andon the trains themselves during certain hours and inhigh crime areas. PATCO has a rather small transitpolice force (20 men), which includes a dog unitthat patrols the property during the rush and baseperiods and rides the train during owl service.BART also has its own police force; but consideringthe size of the property, the force is small (99 mem-bers, of which 63 are in patrolling platoons). In con-trast, CTA has no transit police force as such;passenger security protection is provided by thepolice departments of the municipalities served.

There is, however, no firm statistical evidence tosupport the contention that presence of operatingpersonnel or police on the train does, in fact, pro-mote passenger security, Crime statistics forfour transit systems (BART, CTA, PATCO, andNYCTA) are presented insystems are not available,tion suggests that the ratesto those shown.

88

table 13. Data for otherbut anecdotal informa-are roughly comparable

Caution should be observed in interpreting thesedata. The four transit systems shown here differgreatly in such characteristics as hours of operation,security measures, and types of communitiesserved. There are also slightly differing definitionsamong the four as to what constitutes robbery orassault. For example, some include purse snatchingin the category of robbery, while others do not.Some list sex offenses separately; some combinesuch crimes with other forms of assault. An attempthas been made to reduce the statistics presentedhere to a common base, but some distortions un-doubtedly remain. Therefore, the rates given in ta-ble 13 should be taken only as an indication of therough dimension of the problem and should not beconsidered to show the relative degree of passengersecurity in the four systems.

TABLE 13.—Passenger Assaults and Robberies for SelectedTransit Systems

ASSAULT/ROBBERY RATESYSTEM (per million passengers)

BART (1973–74) 2.96CTA (1969–73) 1.44NYCTA (1973/74)” 3.49PATCO (1969–73) 0.24

● July 1973 to June 1974.

While only limited conclusions can be drawnfrom this sample of data, there does not seem to beany clear relationship between crime rates and thenumber of operating personnel on the trains, Forexample, PATCO with only one operator on thetrain and unmanned stations has a rate that is anorder of magnitude lower than NYCTA, wherethere are two men on board and police activelypatrol trains and stations. Also, the rates in BARTand NYCTA do not appear to differ substantiallyeven though the two systems are vastly different interms of automation and the level of train and sta-tion manning.

The dominant factors influencing security seemto be the size of the city and the sociological charac-teristics of the areas served. It should also be ob-served that, if ATC has any influence at all, it islikely to be small since the preponderance of crimein rapid transit systems (75–80 percent) does not

FIGURE 52.—Approach to Brightly Lighted Station

take place on trains, but in and near stations. A and that patrols are concentrated there. In light ofstudy conducted by the American Transit Associa- this, it is perhaps significant that most transit-prop-tion (1973) concluded that station security was by erties list all assault and robbery statistics under thefar the more critical problem and that station crime general heading of station incidents.was concentrated in neighborhoods of generallyhigh crime, usually near the residence of the crim- As a final comment, a distinction must be madeinals, Anecdotal evidence from transit properties between the real (i.e., statistically measurable)interviewed also indicates that the areas of greatest security of passengers and their perception ofconcern are stations, access ways, and parking lots security while using a transit system, In the area of

FIGURE 53,—Lonely Station at Off-Peak Hour

89

perceived security, most transit operators and plan-ning agencies agree that the on-board employeeplays a useful and reassuring role. Communicationsof any and all forms are also believed to be usefulfor enhancing perceived (and real) security ofpatrons. Two-way communication with passengersis regarded as mandatory for systems with unat-tended vehicles. Surveillance of train interiors byclosed-circuit television is technically feasible, butmost properties consider the cost of purchasing andoperating the equipment to be prohibitive in com-parison to the potential benefits.

Data on the perceptions of passengers them-selves do not exist in any meaningful quantity. Inone of the few studies made of passenger attitudes,a telephone survey of 1,586 bus and rail rapid transitpatrons in Chicago, it was found that passengerswould derive the greatest sense of security from thepresence of a police officer on the train or platformand from the knowledge that help was availablequickly from station personnel or the police. Fewrespondents (8 percent) mentioned the presence ofa conductor or motorman as a reassuring factor.This survey also found that CTA patrons tended toregard subway stations and elevated platforms asmore dangerous than the trains themselves. (ATA,1973)

PERFORMANCE

The operational characteristics of ATC canaffect the general performance of a transit system inseveral ways. Some may be qualititative; othersquantitative. Some may directly affect transitpatrons and be perceived by them as benefits. Otherperformance characteristics may be of concern pri-marily to the operating authority and go largely un-noticed by the riding public. Those selected forexamination here are the more tangible andmeasurable aspects of system performance, wheredifferences between manual and automated formsof train control might be manifested as benefits foreither the transit patron or the operating authority.They are:

Ride Quality—the smoothness and comfort ofthe ride, expressed in terms of speed and its deriva-tives (acceleration and jerk);

Level of Service—the convenience and depend-ability of the transit system, measured as headway,trip time, available seating, and adherence toschedule;

Availability—the ability of the system to sustainthe required level of daily service, as indicated bythe reliability and maintainability of equipment.

As in the preceding discussion of safety, the per-formance of ATC systems is treated as a series ofissues, with operational experience from variouscities presented in tabular format to substantiate theconclusions. This method of presentation tends toinvite comparisons among transit systems; and it isintended that the reader do so, but only within thelimits set forth in the discussion of the issue. Somedifferences are more apparent than real, They ariseeither from different definitions and recordkeepingmethods or from differences among systems thathave nothing to do with train control (e.g., trackgeometry, right-of-way conditions, station spacing,environmental factors, age of equipment, and soon). An effort has been made to reduce all data to acommon base and to use standardized terms, butthere still remains a need for caution in makingdirect comparisons across systems.

ISSUE O-5: RIDE QUALITY

What effects does automatic train operation(ATO) have on ride quality and comfort?

ATO systems provide a ride quality equal tothat of manual modes of operation. Some con-sider ATO systems superior in that the ridequality is more uniform.

Ride quality is a generalsmoothness and comfort ofceived by the passenger. It is

term referring to thetrain motion as per-measured in terms of

the acceleration and deceleration characteristics ofthe vehicle while running at speed and during ar-rival and departure from stations. Ride quality is in-fluenced by many factors—propulsion and brakingsystem characteristics, vehicle suspension, trackgeometry, condition of the right-of-way, wheel-railadhesion, signal system design, and speed regula-tion technique. Of these, only the last two fall with-in the province of the train control system, and theyusually do not have a major influence on ridequality. Vehicle and track characteristics are by farthe dominant factors. However, the train controlsystem can play a part in enhancing ride quality orin compensating for adverse effects produced byother factors.

90

In terms of train control functions, ride qualigoverned by those elements of the systemregulate speed and execute programed station s

ty isthat tops.

Three aspects of motion must be controlled: speed,acceleration, and the rate of change of acceleration.

Acceleration and deceleration (the rate at whichspeed changes) is related to, but not actually a partof, the speed regulation function of the train controlsystem. 52 For passenger comfort, as well as safety,the changes in velocity must be kept within certainlimits when running the train up to speed and whencoming to a stop at stations. Different rates may beemployed-a nominal rate for service braking and asomewhat higher rate for emergency stops.

It is important to control not only accelerationbut also jerk—the rate of change of acceleration, sonamed because of the uncomfortable (and poten-tially hazardous) effect produced by abrupt changesin acceleration or speed. 53 Control of jerk, morethan control of acceleration itself, contributes to asmooth ride and, for the standing passenger, asomewhat safer one. Jerk limiting applied duringstopping is sometimes called flare-out control. It isidentical to the technique employed by a skilledautomobile driver when coming to a stop. By easingoff on the brake, the transition from deceleration tofull stop is smoothed or feathered out. Becausethere are safety implications to relaxing brakingeffort while stopping, flare-out control (a trainoperation function) is overridden by the train pro-tection (ATP) system such that flare-out is pre-vented during emergency braking.

Maintaining optimum wheel-rail adhesion iscalled slip-slide control. Slip denotes the slipping orspinning of wheels during the application of power.Slide denotes the sliding or skidding of wheelswhen brakes are applied. Both are operationally un -

~~Acceleration and deceleration control is considered bytransit engineers to be a part of the traction system. While it istrue that the equipment controlling acceleration and decelera-tion is physically a part of the traction system, the functionalboundary between this system and the train control system issomewhat fllzzy, and a case can be made for treating accelera -tion and deceleration control as part of either one, In practicalterms, th[~ (Distinction is [unimportant since speed regulation, ac-celerat ion and deceleration control. and jerk limiting all intera(; tto prodm:e a smooth ride.

5.JJerL li~l tjng IS also considered te(;hn ically a function of thetraction system. The train control system commands a specificlevel of acceleration. and the propulsion system responds by ap-pl i cat ion of power or brah i ng to produce accelera -tion/{lecelera tion a t a rate w i th i n allowable equipment orhuman tolerances,

desirable because they represent inefficiency inrunning the train and may cause damage to tracks,wheels, or the propulsion and braking system of thetrain. For the passenger’s perception of ride quality,slip-slide control is only marginally important, butit does affect jerk characteristics. There are alsosafety implications; the system is usually designedso that failure of slip-slide control does not allowrelease of brakes when safety requires that they beapplied.

In transit systems where trains are operatedmanually, speed regulation, slip-slide control, andflare-out are usually performed by the motorman.54

The ride quality resulting for the passenger is thusdetermined by the skill or artistry of the individualmotorman and the consistency with which he ap-plies proper technique. In transit systems withATO, these three functions are usually automatic.The use of automatic mechanisms is generally con-sidered to offer two advantages. First, the train ismore likely to be operated within the limits accept-able for passengers and equipment because the con-trol system is designed to preclude human error andimproper technique. Second, automatic operationleads to less variation; human control varies con-siderably with individuals and time.

Table 14 is a summary of the speed regulation,jerk limiting, and slip-slide control methodsemployed in five operating transit systems. Thenew transit systems planned for Washington,Atlanta, and Baltimore will all employ automatictechniques similar to those of PATCO or BART.

There is almost no empirical evidence to supportor refute the advantages claimed for ATO ontheoretical grounds. Systematic studies in experi-mental settings or under actual operating conditionshave not been conducted, and there is no effort nowunder way to do so. The opinion of some transitsystem engineers is that ATO leads to a ride qualityand type of train operation that is at least as good asmanual control, and perhaps even superior becauseof the ability of automatic devices to operate withinprescribed tolerances more consistently. Transitsystem managers also seem inclined to this view.There is, however, some dissenting opinion fromboth engineers and managers. Perhaps the mostconclusive indication that ATO is preferable tomanual control is that all the transit systems nowunder development and most of the proposed ex-tensions or improvements to existing systems will

S~Jerk limiting is a u toma tic on a 11 operating transit systemsand is built into the propulsion system.

.

FIGURE 54.—Comfort Features of Modern Transit Cars

TABLE 14.—Train Operation Methods Related to Ride Quality

TRANSIT ACCELERATION RATE

SYSTEM ACCELERATION JERK LIMITING FLARE-OUT

NYCTA Automatic] Automatic Manual, except on new R–44 andR-46 cars when operating withATO

CTA Automatic] Automatic ManualMBTA Automatic Automatic Manual

PATCO Automatic Automatic Automatic on service brakes, exceptin manua1 backup mode: none onemergency brakes

BART Automatic Automatic Automatic on service brakes, exceptin manuall back-up mode: none onemergency brakes

SLIP-SLIDE

Manual, except on new R–44 andR–46 cars

ManualManual, except on new Red Line

carsAutomatic in all propulsion and

b r a k i n g m o d e s , i n c l u d i n gemergency braking

Automatic in all propulsion andb r a k i n g m o d e s , i n c l u d i n gemergency braking

1Inherent in propulsion system design.

92

incorporate automatic control of acceleration, jerk, run according to schedule, making the prescribedflare-out, and slip-slide. stops, and with the requisite number of cars.55

ISSUE O-6: LEVEL OF SERVICE

Do transit systems with ATC provide a levelof service that is comparable to manually con-trolled systems?

Generally yes, although some systems withATC have encountered difficulty in maintainingschedules, especially during the initial months ofservice.

Table 15 is a summary of the service-related per-formance characteristics of five transit systemswith various degrees of automation of train opera-tion and supervision functions. Also shown are theservice characteristics of the AIRTRANS system atDallas-Fort Worth Airport. Although AIRTRANS isan airport people-mover system in the AGT classand not a true rail rapid transit system, it has beenincluded as example of a wholly automated systemoperating without on-board personnel. No existingrail rapid transit system operates in this manner.The data for AIRTRANS, BART, PATCO, andNYCTA apply to the entire system. The CTA andMBTA data are for only the most automated lines.

Level of service is a general term that includes The speeds and headways for the two rail rapidboth the characteristics of the service offered transit systems with ATO (BART and PATCO) are(speed, trip time, frequency of trains) and the de- generally equivalent or superior to those of thebendability of that service. Designers of transit systems with manual train operation.56 It must besystems consider these aspects of service, along noted, however, that maximum speed is little in-with comfort and convenience, to be determining fluenced by ATO. Speed is mainly a function offactors in gaining and holding public patronage. The track condition, vehicle characteristics, age ofassumption is that if travel time can be saved byusing rail rapid transit, if service is available whenwanted, and if there is assurance that the trip willbe completed according to schedule, a large share ofthe public will choose rail rapid transit over othermodes of transportation, Advocates of automationcontend that ATC offers the means to upgradeservice by making it possible to operate trains atgreater speeds, on shorter headways, in closer con-formance to schedule, and with greater regularity,

Maintaining a high level of service depends onhow well both the train operation and train supervi-sion functions are carried out. The elements of thesystem responsible for operating trains, whether themotorman or an automatic device, must assure thattrains are run at the prescribed speeds, making thescheduled stops and departing from stations afterthe specified dwell times. The train supervisionfunction, either by humans or computers, entailsmonitoring the performance of individual trains inrelation to overall passenger demand and makingcompensating adjustments of schedule, runningtime, station stops, and dwell time as necessary toovercome irregularities of train operation, varia-tions in demand, or adversities of weather. The suc-cess of this combined train operation and supervi-sion activity is measured by schedule adherence,i.e., the percentage of time that trains are actually

equipment, and station spacing, to name a few.Thus, the higher speeds attained in BART andPATCO do not necessarily reflect an y superiority ofATO over manual operation. These systems arenewer, in better condition, and built for differentpurposes. 57 The track and rolling stock have beendesigned for high-speed operation. Station spacingpermits longer runs at maximum speed, therebyraising the average line speed, Still, the data do sug-gest that systems with ATO are capable of provid-ing a level of service at least equivalent to that ofmanual systems,

With regard to headways, ATO does seem tooffer advantages over a manual system. Headway isbasically determined by the level and qua lit y of sig-nal protection (ATP) and the regularity with which

Ssu]timate]y, level of service depends more on managementpolicy than technological features, since it is management thatsets the desired level of service and determines the degree ofcommitment to maintaining service in the face of adverse cir-cumstances,

SoSpeed in the AIRTRANS system is substantially lower, butthis is a result of very c lose s t a t i o n s p a c i ng (typica]ly1,000–2,000 feet) which does not permit vehicles to operate atmaximum speed for more than a few seconds.

5 7BART and PATCO are basically interurban systems morecomparable to the Long Island Rail Road or Penn Central’s com-muter services than to the NYCTA or CTA urban transitsystems <

93

— . —— .—

TABLE 15.-Service Characteristics in Typical Transit Systems

One-WayMaximum Trips/Day Trip Time

AUTOMATION SPEED (mph) HEADWAY (min.) Train (each way) in PeakTRANSIT Length PeriodSYSTEM ATP ATO ATS2 Max, Av. Peak Base (cars) (min.)

NYCTA J 50 20 2 10–12 11 8,000 359CTA (Dan Ryan) ~ 55 30 3 5 8 225 42 1/2

MBTA (Red Line) 4 J 4

50 30 2 ‘/2 4 ‘ /2 4 255 25PATCO l l J 75 40 2 10 6 182 22 1/2

BART J J J 80 40 56 6 105280 54–57

AIRTRANS J d d 17 961 — 2 —6 —6

1A check (~ indicates the function is automated. All but AIRTRANS have an on-board operator to run the train or monitorautomatic system performance.

ZAutomation here specifically means computer-aided central control.Ssystem-wide average; trip time on individual lines varies considerably as a function of line length and whether service is local or

express.4A portion of the rol~te is eqllipped for AT() but current]y operates under manual control, Cars are capable of 70 mph top speed

but are governed to so mph for manual operation.SThe figures are for interim level of service; when fully operational, approximately 600 trips per day will be run at headways of 2

minutes during peak periods and 4 minutes during the base period.BAIRTRANS operates 17 overlapping loop routes of varying length. Trains circulate continually throughout the day on a schedule

determined by aircraft arrivals and departures.

trains are operated, i.e., the invariance of runningtime. There is a large, but not unanimous, body ofopinion among transit engineers and managers thatATO is necessary in order to operate trains at highspeeds on short headways. Given a signal and trainprotection system of good quality, trains can be runmanually on short headways, viz., NYCTA or CTA,where scheduled headways on individual lines areon the order of 1–2 minutes and composite head-ways on merged lines sharing a single track may be40–50 seconds. Given the proper equipment andtrack conditions, trains can also be run at high speedunder manual control. Metroliners have operatedmanually in regular service at speeds of up to 130mph. But some transit engineers and plannersbelieve that the combination of high speed andshort headway cannot be attained without the helpof ATO to eliminate the variability of manualoperation.

Data to support this contention are scarcebecause there is only one transit system (PATCO)where manual and automatic modes of operationcan be directly compared. The PATCO trains arenormally run under ATO, but full-speed manualoperation is possible as an alternate mode and is, infact, required of each train operator once a day as a

means of maintaining proficiency. The PATCO ex-perience has been that the trips run under manualcontrol average about 20 seconds longer and are ofmuch greater variability than ATO runs, Sincethese manual proficiency trips are not run duringpeak periods, the impact of longer and more varia-ble running time on headway is hard to assess, butthe effect might be to increase headway and solessen the overall throughput of the system. On theother hand, the PATCO results may be misleadingbecause they were obtained while running with aclear track ahead. Some transit engineers contendthat, when trains must follow closely or when trackand weather conditions are adverse, the manualoperator is superior to the automatic device; andtrains can be run more uniformly, at closer head-ways, and with shorter running times,

For the transit patron, the schedule of train serv-ice is only part of the equation. The patron also re-quires assurance that the schedule will be main-tained with a high degree of consistency. That is,the performance history of the system must lead thepatron to the conclusion that he can rely on the tripbeing completed, on t ime, without skippingscheduled stops, andcar space available.

with the customary amount of

94

FIGURE 55.—The Wait . . . . . . . . . . and the Rush to Leave

Schedule adherence of transit systems is notstrictly comparable because of differing definitionsof on-time performance and dissimilar methods ofkeeping operational logs. For example, somesystems consider a train on time if it arrives at a ter-minal within the turnaround time, i.e., in time todepart on schedule for the next run. Others use anarbitrary definition, such as a delay not exceeding 5minutes, either at a terminal or at checkpoints alongthe route. Still others, such as BART, use a moredynamic and detai led measure of scheduleadherence that takes into account the impact of in-dividual delays on total system performance.

Schedule adherence is also influenced bystrategy employed in setting a schedule. One

theap-

proach is to base the schedule on maximum trainperformance (maximum attainable speed, accelera-tion, and deceleration and minimum coasting time)with the expectation that maximum throughputwill be” achieved except for a small fraction of thetime when complications arise. An alternative ap-proach is to schedule trains at something less thantheir maximum performance, thereby creating abuilt-in reserve of performance that can be used tomake up delays en route. This approach sacrificessome throughput but offers greater assurance thatthe schedule can be met.

Because of these dissimilarities in settingschedules and defining on-time performance, directcomparisons across transit systems cannot be made,

95

—

The following data, therefore, should be regardedonly as individual examples of schedule adherencefor representative transit systems.

P A T C O

A train is considered late in PATCO if it arrivesat a terminal more than 5 minutes behind schedule.PATCO keeps a daily log of lateness and otherschedule anomalies such as trips annulled, stationstops missed, and trips made with less than thescheduled number of cars (short consist). Table 16shows the performance figures for 1974 and for anaverage year in the period 1970– 74.

PATCO also computes an overall index ofschedule adherence:

where:T s = trips scheduled

T a = trips annulled

CTA

CTA has a very stringent definition of latenessand employs a complex strategy to compensate fordelays, A train is considered late if it is more than 30seconds behind schedule at a terminal or intermedi-ate checkpoint, When this occurs, preceding andfollowing trains are deliberately delayed also so asto minimize irregularity in headways and balancethe service.

For the purpose of this report, a special study wasmade of schedule adherence on one CTA line, theWest-South (Lake-Dan Ryan), which is one of thenewest lines and operates with cab signals. On-timewas defined to be arrival at a terminal with a delaynot exceeding the scheduled turnaround time, i.e.,the actual time of arrival was not later than the nextscheduled departure of the train. Depending on thetime of day, turnaround time on this line is between5 and 7 minutes-a standard roughly comparable tothat of PATCO. In addition to delay, the analysisalso considered the number of trips annulled,scheduled station stops bypassed, and consist short-ages, Table 17 is a summary of findings for the year1974 and for the 5-year period 1970–74.

T I = trips lateTABLE 17.-Schedule Adherence on CTA Dan Ryan Line,

Sb = stations bypassed 1970–74

Applying this formula gives a figure of 98.71 per-cent schedule adherence in the 5-year period1970–74 and a figure of 98.34 percent in 1974. It isworth noting that in 1974, despite a derailment dueto traction motor failure and a subsequent scheduledisruption caused by replacement of motor bearingsfor all cars in the fleet, PATCO was able to sustain alevel of performance nearly equal to that of the pre-ceding 4 years—98.34 percent in 1974 versus 98.80percent in 1970–73.

TABLE 16.-Schedule Adherence in PATCO, 1970–74

FIVE-YEARPERFORMANCE 1974 AVERAGE

(1970-74)

SCHEDULED TRIPSPercent on time 98.36 98.75Percent late 1.16 1.03Percent annulled 0.48 0.23

S C H E D U L E D S T O P SBYPASSED (%) 0,18 0.40

T R I P S M A D E W I T HSCHEDULEDNUMBER OF CARS (%) 99.66 99.75

Five-YearPERFORMANCE 1974 Average

(1970-74)

SCHEDULED TRIPSPercent on time 96,26 97.37Percent late 3.65 2.50Percent annulled 0.09 0.13

SCHEDULED STOPSBYPASSED (%) 0.34 0.26

T R I P S M A D E W I T HSCHEDULEDNUMBER OF CARS (%) 99.93 99.89

Despite certain basic differences between PAT-CO and the CTA Dan Ryan line in route complex-ity, track geometry, and station spacing, theperformance histories of the two systems areroughly comparable when logged on essentially thesame basis. The on-time records of both are on theorder of 97-78 percent, and the percentage of stopsmade and the percentage of trips run with a fullconsist are nearly 100 percent. Thus, it would ap-pear that a manual system with ATP (CTA) and an

96

automated system with ATP and ATO (PATCO)can achieve equal levels of schedule adherence.

NYCTA

The rapid transit system operated by NYCTA isthe largest and most complex in the United States.Automation is minimal, consisting of automatictrain protection by wayside signals with trip stopsand some automated dispatching. Train operation iswholly manual.

In 1974, the on-time performance record ofNYCTA was 97.03 percent, where a train is con-sidered on time if it arrives at a terminal within 10minutes of the schedule. During 1974, there were32,515 delays of unspecified length, or about 90 perday or three per line,

AIRTRANS

AIRTRANS at the Dallas/Fort Worth Airport hasa fully automatic train control system. Automatedtrains operate on 17 intermeshed routes over about13 miles of one-way track. The system is still in the

process of shakedown and debugging, havingopened for operation in January 1974.58

Figure 56 is a plot of the availability of thesystem on a weekly basis from May to October1974, where availability is expressed as the ratio ofactual hours of operation to scheduled hours ofoperation. The figure also shows the number ofservice interruptions experienced each week.

It can be seen that, during the month of May, arelatively few service interruptions caused longdelays. In June, the schedule of operation was in-creased from 105 to 168 hours per week, and thenumber of service interruptions increased sharplyto over 160 per week, or about one per hour. As ex-perience was gained and debugging of the systemcontinued, the length of delay per interruptiondecreased. By October, system availability averagedover 99 percent, while the number of service inter-ruptions declined to about 40 per week. While serv-

wIn November 1975 the system was shut down as a result of acontract dispute between the airport management and themanufacturer.

FIGURE 56.—AIRTRANS Availability and Service Interruptions

97

ice interruptions aretrips, it may providePATCO experienced

———.— —

not truly equivalent to lateperspective to consider thatabout 20 late trips per week

and the CTA Dan Ryan line about 54 per week dur-ing the first year of operation.

BART

BART has ATO and employs a computer-basedATS system for maintaining trains on schedule,The basic performance index is “total systemoffset, ” an expression of the aggregate delay for alltrains operating in the system after application ofcorrective scheduling algorithms. This measure ismore complex than that used by other transitsystems, not only because it incorporates more fac-tors, but also because it considers the compensatingadjustments which have been applied to followingand leading trains, in addition to the late train itself,Thus, a train that is 30 seconds late will result indelays of 5 to 15 seconds for as many as threefollowing and two leading trains, producing a totalsystem offset of as much as 65 seconds while thecentral control computer respaces the trains andsmooths out the traffic flow.

During the first 9 months of operation, under apartial schedule with lo-minute headways, BARTexperienced severe service disruptions. In the weekof 25–29 June, 1973, for example, total system offsetaveraged about 12 minutes in the morning and in-creased to over 45 minutes by the evening rushhour. Delays of over 10 minutes were experienced

five times during the week, and short consists wererun 16 times for periods ranging from 16 minutes to3 hours.

Table 18 shows a larger sample of data, consist-ing of weekly performance summaries selected atapproximately 4-week intervals from August 1973to August 1974. During this period, which coversroughly the second year of operation, transbay serv-ice had not yet been inaugurated, and BART wasrunning what amounted to two separate systems:Fremont/Richmond/Concord service in the EastBay and San Francisco/Daly City service in theWest Bay. Service was limited to the hours of 6 a.m.to 8 p.m., weekdays only.

Examination of the data for the period indicates aslight improving trend with respect to delays, carshortages, and total system offset. The opening ofthe Transbay Tube in September 1974 caused asharp decline in the regularity of service for a fewweeks; but by the last week of 1974, total systemoffset was running at an average of 3.6 minutes inthe morning and 20.4 minutes in the evening. Thesefigures are roughly comparable to those of August1974, the month preceding inauguration of transbayservice, Still, it appears that the BART system hasnot yet attained a level of service dependabilitycomparable to other rail rapid transit systems.

Other Transportation Modes

To assess the general quality of service providedby rail rapid transit, it is useful to make some rough

TABLE 18.—Schedule Adherence in BART, August 1973–August 1974

TOTAL SYSTEMWEEKLY TOTAL OFFSET (minutes)

Trains Delays Over Short Daily Average

WEEK Dispatched 10 min. Consist 7:00 a.m. 4:30 p.m.

20–24 Aug. 7317–21 Sep. 7315–19 Oct. 7312–16 Nov. 7 31

IO– 14 Dec. 737 –II Jan, 74

18–22 Feb. 7418–22 Mar. 748–12 Apr. 74

13–17 May 7410–14 Jun. 748–12 Jul. 745–9 Aug. 74

—AVERAGE

116

124

135

149

166

166

170

172

170

145

162

164

185

8

10

18

10

9

9

5

5

8

6

5

2

8

1112

21

17

26

38

17

19

23

25

28

18

26

8.2

4.8

9.2

9.0

9.6

11.8

10.8

3.4

7.8

2.8

2.2

6.6

0.8

36.023.045.439.819.028.622.636.624.616.823.621.615.6

156 8 22 6.7 27.21West Bay service began on November 5 1973

comparisons with other modes of public transporta-tion. The on-time performance records of the railrapid transit systems examined here range from 97percent for an essentially manual system (NYCTA)to almost 99 percent for a system with ATP andATO (PATCO). The on-time performance of morehighly automated systems such as BART andAIRTRANS cannot be determined from the dataavailable, but it appears to be not lower than 90 per-cent.

The Metroliner operating between New Yorkand Washington is comparable to rail rapid transitsince it operates on a fixed guideway in an ex-clusive right-of-way and employs similar train con-trol technology, The on-t ime record of theMetroliner is currently running at about 53 percent,where a train is counted late if it arrives more than15 minutes behind schedule on a trip of 3 hours. On-time performance for railroads in general exceeds90 percent for many lines and in some cases reaches95 percent (Reistrup, 1975).

Air carrier service is a more remote comparison,but still generally valid if limited to flights of aboutthe same duration as a typical rail rapid transit run.The on-time performance record in September 1974is given below for air carrier service between threepairs of cities about one flight-hour apart:

New York–Washington 79 percentLos Angeles–San Francisco 84 percentLos Angeles–Las Vegas 84 percent

(Air Transport World, 1975)

A flight is considered on time if it arrives within15 minutes of schedule, a less stringent standardthan the 5–10 minutes used in the rail rapid transitsystems cited above,

ISSUE O-7: RELIABILITY

effect has ATC equipment reliabilityhad on the performance of transit systems?

ATC equipment poses reliability problems,especially during the initial period of systemoperation. However, in comparison with othercomponents of the transit system, ATC equip-ment does not cause a disproportionate share ofservice disruptions. The problems do not seem tostem from automation per se but from the in-creased complexity of all new transit systemequipment,

The general trip dependability, or scheduleadherence, of rail rapid transit systems employingmanual or automatic train control was examined inthe previous issue. It was found that the method oftrain operation, either manual or automatic, did nothave a major influence. The principal cause ofschedule irregularity and service disruptions is nothow dependably the train is operated, but how serv-iceable is the transit equipment itself. Thus,schedule adherence ultimately reduces to a ques-tion of whether the equipment can render servicewhen needed.

Technically, the ability of equipment to renderservice when needed is known as availability andembraces two separate concerns:

(1) Reliability—the ability of the equipment tooperate as required at any giventime,

(z) Maintainability—the ability of the equipmentto be restored to operatingcondition after failure.

The two are closely related, but only the matterof reliability will be examined here. Maintainabilityis taken up as the next issue. To provide someperspective for these issues, however, a briefdescription of the general nature of reliability,maintainability, and availability (RMA) is in order,

Reliability, maintainability, and availability arelinked in a relationship that can be expressedmathematically as:

where A =

R =

M =mean time to repair (or restore) toserviceable condition (MTTR)

In effect, the entire expression reduces to a state-ment of the probability that the equipment will beavai lable in working condi t ion, or that thepassenger will find the transit system fully opera-tional at any given time,

99

72 .fi8’{ ( ) - 76 - f{

The general standard in transit systems is for thereliability (MTBF) of major assemblies or sub-systems to be on the order of 1,000 hours or more.Repair time (MTTR) is typically 1 or 2 hours. Com-bining the separate MTBF and MTTR for all sub-systems yields on expected availability of roughly98 to 99 percent for the entire system. The issues tobe examined here are whether this expectation is,in fact, realized and what part is played by ATCequipment in the overall RMA picture.

Despite the recognition in the transit industrythat reliability is perhaps the single most pressingtechnical problem, this study did not uncover a sig-nificant body of operational data on the perform-ance of vehicle and wayside equipment compo-nents. Some transit agencies attempt to maintain asystematic data bank of reliability information,with computer analysis and calculation of compo-nent reliability rates (MTBF). Others have a lessformal system consisting of shop logs, summaries ofindividual failure reports (“bad orders”), and othersuch working records. The methods of recordingfailures differ among transit systems. Some recordfailures at the component level, others group thesefailures in higher order assemblies, such as sub-systems or replaceable modules. The definition ofwhat constitutes a failure also varies. Some countreports of failure by train operators; others countonly failures confirmed by shop personnel and ex-clude the so-called “false bad order” or intermittentfailure. Still others count only those failures thatdisable a train or cause it to be removed or withheldfrom operation.

For those that calculate MTBF, some use a timebase that includes all the hours the equipment is ac-tually in operation, counting the time in revenueservice as well as the time in yards or on storagetracks when the equipment may be energized butthe train not running. Others count only revenueservice hours. This difference alone can have sig-nificant impact on the calculated failure rate. A tBART, for example, it is estimated that yard time isabout twice the revenue hours.

As a result, a quantitative analysis of reliabilitycould not be performed in such a way to permitdetailed comparison of experience with ATC equip-ment among transit systems. The following sum-maries of equipment failure and reliability informa-tion for individual systems are therefore not to becompared, except at the most general level and onlywithin the limits noted in the discussion.

FIGURE 57.--Carborne and Wayside ATC Equipment

PATCO

reliability andproduces sum-

PATCO has a computer-basedmaintenance record system thatmaries of failure data at 4-week intervals. Table 19is a sample of car component performance data for arepresentative 16-week period from mid-July to theend of November 1974. Only certain categories ofequipment failure have been selected—ATC equip-ment and a sample of other major componentsgenerally considered reliable. Data on periodic in-spection and preventive maintenance have alsobeen included to indicate the proport ion ofscheduled to unscheduled maintenance events.

It can be seen that ATC equipment failure ac-counts for about 6 percent of all maintenanceevents—roughly the same as the propulsion controlequipment (cam controller) and air brakes, two con-ventional items of car equipment that are generally

100

TABLE 19.—PATCO Car Component Performance, July-October 1974

NUMBER OF FAILURES PER-CENTAGE

COMPONENT 13 JuL- 10 Aug.– 7 Sep.– 5 Oct. -TOTAL

4-WEEK OF ALL9 Aug. 6 Sep. 4 Oct. 1 Nov. AVERAGE FAILURES

ATC. Air Brake

Cam ControllerCommunicationControllerCouplerMaster ControllerMotor-GeneratorAll Other

667447262060

534

1201

735289311772

136

698

100101

843042

1977

56788

454868313195

442

634

284275268118110424

17168

3321

7169723028

1064

42830

5.75.55.82.42.28.50.33.4

66.3

Periodic Maintenance 219 270 449 275 1213 303

IHigh voltage switches.Zoperator’s control unit in cab.

regarded as reliable elements. The incidence ofcoupler failure is about one and one-half times ashigh as that of ATC equipment. PATCO was ex-periencing a problem with couplers at that time,necessitating a redesign and replacement of theoriginal equipment. The failure rate for couplerswas therefore unrepresentatively high during thesample period. From these data, it can be concludedthat ATC equipment at PATCO, accounting forabout one failure in eighteen for the all carbornecomponents, is not a reliability problem of dis-proportionate magnitude.

A separate analysis, performed by Battelle Co-lumbus Laboratory in support of this study, con-sidered only disabling failures59 and covered a 1-year period from August 1973 to July 1974. Thesedata, presented in table 20, indicate that ATCfailures accounted for about 10 percent of all trainremovals during the year, but with considerablevariance. ATC failures, expressed as a percentageof all disabling failures, ranged from as low as 7 per-cent to as high as 22 percent. Using these data, Bat-telle also calculated MTBF for vehicles as a wholeand for carborne ATC equipment. Vehicle MTBFwas found to be 23.9 hours, and the ATC MTBFwas about 227 hours, Since cars were operated anaverage of 30 hours per week, each car had about1.2 disabling failures per week.

WA disab]ing failure, as defined by PATCO, is o n e t h a twould require removal of a train or car from service or preventits return to service after leaving the line at the end of ascheduled run.

ATC accounted for about one-tenth of theremovals, or roughly one removal per car every 8weeks. Thus, ATC reliability accounted for 6 per-cent of all failures but about 10 percent of removalsfrom service, a reflection of the criticality of ATC totrain system performance. Still, the magnitude ofdisabling failures due to ATC was not large—repre-s e n t i n g a b o u t o n e i n c i d e n t p e r c a revery 8 weeks or, for the whole fleet of 75 cars, 488removals due to ATC out of the 4,797 experiencedin a year,

BART

Like PATCO, BART has a computer-basedrecordkeeping system for reliability and main-tainability information. However, because ofdifferences in the definition of failure and theequipment categories in which data are tabulated,reliability data for the two systems cannot bedirectly compared. Table 21 is a summary ofreported failures by major equipment categories forthe period January 1, 1974, to January 21, 1975. Twomajor classes of equipment are included, carborneequipment and wayside equipment, The latter classincludes a substantial amount of train controlequipment required for ATP (interlocking control),ATO, and ATS.

The failure of BART carborne ATC equipmentaccounts for about 11 percent of all carborne equip-ment failures, a proportion almost identical to thatof PATCO, if it is assumed that all the BART

101

—.

TABLE 20,—Summary of Disabling Equipment Failures in PATCO, August 1973–July 1974

Disabling Failures2 ATC Failures3

Four-Week Total Percentage Percent ofInterval Ending Failures l Number of total Number Disabling

8/10/739/7/73

10/5/7311/2 /7311130/7312/28/73

1/25/742/22/743/22/744/19/745/17/746/14/747/12/74

Total

Average

75511611339123411971180119313991298

962110511971206

425777913835769788716839807541690682682

56.366.968.276,764.266.860.060.062.256.262.457.056.6

734795847856576953

12091

10866

17.26.1

10.410.110.1

7.18.08.26.6

22.213.215.8

9.7

15,226 9,464 62.2 997 10.5

1,171 728 62.2 76.7 10.5

Battelle calculations, based on PATCO data)IDoes not include preventive maintenance or cleaning.ZDefined by pATCO to be critical fai]ums that would require removal of a train or would prevent its return to service after leav-

ing the line at the end of its scheduled run.3Does not include communications since PATC() does not consider this disabling.

TABLE 21.-Summary of Equipment Failure in BART, 1974–751

COMPONENTNumber of Average Percent of Failures per car

failures per month total failures per month2

Carborne Equipment:ATCAir ConditioningAuxiliary ElectricalCar BodyCommunicationDoorsFriction BrakePropulsion ●

SuspensionTruck

1,295504834

1,676500598

1,3754,158

222614

1024066

1324047

109329

1849

10.94.37.1

14.24.35.0

11.735.3

1.95.3

0.350.140.220.450.140.160.371.150.060.17

TOTAL CARBORNE 11,774 932 — 33.16

Wayside ATC Equipment:ATO 339 27 21.3 N A4

ATP5 696 55 43.3 NAATS (Central) 41 3 2.4 NAPower 31 2 1.6 NASwitch & Lock 198 16 12.6 NAYard Control 299 24 18.9 NA

TOTAL WAYSIDEATC EQUIPMENT

1,604 127 — —

IThe period covered is from ]anllary 1, 1974 to January 21, 1975, 12,65 months. *z~sed on an average fleet size of 295 (145 A-cars, 150 B-cars) during the period.3Does not sum because of rounding in individual calculations.

qNot applicable.SIncludes multiplex and interlocking control equipment.

102

failures should be counted as disabling. To this,however, must be added the failures of waysideequipment, which in BART accounts for a sizableshare of the train control system. BART waysideATC equipment, including central supervisory(ATS) equipment, experiences about 127 failuresper month, the equivalent of 6 per day.60 In com-parison with carborne equipment failures, waysidefailures tend to have more widespread conse-quences because all trains operating in the vicinity(or, if a central control failure, all trains in thesystem) are affected.

Reliability of equipment has been a majorproblem in the BART system. For example, ananalysis of the operating logs for the period May1974 to January 1975 shows that only slightly overhalf of the car fleet was available for service at anygiven time and that availability declined regularlythroughout the day and week. The problem wasparticularly severe with the A-cars, which containthe train control electronics. In an average weekduring this period, only 71 of the 148 A-cars (48 per-cent) were in running condition. From Monday toFriday, availability declined by an average of 8 cars,often leaving fewer than 65 A-cars in service by Fri-day.

The extent to which ATC equipment reliabilitycontributes to the overall pattern of car problemsand service disruptions could not be determinedconclusively. The BART staff estimated that ATCwas initially cited as the reason for about 20 percentof all train removals, but the actual figure may besomewhat lower if “false bad orders” are dis-counted and only confirmed ATC failures are con-sidered. Even so, ATC is not the single largest causeof train removal. Propulsion motors, car bodydefects, and brakes each account for a larger shareof car system failures than ATC.

CTA

Automatic train control equipment in CTA con-sists of wayside signals with trip stops on someparts of the system and cab-signaled ATP on others.Since the extent of train control automation islower than in PATCO or BART, it would be ex-pected that the proportion of train removals due toATC failure would also be lower. This hypothesis

ISOBART operates only on weekdays, or about 20–21 days permonth.

cannot be conclusively affirmed because CTA doesnot maintain a formal equipment reliability recordthat would allow MTBF to be calculated directly.However, a partial analysis, performed as part ofthis study, sheds some light on the situation.

An analysis of carborne equipment reliability onthe West-South route for a representative 16-weekperiod in 1974 was performed by CTA personnel atthe request of the OTA staff, The results are shownin table 22. Because two different types of cars areoperated on this line (180 2000-series cars and 782200-series cars) failures for each are tabulatedseparately. Cab signal equipment, although listed asa single entry, is of two types-one a rather simpleand conventional design and the other more com-plex and technologically advanced.

Cab signals are the largest failure category forequipment on the West-South route, accounting for44 percent of the sample of cases reported; but thereare several factors operating here that may havedistorted the results. First, this is only a partial list-ing of failures. When considered in the context ofall equipment failures, cab signal failures wouldrepresent a lower proportion. CTA maintenancepersonnel estimate that cab signal failures accountfor no more than 20 percent of all “bad orders.”Second, it should be noted that the total of 307 cabsignal failures listed in table 22 are reported failures.Shop personnel confirmed only about 60 percent ofthis number—the remainder being either erroneousreports by motormen or intermittent failures thatcould not be duplicated in shop tests. This illus-trates the general problem of confidence inreliability statistics, where the basic data may bequestionable because of incorrect initial diagnosisor the inherent diff icul ty in t roubleshootingelectronic equipment. Third, the cab signal failuresreported here are not all disabling failures. Some aremalfunctions of nonessential features, such asburned-out indicator bulbs, that do not affect theperformance of the equipment for basic train pro-tection functions. Fourth, the West-South route wasin the process of converting to cab signal operationduring the time period considered in this sample,The general experience of CTA has been thatequipment reliability is particularly troublesomeduring the initial installation and check-out period.This is true not only of cab signals but any othernew and complex type of transit equipment in-troduced in an established system.

103

TABLE 22.—Car Component Performance on CTA West-South Route, July–October 1974

Number of FailuresFailures

COMPONENT 13 Jul. – 10 Aug.– 7 Sept.– 5 Oct.–TOTAL

4-WEEK per car9 Aug. 6 Sept. 4 Oct. I Nov. AVERAGE per week

Cab SignalsReported Defective(Confirmed)(Unconfirmed)

(:)(34)

(:;)(22)

(::)(25)

(::)(41)

307(185)(122)

(::)(31)

0.08(0.04)(0.03)

Doors2000 -Series1

2200 -SerieslAll cars

(13)(26)39

(21)

(14)

35

(29)(15)44

(29)(16)45

( 92)

( 71)

163

(23)(18)41

(0.03)(0.06)0.04

Dynamic Brakes2000-Series2200-SeriesAll cars

(19)( 4)23

(12)( 5)17

(19)( 5)24

(19)( 2)21

( 71)

( 16)

87

(18)( 4)22

(0.03)(0.01)0.02

Friction Brakes2000-Series2200-SeriesAll cars

( 7)

( l o )

17

( 7)

(14)

21

(lo)(25)

35

(19)(22)41

( 43)( 71)114

(11)(18)29

(0.02)(0.06)0.03

Traction Motors2000-Series2200-SeriesAll cars

( 3)( 3)

6

( 5)( 2)

7

( 8)( o)

8

( 4)

( o)

4

(20)( 5)

25

( 5)( 1)

6

(0.01)—

0.01

ITwo types of cars are operated : 180 200()-series cars (purchased 1964) and 78 2200-series cars (purchased 1969–70).

cars. The older equipment, despite having been inservice much longer, was five to ten times morereliable than the newest equipment—the R–44series cars. For example, the R–36 cars (purchasedin 1962) had 4,048 hours MTBF; and the R–38 cars(dating from 1965), had 2,126 hours MTFB.62 In con-trast, MTBF for the new R–44 cars was only 421hours-or about half that of the fleet as a whole.63

Preliminary indications are that the newest equip-ment, the R–46 series now being delivered, haveeven less low reliability.

This experience suggests that some of thereliability problems experienced by new systemssuch as PATCO and BART result not so much fromtrain control automation as from the general com-plexity of the newer transit vehicles, All types of

NYCTA

NYCTA has wayside signals and trip stops forATP and virtually no carborne ATC equipment ex-cept on the R–44 and R–46 cars.61 The experienceof NYCTA with equipment reliability is, therefore,a useful baseline from which to estimate thegeneral performance of car components other thanATC,

During 1974, there were 32,515 delays in servicein NYCTA, about 90 per day. Of these 16,872 (52percent) were chargeable to car equipment failure.During the same period, wayside signal failures ac-counted for only 1,435 delays, or 4.4 percent.

Us ing NYCTA da ta , Ba t t e l l e Co lumbusLaboratory estimated that the reliability of NYCTAcars was about 842 hours MTBF. However, therewas great variability among the different models of

GIThe R–44 and R–46 cars are equipped with cab signa]s; but

since the wayside equipment associated with cab signaling hasnot yet been installed, the cars are run with the cab signal unitscut out.

1 0 4

8ZThe R–36, R–38, and R–44 cars were all purchased fromthe same manufacturer.

f33The average age of the NYCTA fleet is 17 years, withalmost one-sixth having been in service over 28 years. All ofthese oldest models had an MTBF greater than that of the R–44cars.

car equipment have grown more complex over theyears. Propulsion motors, suspension systems, dooroperat ing mechanisms, air condit ioning, andcouplers are but a few of the mechanisms that havebecome more complicated and sophisticated. Thus,ATC equipment may produce reliability problems,not because of automation per se, but because itrepresents the introduction of one more complexpiece of equipment in an already complex vehicle.The general rule of reliability is that as the numberof interacting components increases, the overallreliability of the system decreases, The experienceof NYCTA, which has no carborne ATC equip-ment, confirms this point.

ISSUE O–8: MAINTAINABILITY

To what extent does ATC equipment main-tainability contribute to the general maintenanceproblems of transit systems?

ATC equipment is considered by transitmanagers to be a major maintenance problem,but probably no more so than other types of com-plex and sophisticated transit equipment. T h eproblem of ATC maintenance is difficult toassess quantitatively because of the scarcity ofdetailed data and the variety of recordkeepingmethods employed by transit systems.

Maintenance of transit system equipment is anever-ending battle. Weather conditions, hard dailyuse, and the demands of meeting train schedules alltax the ability of equipment to perform as requiredand increase the pressure to restore equipment toservice when failures occur. The promptness withwhich maintenance is performed and the effective-ness of the repair action play a role almost as impor-tant as equipment reliability itself in sustaining therequired level of service to transit system patrons,The overall importance of maintenance in thescheme of transit operations is illustrated by thefact that in most systems the maintenance force isequal to or larger than the force required to operatethe trains. Maintenance of ATC equipment ,because it is vital to the safety and efficiency oftrain operations, is of special concern.

The i n f l uence o f ATC equ ipmen t ma in -tainability on the general maintenance picture ishard to determine. Most transit systems do not keep

detailed and formal records that would allow themaintenance problems of ATC (or any otherspecific kind of equipment) to be analyzed andevaluated in precise quantitative terms. Shop logs,workmen’s time records, and repair tickets areuseful as working documents, but they do not lendthemselves to treatment as a data base for calculat-ing maintainability statistics such as mean time torestore (MTTR). The following observations,therefore, are based primarily on interviews withtransit system maintenance personnel and con-stitute largely opinion and anecdotal evidence. Thisis supplemented with a small amount of data ob-tained from BART and PATCO, where detailed andquantitative maintenance records are kept.

The general feeling among transit system per-sonnel is that ATC equipment poses especiallydifficult maintenance problems. Because this viewis widely held by those intimately acquainted withthe maintenance situation, it must be accepted.However, the data from PATCO, and perhapsBART also, do not entirely bear this out. This is notto deny that maintenance of ATC equipment re-quires substantial effort but simply to suggest thatthe size of the effort is not disproportionate in rela-tion to that required for other types of transitsystem equipment of similar complexity andreliability, An examination of the data fromPATCO will help to clarify this point.

Table 23 is a summary of maintenance time forseveral types of equipment in PATCO during a re-cent 16-week period. Maintenance time is ex-pressed in terms of mean time to restore or repair(MTTR) and as a percentage of the total mainte-nance effort. For comparison, the frequency offailure for each type of equipment is also shown,expressed as the percentage of total failures,

In terms of both average repair time (MTTR) andproportion of the maintenance effort, ATC equip-ment is not significantly different from other typesof equipment. MTTR is slightly over 3 hours forATC, the same as for the master controller and onlya few minutes longer than for the cam controller ormotor generator. It is also significant that the timerequired for ATC repairs is in the same proportionto the total maintenance effort as ATC failures areto total equipment failures.

Interviews with maintenance personnel fromother transit systems suggest, however, that thePATCO situation may not be typical, The ex-perience in these other systems, notably older

105

TABLE 23.—Maintenance Time for Selected PATCO Car Components

COMPONENT

AverageNumber of Total Repair Repair Time Percent ofFailures or Time (MTTR) All Maint- Percent of

Events (hours) (hours) enance All Failures

ATCAir BrakeCam ControllerCommunicationControllerCouplerMaster ControllersMotor-GeneratorAll Corrective Maintenance

284275288118110424

17168

5,005

881636803165270582

53449

12,007

3.12.32.81.42.51.43.12.72.4

5.13.74.60.91.63.30.32.6

69.0

5.75.55.82.42.28.50.33.4—

Periodic Maintenance 1,213 5,387 4.4 31.0 —

IData are for a 16-week period, July 1–November 1, 1974.ZHigh vo]tage switches.Soperator’s control unit in cab.

systems converting totrain control, indicates

more automated forms ofthat ATC equipment takes

longer to repair than other kinds of equipment. Thisis probably true if the comparison is made to con-ventional mechanical components. It could not beestablished how ATC repair time compares to thatfor other kinds of complex electronic equipment, inpart because there is relatively little such equip-ment in use, except for radios and some elements ofthe propulsion control system.

Several reasons are cited by maintenance person-nel to support this view that ATC equipment isdifficult to maintain. Troubleshooting and faultisolation are more difficult procedures. It may takea substantial amount of time to confirm the trainoperator’s report of trouble. Some kinds of failureare intermittent; others are difficult to reproduceunder shop conditions. Also, the description of themalfunction reported by the operator may be er-roneous or imprecise. Once the fault is diagnosed,the repair process may be time-consuming, bothbecause of the type of work required and because ofthe need to check out addit ional secondaryproblems. A recurring problem in electronic main-tenance in general, and ATC in particular, is thedifficulty in ascertaining the effectiveness of therepair. This is the so-called repeating failure. InBART, for example, it is estimated that about one-third of the cars account for over two-thirds of therepairs; and a car delivered to the shop for a specificrepair may be returned one or more times on suc-cessive days for the same reason. This has led somemaintenance managers to the conclusion thatrealistic work planning must be based on theassumption that corrective maintenance for A T Cequipment will be from 1.25 to 2 times the equip-ment failure rate,

It is widely agreed that the maintenance ofelectronic equipment, of which ATC equipment is aprime example, calls for a different type of mainte-nance skill than conventional transit system equip-ment. The human factor aspects of this problemwill be treated later in a separate issue, but it shouldbe noted here that the qualifications and experienceof the shop force have a sizable influence on thesuccess of ATC maintenance operations. Relatedproblems are the shortage of qualified maintenancetechnicians and the more extensive training re-quired to bring in new personnel or reassign the ex-isting shop force. These manpower problems areespecially keen in established transit systems goingthrough a process of installing a new ATC equip-

FIGURE 59.-Cab Signal Maintenance

ment or adding new lines. New systems tend torecognize these problems in advance and make pro-vision to solve them in the preparatory periodbefore inaugurating operations. Even so, this an-ticipatory action is not always successful, and newsystems such as BART have had trouble in acquir-ing and training a suitable shop staff for electronicmaintenance.

A related problem is that of facilities and shopequipment. The work space and tools required tomaintain electronic equipment are very differentfrom that of the conventional car shop. Most transitmaintenance is dirty, heavy work that is largelymechanical, Electronic maintenance calls for afacility more like a laboratory or television repairshop. Special tools and test equipment are needed,and many transit systems have had to build suchitems themselves because of a lack of a suitable ver-sion on the general market, Older systems like CTAand MBTA have also had to build new maintenancefacilities or remodel existing ones in response to thespecial needs arising from introduction of cab sig-nals and related ATC equipment. But here again,the problem is not peculiar to ATC but stems fromthe more general trend in rail rapid transit to con-vert to a different form of technology.

107

As a final point, it should be noted that thedesign of ATC equipment and its placement ontransit cars may aggravate the problems of mainte-nance. Access to equipment cases or individualcomponents within them may be difficult; and thetime to remove and replace an item may exceedrepair time itself. In some instances, the equipmentis not designed modularly so that defective ele-ments can be quickly replaced and the car restoredto service.

Repair of electronic equipment while it is inplace on the vehicle is generally not an efficientmaintenance strategy; but in many cases, thestrategy of on-vehicle repair has been forced onmaintenance personnel by a lack of spare parts.Nearly all transit maintenance and operating per-sonnel interviewed during this s tudy ci tedavailability of spare parts as a major problem.Several factors seems to be at work here. First, thereis the generally low reliability of new equipment;components are wearing out or becoming un-

serviceable at a much higher rate than anticipated.Second, there has been some instances of inadequ-ate provisioning of spare parts in the initial procure-ment order. The lead time for replenishing stocks isoften long, which tends to exacerbate the spareparts problem once it is detected. Third, some sup-pliers do not find it profitable to keep a supply ofitems that may be peculiar to a single transit systemor to only a single procurement order by thatsystem, Transit systems, old and new alike, havefound it increasingly difficult to locate alternativesources of supply, The shortage of spare parts is notrestricted to ATC equipment. It is a general problemin the transit industry, cited here to indicate all thefactors that influence the maintainability of traincontrol equipment,

The car availability problems that have plaguedthe BART system have received widespread atten-tion in the transit industry and in the public at large.Equipment reliability, and often ATC systemreliability, is cited as the major cause. Upon closer

108

examination, it appears that maintenance may alsobe an important part of the problem. A recentmanagement audit of BART (Cresap et al., 1975)stated that maintenance was the prime problem tobe solved by BART and recommended that ap-proval for a full 20-hour, 7-day operating schedulebe withheld until the maintenance backlog iscleared up and continued operation of the full 450-car fleet could be assured, Fig. 61 is a summary ofthe maintenance situation that existed in BARTfrom May 1974 to January 1975, roughly the periodduring which the management audit was con-ducted. These findings are offered not in order tosingle out the BART system for special criticism butonly to illustrate the impact that maintenance canhave on car availability and transit system per-formance, In this regard, the categories of “Back-logged for Corrective Maintenance” and “AwaitingParts” are particularly noteworthy, Estimates byBART officials indicate that ATC equipment main-tenance makes up 10 to 20 percent of the total main-tenance burden, a proportion roughly equivalent tothe ratio of ATC failures to all equipment failures.

Average Fleet Size:A-Cars 148B-Cars 174

Total 322

FIGURE 61.—Influence of

COST

The costs of automatic train control, both theinitial capital cost to design and install ATC equip-ment and the cost to operate a transit system withATC, raise several important issues,

1n the area of capital cost, there is a need to ex-amine the expense of acquiring an ATC system, inabsolute terms and relative to the cost of the wholetransit system. It is also important to examine theincremental capital costs associated with increasingthe level of automation from a simple ATP systemto one including ATO and ATS as well,

With regard to operational cost, the general issueis the comparative expenses of transit systemsemploying different levels of automation. Withinthis issue are specific questions relating to man-power and labor cost savings that may be derivedfrom automation. There is also the question ofenergy savings that may be achieved by the moreefficient train operation claimed to result fromATO and ATS,

Maintenance on Car Availability in BART, May 1974–January 1975

109

Ultimately, the matter of cost reduces the ques-tion of whether the greater expense required to ac-quire an ATC system can be recovered by opera-tional savings over the life of the equipment, Thismatter is important, not just because of the publicfunds involved in capital grants and operating sub-sidies, but also because advocates of automationclaim that ATC more than pays for itself in the longrun,

System Size and Configuration—miles of track,number of interlocking, number of stations andterminals, the number of trains or vehicles oper-ated, and the nature of the train consist (i. e., A andB cars, married pairs, single-car trains, etc.),

Condi t ion o f Ins ta l la t ion—ins ta l la t ion as par tof the original construction of the system or as anadd-on to a system already in service. (The latter isgenerally more difficult and expensive.)

C u s t o m i z e d D e s i g n s—the degree to which aspecific ATC installation differs from other ATCdesigns in use within the system or elsewhere andthe degree of custom engineering required to meetlocal requirements.64

ISSUE O–9: CAPITAL COSTS

What are the capital costs of automatic traincontrol ?

ATC equipment costs are roughly 3 to 5 per-cent of the total capital costs for a rail rapid tran-sit system. Ninety percent or more of the ATCcost is for wayside equipment.

Table 24 is a summary of capital costs on transitsystems recently built or now under construction,Because of the factors cited above and the effects ofinflation, the costs of these systems cannot bedirectly compared, However, the data do indicatethe general range of costs incurred in recent yearsby transit agencies building completely newsystems with advanced levels of ATC.

The capital costs of an ATC system are in-fluenced by a number of factors, primarily:

Level of Automation—the number of ATP,ATO, and ATS functions which are automated andthe degree of operational sophistication (the num-ber of running speeds, degree of supervisory con-trol, or station stopping accuracy).

64One supplier of ATC equipment estimated that specialengineering of just the speed regulation and station stoppingequipment for a new installation can cost between $100,000 and$200,000.

TABLE 24.-Capital Costs in New Transit Systems●

PATCO BART WMATA MARTAMTA

(Baltimore)

INITIAL SERVICETOTAL SYSTEM

Cost ($ million)MilesCost/Mi. ($ million)

VEHICLESCost ($ million)NumberCost/Vehicle ($)

TRAIN CONTROLAutomation Level6

Cost ($ million)

1969 1972 1976 1979 1981

113514

9.6

1,586

71

22.3

24,65098

47.4

2 2 , 1 0 0

5042

345031 5

330

15

75

200,000

143

450

318,000

199556

358,000

N A4

338NA 4

N A4

N A4

N A4

ATP, ATO, ATS2100

ATP, ATO4.5

ATP, ATO, ATS540.5

ATP, ATO, ATSN A4

ATP, ATO, ATS3 2 5

I ] n{:] I I{ifIS t:,] 1) i I il I c;{}st (If n[It\ ( ;on~ I rl It: t i on i] nd (’[] I I i preen t. N’ no ~’,] t ion, iI n(l IISI i m i] t(’(1 v, I I I I (I () f [)rII~I x i st i n~ ri~ h t -of-~v,l \ ,1 n(l

st rl I(:t I I r[)s

~Cllrrent estimate, cost Ily Comp]etlon date will probab]y b e h i g h e r .:) Estimate for phase I, I ~-mile partial system (1970 dollars).aNot available,sInc]lldes a(](i itional work; original bid was $26,2 million.GATS here means computer-aided Central control.

110

Since there are so many local and temporal fac-

tors at work, and because so few new systems have

been built, historical data on procurements in suchsystems as PATCO, BART, and WMATA and pro-jections for MARTA and MTA do not provide ameaningful picture of the capital cost of ATC. Adifferent perspective is provided by Table 25, whichcontains estimated capital costs based on interviewswith manufacturers and consultants concerning thecurrent prices (1975 dollars) of major ATC systemcomponents.

Table 25 separates ATC equipment into twocategories: carborne equipment and wayside equip-ment (including central control and ATS equip-ment). Within each category, successive levels ofautomation are identified and priced. The prevail-ing view in the transit industry today is that cab sig-nals, overspeed protection, route interlocking, amodest supervisory system, and the associated com-munications equipment represent the minimumATC system that will be installed. Thus, the firstentries in the vehicle and wayside categories of ta-ble 25 should be considered a baseline system. Ad-ditional features incur additional costs as indicated.

To obtain an estimate of the total cost of a typicalATC installation, consider the example of a hy-pothetical transit system consisting of 50 miles of

double track (100 single-track miles) and zOO car-borne controlled units (400 cars operating as mar-ried pairs with one ATC package per pair). Thetotal cost of a baseline ATC installation (ATP only)in such a transit system would be approximately$59.5 million ($57.5 million for wayside and $2million for carborne equipment) .65 This would be asystem with a level of automation roughlyequivalent to the MBTA Red Line or the CTAWest-South Line. The addition of ATO (the secondentry in the wayside and carborne categories of ta-ble 25) would raise the cost to almost $70 million($65 million wayside, $4.5 million carborne), Thiswould be a system resembling PATCO. The addi-tion of ATS, to build a system with a level ofautomation similar to BART, would raise thecapital cost to $87 million ($82.5 million wayside,$4.5 million carborne). Note that the addition ofATS does not increase the cost of carborne ATCequipment since virtually all the additional equip-ment needed for ATS is in the central controlfacility.

While the absolute cost of an ATC system maybe large, ranging up to $100 million or more for alarge system with a high level of automation, its

sSTheSe estimates assume vaiues for carborne and waysideequipment costs in the middle of the ranges given in table 25.

TABLE 25.—Cost Estimates for ATC Equipment

UNIT OF MEASUREAPPROXIMATE UNIT

C O S T

Single-Track Mile 500.000-650,000

CARBORNE EQUIPMENTCab signaling and overspeed protec - Controlled Unit2 $ 9,000-11,000

tionAbove, plus speed maintaining, pre- Controlled Unit2 18,000-25,000

cision stopping, performancelevel adjustment, and trainidentification

WAYSIDE EQUIPMENTCab signaling, overspeed protection,

r o u t e i n t e r l o c k i n g , d a t atransmission, modest super-visory system

Above, plus precision stopping, per- Single-Track Mile 550,000 -750! 000formance level adjustment,and train identification

Above, plus sophisticated ATS with Single-Track Mile 750,000-900,000computerized control

11975 dollars2A c~ntr~lled \lnit may be more than one vehicle. e.g., a married pair of cars typically has only one set of ATC eq~lipment,

(SOIJRCE: FX]ttellf’ from man~lfacturer and consultant interviews.)

‘11 1

— — —

cost relative to the total capital cost of the system islow. A rail rapid transit system typically costs $30million to $45 million per double-track mile tobuild. Transit vehicles cost in the range of $200,000to $350,000 each, depending upon their size andcomplexity. On this basis, wayside ATC equipmentrepresents something on the order to 3 to 6 percentof the cost per track mile. Carborne ATC accountsfor 5 to 12 percent of vehicle cost.

Returning to the example of the hypotheticalsystem, the total cost would be about $2 billion.66

The ATC system, depending upon the level ofautomation selected, would run between $60million and $87 million, or 3 to 5 percent of the totalcapital cost, Note that the cost increment associated

o6(50 doub]e-track miles X $35 mi]]ion per mile) + (400 cars,i.e., 200 married pairs, x $300,000) = $1.88 billion + $0.lz billion= $2 billion.

BART Elevated Guideway

Section of BART Transbay Tube

WMATA Pentagon Station

FIGURE 62.—Transit System

WMATA Judiciary Square Station

Construction

112

with selection of an ATC system with a high levelof automation instead of a baseline system withATP alone, would amount to only 2 percent or so ofthe total capital cost of the transit system. Note alsothat the bulk of the expense, either for a baseline ora highly automated ATC system, lies in waysideequipment—90 percent or more.

ISSUE 0–10: OPERATIONAL COST

How do the operating costs of systems withautomatic train operation compare to those ofsystems where trains are run manually?

The costs of operating trains are somewhatlower in systems with ATO, but the mainte-nance costs are higher. In general, ATO reducesthe proportion of personnel-related costs inoperating a transit system.

One of the purported advantages of automatic

train control (particularly automatic train opera-tion) is that it can reduce the operating costs of a

t rans i t sys tem. This reduct ion would be brought

about primarily by decreasing the number of per-sonnel needed to operate the system. The question

of workforce reduction is thus a pivotal issue thatneeds to be examined f rom severa l aspec ts . The

purpose here is to look at operating cost in general

terms to provide a background for the specific dis-

cussions of workforce reduction in the two follow -

ing issues.

Table 26 is an analysis of operating costs for themost recent year in five transit systems, Since thesesystems vary greatly in size and service level, thedata are normalized by expressing cost in terms ofdollars per revenue car mile and as percentages oftotal operating expenses for each system. Costs areallocated to three categories: transportation, main-tenance, and administration, The transportationcategory includes all costs incurred in providingpassenger service. Payroll and fringe benefits fortrain crews, central control personnel, stationattendants, and supervisors are the largest compo-nents; but the category also includes electric powercosts and all other expenses associated with transitoperations. 67 Maintenance includes all personnel-related costs for vehicle, track, signal, and struc-tures maintenance as well as the cost of materialand supplies. Administration is made up of all ex-

penses associated with management, support, andadministrative services and all general expenses notdirectly attributable to either transit operations or

m a i n t e n a n c e ,

The five systems are arrayed in an order thatrepresents an increasing level of automation, fromleft to right, but the principal distinction is betweenNYCTA, CTA, and MBTA with conductors on thetrains and PATCO and BART without. Note,however, that technology is not the only factordetermining the size of the train crew, Local labor

67Transit police expenses have been excluded since not allsystems have an internal police force.

TABLE 26.—Summary of Rail Rapid Transit Operating Costs

NYCTA CTA MBTA PATCO BART(1973/74) (1974) (1974) (1974) (1974/75)

OPERATING COST ($/revenue car mile)Transportation 1.05 0.95 2.15 0.72 0.89Maintenance 0.62 0.44 1.45 0.59 11.33Administration 0.26 0.15 1.22 0.16 0.30

Total 1.93 1.54 4.82 1.47 12,52

PERCENT OF OPERATING COSTTransportation 55 62 45 49 36Maintenance 32 29 30 40 52Administration 13 9 25 11 12

RATIO OF MAINTENANCE COST TOTRANSPORTATION COST 0.59 0.44 0.68 0.82 1.48

SALARIES. WAGES & BENEFITS AS PERCENT-AGE OF OPERATING COST 82 82 80 64 74

I For stable year operation, BART forecasts a maintenance cost of about $0,83 per revenue car mile, with transportation and ad-ministrative expenses remaining a t present levels, If the reduction of maintenan(;e is achieved, the total cost per revenue car milewould be $Z. OZ and the maintenance-transportation cost ratio would he 0,93.

113

agreements and operating philosophy also playstrong roles. Thus, any cost differences amongthese systems are not purely the result of train con-trol automation.

Examination of the revenue costs per car milereveals a wide variation among the five systems,”with no clear-cut pattern. PATCO, a system withATO and a single train operator, has the lowestoveral l operat ing cost ;68 but BART, which isequally automated in the area of ATO, has costssubstantially higher than any system except MBTA.

B~The pATcO figures are somewhat deflated in the area oftransportation and administration. PATCO stations are largelyunattended, while all the other systems have station attendants.Many administrative functions normally carried out by a transitagency are, in the case of PATCO, accomplished by its parentorganization, the Delaware River Port Authority. If allowance ismade for these factors, the transportation-related costs ofPATCO might be on the order of 80 to 85 cents per revenue carmile and the administrative costs 20 to 25 cents per revenue carmile.

Nevertheless, it does appear that transportationcosts are lowest in the two systems with ATO. Italso appears that maintenance costs are somewhathigher than in systems with manually operatedtrains. In the case of BART, this is probably a reflec-tion of the general maintenance problems that haveplagued the system and not a specific effect ofATO.

The reciprocal relationship of maintenance andtransportation costs appears most pronouncedwhen they are expressed as percentages of the totaloperating cost of the respective systems. As the pro-portion of transportation costs goes down, the main-tenance proportion rises; and the sum of the two is aroughly constant 80–90 percent of the whole.69 Thetendency of the relative cost of maintenance to in-

~gThis generalization does not hold true for MBTA, wherethe percentage of administrative costs is unusually high andwhere absolute costs are about double those of any other system,

+’.

FIGURE 63.—Winter on the Skokie Swift Line

114

crease as a function of automation also appears

when maintenance cost is expressed as a ratio oftransportation cost.

Another apparent, and logically expected, effectof automatic train operation is the lower proportionof payroll-related costs in PATCO and BART.Labor accounts for 80 to 82 percent of operating costin the systems with manually operated trains andtwo- or three-man crews. In PATCO, labor costs areonly about two-thirds of total cost—partly due toone-man operation and partly due to the absence ofstation attendants. In BART, the percentage is high-er, although still lower than NYCTA, CTA, andMBTA. BART officials forecast that the labor com-ponent will drop to something like 65 to 70 percentwhen the debugging period is passed and the main-tenance situation becomes more normal.

While some of these observed differences u n -doubtedly arise from causes not related to automa-tion, it does appear that ATO (insofar as it leads to areduction of train-crew size) has the effect oflowering labor cost and, perhaps, overall operatingexpense, This conclusion must remain tentative atthis point because the data are limited to such a fewcases. However, it deserves further examination in

the following issues, which deal more specifically

with the manpower effects of ATO,

ISSUE 0–11: WORKFORCE REDUCTION

Does automatic train controJ lead to a reduc-tion of the workforce?

Automation of train operation functions, per-mitting reduction to a one-man train crew, leadsto small but significant workforce savings.Further automation, but short of total automa-tion, has little effect.

As a concept, automation implies the replace-

ment of human labor with machines. In some cases,

a u t o m a t i o n r e s u l t s s i m p l y i n l e s s e n i n g t h eworkload for operating personnel without changing

the manning level of the system. In other cases, it

m a y b e p o s s i b l e t o r e p l a c e a h u m a n o p e r a t o ra l toge ther - -e i ther by ass ign ing a l l func t ions tomachines or by consolidating several partially auto-mated functions into a smaller number of operator

pos i t ions . The potent ia l economic advantages o fautomation are large, Rapid transit is a labor-inten-

sive system, in which personnel costs (salaries andbenefits) typically account for 65 to 85 cents of ev-ery dollar of operating expense. Clearly, even asmall manpower reduction of 10–15 percent wouldhave enormous leverage and might make thedifference between an operating deficit and break-ing even,

Historically, rail rapid transit has pursued acourse of consolidation by successively reducingthe number of conductors in the train crew, In theearly days, conductors were assigned to each car orpair of cars to collect fares and operate the doors. Asfare collection was transferred to stations and assemiautomatic and power-assisted door mecha-nisms were introduced, the conductor workforcewas reduced to one per train, with even greaterrelative reductions brought about by running longertrains. In newer systems such as PATCO andBART, t he conduc to r ha s been e l imina t edaltogether, and the door operation function hasbeen transferred to the train operator (PATCO) orautomated entirely (BART). The ultimate step is afully automated system like AIRTRANS, whichoperates unmanned vehicles.

Table 27 shows the general effect on theworkforce produced by various levels of ATC.Representative transit systems are listed by increas-ing level of automation. Because these transitsystems vary greatly in size and organizationalstructure, the data have been normalized by ex-pressing workforce as the ratio of operations andmaintenance personnel to vehicles. Personnelresponsible for administrative, support, planning,developmental engineering, station operation, sta-tion maintenance and police activities are excludedin order to confine the comparison to the area mostdirectly affected by ATC.

For MBTA, NYCTA, and CTA, where automa-tion is the least and the train crew is two or three,the employee/vehicle ratio is between 3.1 and 2.4.In PATCO, where ATO has permitted reduction ofthe train crew to one, the ratio is lower than inMBTA and NYCTA but higher than in CTA, ThePATCO ratio might be lower if PATCO were morenearly the same size as the others. There are un-doubtedly economies of scale in a large organiza-tion that cannot be obtained in a transit propertywith only 75 vehicles and 203 operations and main-tenance employees.

The more advanced level of automation repre-sented by BART does not result in a manpower

1 1 572-6 H:i ( ) - 7’ (1 - ‘1

TABLE 27.—Effect of Automation on Size of Workforce

TRAIN O&M TRANSIT EMPLOYEESSYSTEM I CREW EMPLOYEES VEHICLES PER VEHICLE

MBTA32 – 3 1,063 354 3.0

NYCTA 2 21,045 6,681 3.1

CTA 2 2,594 1,094 2.4

PATCO 1 203 75 2.7

H A R T ( 1 9 7 4 / 7 5 ) 1 1,000 350 2.9

BART (Stable Year)4 1 1,192 450 2.6AIRTRANS (1974) o 142 68 241AIRTRANS (Stable Year)4 o 122 68 1.8

1 All {]ata are for the most r(’cently (;ompleted opt?rationa] yt?ar.~lnclll{hw only personnt’1 to opc’rate an(l maintain trains, with immmiiate sllpervisors.ITra in crew consists of motorman and one train gllarti for each pair of cars,~Aft[~r (iel}(l~g ing anti transition to fill] operat iona] statlls.

r e d u c t i o n , B A R T a t p r e s e n t h a s a n

employee/vehic le ra t io about equa l to MBTA orNYCTA. In pro jec ted s tab le -year opera t ion , the

ratio will decline to a level comparable to that of

PATCO, The reason for the ra ther h igh ra te in

BART a t present i s apparent ly connec ted to theprob lem of equipment re l i ab i l i ty , which necess i -

tates a large maintenance force. Further examina-tion of this point will be deferred to the next issue,where the composition of the workforce in BART

and other systems will be analyzed,

The employee/vehic le ra t io for AIRTRANS, a

fully automated system with unmanned vehicles, isabout the same as PATCO, where there is a one-man t ra in c rew, AIRTRANS i s , however , a new

sys tem s t i l l undergo ing opera t iona l shakedown.

The present operating force includes 36 passenger-serv ice employees requi red to he lp pa t rons f ind

their way around the airport. It is anticipated that

the need for such employees will decrease once bet-

ter signing has been installed. It is also expected

t h a t t h e m a i n t e n a n c e f o r c e w i l l b e r e d u c e d a sdebugging and break-in of the equipment is com-

pleted and more operating experience is gained, It is

anticipated that the total of O&M employees would

go down to about 122 in a stable year, producing anemployee/vehicle ratio of 1.8, a figure substantially

lower than that of any manned system,

From these data it appears that ATO, insofar as it

allows consolidationfunctions in a single

of conductor and motormantrain operator position, will

produce a small but significant manpower saving.70

Automation to levels beyond the minimum re-quired for such consolidation, but short of fullautomation, does not seem to lead to further man-power savings because of offsetting increases in therequired maintenance force,

ISSUE 0-12: WORKFORCE DISTRIBUTION

What effect does automatic train control haveon the composition and distribution of theworkforce ?

As the degree of automation increases, thenumber of operation employees goes down, butthe number of maintenance employees goes up.The net result is a shift in the balance of theworkforce without a substantial decrease in thetotal O&M force.

ToTransit system professionals point out that automation isonly one factor influencing the size of the train crew. Unionagreements and work rules, especially in established transitsystems, may play a part in keeping the conductor on the traineven though the train could be satisfactorily operated by oneperson at the existing level of automation. In some circum-stances, transit system management officials may also concludethat the conductor position should be retained for reason ofpassenger safety in emergencies or as a way of offering informa-tion and other assistance to patrons on long trains.

1 1 6

CTA Train With Conductor

BART Train Without Conductor

FIGURE 64.—Reduction of Train Crew

In the discussion of the preceding issue, it was

c o n c l u d e d t h a t a u t o m a t i c t r a i n o p e r a t i o n ( A T O ) ,

insofar as it permits reducing the train crew 10 one,p r o d u c e s a s m a l l d e c r e a s e i n t h e t o t a l O & M

workforce. This decrease, however, is not commen-

sura te wi th the number o f conductor pos i t ionseliminated, It is therefore necessary to examine the

composition and distribution of the workforce atvarious levels of automation to see what counter-vailing effects are at work.

Figure 65 shows the relative size of the opera-tions and maintenance forces in five transitsystems. To illustrate the effect of full automation(i.e., elimination of all on-board personnel), similarfigures are also given for AIRTRANS, even thoughit is not a true rail rapid transit system. Operationsemployees are all those necessary to operatetrains-dispatchers, trainmasters, stationrnasters,

t o w e r m e n , c e n t r a l c o n t r o l l e r s , a n d y a r d m o t o r m e n

as well as the train crew i tself . Maintenance person-

n e l i n c l u d e t h e e m p l o y e e s i n c a r s h o p s , a n d t h o s e

n e e d e d t o m a i n t a i n w a y , p o w e r , a n d s i g n a l s . The

size of the operat ions and maintenance forces is ex-

p r e s s e d a s a p e r c e n t a g e o f a l l e m p l o y e e s f o r t h e

r e s p e c t i v e t r a n s i t systemsoTl

While there is considerable variation in the data,there does appear to be a discernible trend. Readingfrom top to bottom, as train operation generallybecomes more automated, the proportion of opera-tions employees declines while the proportion ofmaintenance employees shows a corresponding in-crease. It appears that ATO results primarily in ashift of the balance of the O&M workforce butwithout significantly changing its size in relation tothe total workforce. More specifically, as conduc-tors and finally the operator are taken off the train,almost equal numbers of new jobs are created in thecar shops and wayside maintenance crews.

A more detailed analysis is presented in table 28,where the workforce in the operations and mainte-nance departments is expressed in terms of thenumber of employees per car. The number of opera-tions employees per car generally declines from1.2–1.4 for systems with manual train operationand a crew of two (NYCTA, CTA, and MBTA) to0.3 for a fully automated system (AIRTRANS).PATCO and BART, with a train crew of one, fallabout midway between. At the same time, themaintenance force increases from 0.8 per car inCTA to 1.8–2.0 for BART and AIRTRANS in thecurrent year.TZ The same trend shows up even moreclearly in the ratio of maintenance to operationsemployees, wherr thrr(] is {i thr[v~ ft)lt I to {[’n felt Idifference between manned systems without ATO(NYCTA and CTA) and the unmanned AIRTRANSsystem, with PATCO and BART falling at roughlyproportional intermediate points.

plTransit police and construction personnel are excluded.TZEstimates of stable year operations for both systems project

a decrease in the ratio of maintenance employees per car to1.5–1.7, a figure comparable to PATCO,

1 1 7

NYCTA (1974/75)

CTA (1974)

MBTA (1974)

PATCO (1 974)

BART (stable yr.) 2

AIRTRANS (Stable yr.) 2

1.2

Excludes transt police and construction personnelEstimated staffing when debugging IS completed and the system becomes fully operational

FIGURE 65.—Proportion of Operations and Maintenance Employees in Total Workforce

TABLE 28.-Analysis of Operations and Maintenance Workforce

BART AIRTRANSNYCTA CTA MBTA PATCO(1974/75) (1974) (1974) (1974) 74/75 S.Y. l

74 S.Y.1

Total Cars in Fleet 6,681 1,094 354 75 350 450 68 68O&M Employees 21,045 2,370 1,063 203 1,000 1,192 142 122Operations Employees 8,350 1,540 482 75 315 415 22 22Maintenance Employees 12,695 830 581 128 684 777 120 100O&M Empl./Car 3.1 2.2 3.0 2.7 2.9 2.6 2.1 1.8Ops. Empl/Car 1.2 1.4 1.4 1.0 0.9 0.9 0.3 0.3Maint. Empl/Car 1.9 0.8 1.6 1.7 2.0 1.7 1.8 1.5Maint. Empl./Opr. Empl. 1.5 0.5 1.2 1.7 2.2 1.9 5.5 4.5

IEstimated stable year operation.ZTrain crew, dispatche~, towermen, central control, and yard motormen.sMaintenance of vehicles, way, power, and signals.

The differences among these systems are notsolely attributable to ATC. A large share of themaintenance force (67–93 percent) is not concernedwith ATC equipment but with other carborne andwayside components, which also tend to need moremaintenance as the transit system becomes morecomplex or equipment and structures grow older.Still, the percentage of maintenance employees in-volved in ATC-related activities shows a generalincrease proportionate to the level of automation.

In NYCTA, with no carborne ATO equipmentand all ATP in the wayside, it is estimated that 1 0

percent of the maintenance force performs ATC-related work (primarily signal maintenance). Theestimated figure for CTA is about 5 percent, abouthalf for wayside equipment and half for cab signals.For MBTA the figure is now 7 percent, but expectedto increase as cab signals are installed on otherlines. PATCO, with cabs-signaled ATP and ATO,has about 15 percent of the maintenance force dedi-

118

cated to ATC (9 percent wayside, 6 percent cars),For BART, the ATC position of the maintenanceforce is now about 31 percent (18 percent forwayside and central ATS equipment, 13 percent forcars)--a distribution that is expected to remain es-sentially the same when stable-year operation is at-tained. From these data, it appears that the progres-sion from ATP (either wayside or cab signals) toATP and ATO results in a doubling of the percent-age of the maintenance force assigned to ATC ac-tivities. The increase to a system with ATP, ATO,and ATS (if BART is typical) causes the percentageto double again.

If the PATCO and BART cases are assumed to berepresentative of the manpower shifts that resultfrom automation of the train control system, it ispossible to draw some tentative conclusions aboutcost savings attributable to ATC. For PATCO, theincorporation of ATO made it possible to run thetrains with a single operator, resulting in theelimination of about 45 conductor positions. At thesame time, about 15 additional shop and wayside

NYCTA(ATP, wayside signals)

ATC CAR O%

personnel were required to maintain ATC equip-ment. This is a net of 30 fewer employees.However, the pay rate for personnel skilled in ATCmaintenance is generally higher than that for con-ductors. Assuming a pay differential of 20 percentfor ATC maintenance workers, the effective savingin payroll costs reduces to about 25 positions, orroughly 9 percent of the annual payroll. Following asimilar line of reasoning, the BART ATC systemeliminated the need for about 315 conductors, butadded about 200 to the maintenance force, a net of115 fewer positions.73 Adjusting for maintenancepay differential, this is equivalent to a saving ofabout 75 positions, or roughly 4 percent of the an-nual payroll. Since labor costs are about three-quar-ters of all system operating costs, ’these calculationssuggest that automatic train operation with a crewof one offers the potential to reduce operating costssomewhere between 3 and 6 percent per year.

73The BART estimate assumes stable-year staffing levels.

FIGURE 66.—Distribution of Maintenance Force as a Function of Automation

1 1 9

One of the arguments often advanced f o rautomatic train control is that ATO and ATS canlead to a more efficient mode of train operation and,hence, lower energy consumption. It is assertedthat, in an automatic system, trains can be run atmore uniform headways and at predeterminedspeed-distance patterns, which provide lower max-imum speeds and more uniform accelerating andbraking rates. This yields a lower power consump-tion per car mile as a direct effect. The moreuniform spacing of trains brought about by operat-ing at optimum conditions also has an equalizingeffect on the passenger load of trains, and in turnproduces more energy savings as a secondarybenefit. More uniform headways also shortenlayover times at terminals, permitting a reduction inthe number of trains operated and still furtherenergy savings. (Irvin and Asmus, 1968)

Theoretically, this argument is sound; but it isdifficult to test its practical validity and to assess themagnitude of energy savings that might actually be

achieved in revenue operations with various formsof ATC. Table 29 is a summary of the energy con-sumption in the five transit systems considered inthis study. Energy usage is expressed in terms ofkilowatt-hours per revenue car mile and perpassenger mile. The latter figure is perhaps the bet-ter index for comparing energy consumption amongthe five systems because it is independent of vehi-cle seating capacity and load factor.

Note that the power consumption figures aresystemwide totals, including traction power and allother uses such as vehicle lighting and air condi-tioning, station operation (lighting, escalators, etc.),parking lots, and maintenance facilities. A purerform of comparison would be the energy requiredfor traction power alone, but such figures could notbe accurately derived from the records of somesystems. Thus, there is some distortion of the datadue to factors other than train operation, but theirinfluence is probably not large since traction powerrepresents the dominant share of all energy use(typically three-quarters or more) .74

The data in table 29 do not indicate differencesamong transit systems that appear to be related toATC. With the exception of CTA, the energy con-sumption per passenger mile is about the same forall systems, regardless of the level of automation. Inshort, there is no conclusive evidence that ATCsaves energy, at least when energy use is measuredat the overall system level.

TQ1n BART, for example, traction power amounts to about 75percent of all power use. 1n PATCO, traction power is 85 to 90percent of the total.

TABLE 29.—Rail Rapid Transit Energy Consumption

ENERGY CONSUMPTIONNYCTA CTA MBTA PATCO BART2

(1973/74) (1974) (1974) (1974) (1974/75)

ANNUAL KILOWATT-HOURS 2055.0 256.2 102.4 39.3 197.9(million)

ANNUAL REVENUE CAR MILES 320.6 46.8 10.3 4.3 21.6(million)

ANNUAL PASSENGER MILES 35480,() 775.2 4 2 6 3 . 2 95.0 446.4

(million)KWH/REV. CAR MILE 6.4 5.2 9.9 9.2 9.2

KWH/PASSENGER MILE 0.38 0.33 0.39 0.41 0.44

lpower consumed for all purposes (traction, station operation, shops, etc.).2Estimate based on operating data for July 1974 to January 1975.:jEst i matr hawi on average trip I cng th of 5 m i 1(:sQEstimate bawd on average trip length of 3,1 mi]es.

120

There may indeed be energy savings due to ATC,but they cannot be discerned by the methodsemployed here. In all probability they are small andmasked by several other factors which account formost of the observed differences among the fivesystems. For example, these transit systems differgreatly in their maximum operating speed andaverage line speed. The two systems with ATO(PATCO and BART) also happen to run trains athigher speeds. Since power consumption variesdirectly as a function of speed. the possible energysavings due to ATC in PATCO and BART are prob-ably offset by the increased energy required to runtrains at 70–75 mph.

The weight of the vehicle has a profound effecton the amount of traction power required to movetrains. There is great variation among transitsystems in the weight of vehicles, and this factoralone probably accounts for most of the differencein power consumption. In this regard, it is signifi-cant that CTA (with 20- to 24-ton cars) has thelowest level of energy use and PATCO (39-ton cars)has one of the highest.

It should also be noted that several other factorsinfluence power consumption. Among these are theaerodynamic properties of vehicles75, route charac-

75The amount of aerodynamic resistance to be overcomevaries according to whether the train is operating in a tunnel, onelevated structure, at grade, or in a cut.

FIGURE 67.—State-of-the-Art Car

121

. - — . — — — —

tained by man, he becomes less an operational ele-ment of the system and more a monitor, overseer,and back-up for automatic elements, which them-selves carry out the direct functions of train control.

Up to this point in the discussion of operationalexperience with ATC, automation has been treatedprimarily in terms of machine performance andengineering concerns. To complete the picture, it isnow necessary to examine the inverse subjects ofthe role of man and the effects that automation pro-duces upon the humans who, perforce, remain anintegral part of the train control system. There aretwo major questions here. First, there is the need toexamine whether man is used effectively and pru-dently in systems with various levels of ATC. Whatuse is made of man’s performance capabilities? Isadequate attention given to human needs as opera-tor and supervisor? Is man well integrated into thesystem? The second major concern is the conse-quences that have resulted from the application ofautomation in transit systems. Specific matters ofinterest are changes in working conditions and jobqualifications for transit system employees and thesecondary effects that ATC may produce for theriding public using the transit system,

ISSUE O–14: THE HUMAN ROLE

Is effective use made of man in systems withATC?

In some cases, new transit systems with ATCdo not make effective use of the human operatorto back up or enhance automatic systemperformance, and human involvement in nor-mally automatic processes tends to degradeperformance, primarily in terms of speed, head-way, and level of service. In systems now underdevelopment, there seems to be a greater concernfor the role of man and for making the ATCsystem more amenable to human intervention.

In considering the role of humans in systemswith automatic train control, it is necessary to dis-tinguish among the parts played by man in each ofthe major functional categories: train protection,train operation, train supervision, and communica-tion.

In train protection (ATP), the motormen (andconductors, if there are any) customarily perform

very few functions, except in a back-up or emergen-cy capacity. Nearly all transit systems have eitherwayside or cab signal equipment that automaticallyassures train separation and prevents overspeed,The human operator’s tasks are track surveillance(for detecting persons and obstacles on the right-of-way or as a back-up to track circuits for detectingother trains) and emergency braking in unusual cir-cumstances that the ATP system is not designed todetect. The operator also acts to restore the systemto operation in the event of ATP system failure,performing such tasks as emergency brake release,key-by, or manual route request. Since the operatoris backing up a highly reliable system, there are sig-nificant problems in maintaining proper vigilanceand alertness, There is also considerable risk ofhuman error in cases where the ATP system is notfunctioning properly or has been deliberatelybypassed (as when closing in on a disabled train).

FIGURE 68.—Student Conductors Trainingon the Job

In the area of train operation (ATO) there is widevariation among transit systems in the tasksassigned to the on-board operator, In NYCTA,CTA, and MBTA (except the Red line) all trainoperation functions are performed manually. InPATCO, only door operation and train starting aremanual in normal circumstances. In BART, all trainoperation is automatic. The role of the on-boardoperator in systems with ATO is mainly limited to

122

monitoring automatic equipment performance, act-

ing as a back-up in the event o f mal func t ion or

emergency, and—in some cases—adjusting the per-

formance level of the ATO system (e.g., by modify-ing speed-acceleration profiles or by ordering the

train to run by a station without stopping). The ma-

jor human per formance prob lems tha t have beenencountered in regard to ATO are the effectiveness(and sa fe ty ) o f manual in te rvent ion in normal ly

automated processes and the adequacy of the con-

trols and displays provided to the operator for pur-

poses of monitoring or manual takeover.

Train supervision embraces a number of diversefunctions, mostly carried out at a remote, centrallylocated facility. Here, too, there is wide variationamong transit systems in the degree of automation.At one extreme virtually all functions except traindispatching are manual operations. At the other ex-treme, scheduling, dispatching, route selection,traffic regulation, and documentation of events arecarried out by automatic devices either wholly orprimarily. Because the supervisory facility is thenerve center of the transit system, there can be sig-nificant workload problems for supervisory person-nel, particularly during rush hours or emergencysituations. These problems may be aggravated insystems with ATS if there is a breakdown ofautomatic equipment or the need for extensivehuman intervention in response to unusual condi-tions to which the computers are not programed torespond. The major difficulties that have been en-countered in systems with ATS are the quality andtimeliness of information available to central con-trol personnel, the flexibility of automatic systemresponse in abnormal or emergency conditions, andthe ability of humans to assume the burden of mak-ing and implementing decisions in areas normallyassigned to machines.

While there has been some automation of thecommunicat ion process in the newer t ransi tsystems, primarily in the area of data transmission,the major thrust of technological innovation has

FIGURE 69.—Motorman at Work been to-provide train crews and central supervisors

FIGURE 70.—Line Supervisors

123

withtion.how

—

communicating with remotely located transitsystem employees.

Table 30 is a summary of the allocation of tasksto men and machines in several operating anddevelopmental transit systems. The table also indi-cates man-machine allocations in AIRTRANSwhich, although not a true rail rapid transit system,may be considered representative of the extent thatpresent technology can go in achieving a fully auto-mated train control system. The systems have beenarrayed in a generally increasing order of automa-

more extensive means of voice communica-The major problems encountered have beento manage communication networks of in-

creased size and complexity, how to limit unneces-sary or excessive exchanges (chatter), and how toimplement various modes of selective and generaladdress. There has also been a general concernabout the ability of improved communicationsystems to compensate for the fewer number of on-board personnel in providing information and in-structions to passengers in special or emergencysituations and in affording passengers a way of

TABLE 30.—Man and Machine Roles in Rail Rapid Transit Systems

KEY: A = AutomatedM = Manual— = Not provided

SITUATIONTRANSITSYSTEM

MMM

MMM

MMM

MMM

MMM

MMM

MMM

MMM

——

M

AAM

AAM

AAM

AAM

AAM

AAM

AAM

AAM

AA

AIM

AA

AIM

AA

AIM

AA

A/M

AA

AIM

AA

A/M

AA

AIM

AA

AIM

AA

AIM

AA

AIM

AA

AIM

AA

A/M

AA

AIM

AA

AIM

AA

AIM

AA

AIM

AA

AIM

AA

A/M

AA

A/M

MMM

MMM

AAM

AMM

AA/M

M

AAIM

M

AAM

AAM

AAM

MMM

MMM

MMM

AMM

AAIM

M

AAIM

M

AAM

A ,AM

AAM

MMM

MMM

MMM

MMM

MMM

AAIM

M

AMM

AAIM

M

A—

M

MM—

MM—

MM—

MM—

MM—

AA—

AAIM

—

AA—

AA—

MMM

MMM

MMM

MMM

AMM

AAM

AAIM

M

AAM

AAM

NormalAbnormal 1

Emergency 2

NormalAbnormalEmergency

NormalAbnormalEmergency

NormalAbnormalEmergency

NormalAbnormalEmergency

NormalAbnormalEmergency

NormalAbnormalEmergency

NormalAbnormalEmergency

NormalAbnormalEmergency

NYCTA

CTA

MBTA 3

PATCO

MTA 4

(Baltimore)

WMATA 4

MARTA4

BART

AIRTRANS

IModerate delays, bad weather, unusually heavy demandzMajor delays, accidents, failure of critical equipment.:] Red Line only.dl.Jnder development, not yet operational.

124

tion to facilitate seeing the overall pattern ofreplacement of the human operator by automateddevices,

It can be seen that the general effect of increasedautomation is for machines to assume a greater andgreater share of operational functions in normalsituations and in certain off-normal conditions, butnot in emergency conditions. At every level ofautomation, man remains the primary means ofsustaining system operations under extreme condi-tions and the back-up element in the event ofequipment failure. On board the train, the result ofautomation is diminished importance of operatingskills for the train crewman and increased emphasison the ability to monitor automatic equipmentfunctions. Man’s primary job is not running thetrain but overseeing train operation and interveningwhen necessary. At central control facilities,automation results in more routine decisionmakingbeing allocated to machines, which monitor trafficflow, adjust schedules to compensate for ir-regularities, and alert supervisory personnel whenspecial action outside the bounds of computerprograms is required.

The conversion of train control from a manual toan automated process has produced problems onboth sides of the man-machine interface. Theseproblems arise not from any inherent inadequacy ofautomation technology as such; almost any level ofautomation is a technically viable solution. Instead,the problems stem from within the design of par-ticular systems and from the way in which the man-machine interface is engineered. The following arespecific examples of successful and unsuccessfulaspects of automated equipment design drawn fromthe experience of transit systems with operationalATC systems.

Train Protection

ATP equipment has proven to be highly reliable;but, in a way, this reliability has also createdproblems, Train operators tend to take ATP forgranted. ATP equipment operates so well so muchof the time that the operator is inclined to neglecthis responsibilities as a monitor and back-up and toforget what he must do to safeguard the train whenATP equipment is inoperative or when it has beenpurposely bypassed. The general experience oftransit systems is that accidents tend to occur whentrain operators revert to visual observation and

rules of the road because the normal automaticmethods of train protection are inoperative.76

A related problem arising from ATP (and fromhighly automated forms of train operation) is that ofvigilance. At first glance, it would appear thatrelieving the train operator of most routine and bur-densome tasks would produce a near-ideal situa-tion, in which he would be free to concentrate on afew surveillance and monitoring tasks and performexcellently in that role, Unfortunately, the result isalmost always the contrary. Given too little to do,one tends to lose vigilance and to exhibit problemsof motivation. For a person to remain vigilant, theevents to be observed must occur with reasonablefrequency. To keep a person motivated, theassigned tasks must be demanding enough to pre-vent boredom and meaningful enough to engage at-tention. “Make-work” tasks, or those perceived assuch, are not satisfactory. The individual must feelthat he has a useful and important role to play.Duties should not appear to be vestigial to machinesor compensatory for their inadequacies. (TSC, 1974)

Train Operation

One of the operator’s primary duties in systemswith ATO is to intervene whenever either equip-ment performance or operational conditions falloutside prescribed limits. In some cases, however,the act of manual intervention results in a furtherdegradation of system performance. For example,in BART where trains are normally operatedautomatically, the design of the system effectivelyprecludes the operator from assuming manual con-trol without causing a delay in service. Train speedin a manual mode of operation is limited to so per-cent of the speed allowable under ATO. Thus,manual takeover inevitably results in a slowing ofthe train and, as a consequence, following trainsalso. Furthermore, taking over manual control re-quires that the train first be brought to a full stop,thus compounding the delay. There is no tech-nological or human impediment to operating transitvehicles manually at high speed or to changing froman automatic to a manual mode while the train is in

TOThe collision of MBTA trains in August 1975 occurred injust such a circumstance. A train operating under line-of-sightrules entered a tunnel and struck a leading train waiting to bekeyed by a defective trip stop, A third train, also operatingunder line-of-sight rules, struck the rear of the second trainabout 2 minutes later.

125

motion. The PATCO ATO system permits a man tooperate the train at full speed, and the WMATAsystem will also, because it was recognized duringthe design process that such was a desirable way forman to augment the performance of an automatedsystem.

The PATCO system also incorporates otherfeatures that promote effective cooperation be-tween man and machine in maintaining the desiredlevel of service. One of the train operator’s respon-sibilities is to help complete the trip on time in casethe ATO equipment should fail. Because failure ofthis sort is not expected to occur often, it is neces-sary to devise a means for the human operator tomaintain his manual skills so as to be able to per-form at his best when needed. In PATCO this isassured by an operating rule that requires eachoperator to make one trip per day in the manualrunning mode. The skill thus maintained also helpsin other circumstances, such as when rails are slip-pery. ATO system performance is not as good as inmanual operation in this condition. Thus, a com-bination of equipment design and procedures per-mits the system to make effective use of the humanoperator as a means of enhancing the performanceof automated equipment. This lesson is being ap-plied in the design of new systems such asWMATA and MTA.

The display of information to the train operatoris an aspect of design that has been somewhatneglected in transit systems. Speed regulation is animportant operator duty on manually operatedtrains, and yet there is no speedometer in the cab totell the operator his actual speed, except in systemsthat have cab signals or ATO. Even with cab sig-nals, the human factors of information display arenot always given proper attention. For example, theBART operator’s console originally contained onlyan indicator of actual speed. The command speed,with which actual speed is to be compared, was notdisplayed, A command speed indication was lateradded, but as a digital readout.77 This form of dis-play does not facilitate the operator’s speedmonitoring task since it requires making com-parisons between two digital indicators, each ofwhich may be changing rapidly. There is a con-siderable body of human factor research that indi-

ppThe WMATA operator’s console has indicators of com-mand and actual speed, but also in digital form-an example oflearning part, but not all, of the lesson to be gained from the ex-perience of others.

1 2 6

catestrend

digital displays are difficult to interpret forand rate of change, factors which are as im-

portant as speed itself in monitoring the relation ofcommand and actual speed, An analog indicator,such as a conventional automobile speedometer, isgenerally a much more effective and informativedisplay for such purposes.

Train Supervision

Train supervision is an area where, historically,there has been very little automation, except fortrain dispatching. All operating transit systems, ex-cept BART, supervise train movement by largelymanual methods. In BART, the central computerhandles tasks such as traffic regulation (scheduleadjustment) and performance level modification.The train control systems under development inWashington, Atlanta, and Baltimore will incorpo-rate similarly automated forms of train supervision,

ATS poses several design problems relating tohuman factors. One important concern is what to dowhen the computer fails, An abrupt change fromautomatic to manual supervision can cause majordisruption of service and may even affect the safetyof transit operations. Attention is being given to thisproblem in the design of the new systems(WMATA, MARTA, and MTA) and in planning forthe addition of ATS to NYCTA, One solution is todesign ATS equipment so that it does not failabruptly and absolutely, but gracefully (i.e., in slowstages) and with sufficient coast time for humansupervisors to assess the situation and decide on anappropriate course of action, New systems are alsoproviding for intermediate levels of operation be-tween manual and automatic. These modes allowthe ATS system to operate under manual inputs orto serve as an information processing aid to humandecisionmaking. The ATS system for MARTAis being implemented in two stages — semiauto-mated first and fully automatic later. After the sec-ond stage is implemented, the first will be retainedas a back-up mode, a training device, and a meansfor central control personnel to retain manual skills,

Central supervisory systems, both manual andautomated, also exhibit the deficiencies of displaydesign noted earlier in connection with operator’scab equipment. Some systems do not have any formof central display board (model board) to allowsupervisors to monitor the progress of trains. Per-sonnel are required to form a mental picture of the

situation on the line by monitoring verbal reportsfrom trainmen, towermen, or station dispatchersand by reading pengraphs or other such nonpictorialindicators. In systems that do have model boards,the supervisor’s task is somewhat easier since thereis a large diagrammatic representation of the tracklayout with lighted indicators to show train loca-tion. Sometimes, however, the model board doesnot indicate track occupancy block-by-block but forlonger sections of track. If there is a stalled train, forexample, the supervisor may know from looking atthe board only that it is between one station andanother but not precisely where. If a following trainis ordered to close up and push the stalled train to astation or siding, the central supervisor cannotfollow the progress of this operation by means ofthe display board. The central control facilitiesbeing designed for WMATA and MARTA will in-corporate special displays that allow supervisors to“zoom in” on selected sections of track or to call updisplay modes of differing levels of detail to suit thetask in hand.

This brief review of human factors problemsassociated with existing ATC installations is not in-tended to be exhaustive nor to single out particularsystems for praise or criticism. The purpose is onlyto indicate the general range of problems encoun-tered and to illustrate the need for more attention tohuman factors in the design of ATC systems.

Neither the recently built systems with ATC(PATCO and BART) nor those now under develop-ment have had a formal human factors program.This is not to suggest that the role of man was notconsidered by the planners and engineers, but thereis no evidence that an explicit and systematicanalysis of human factors was made a part of thedesign process. An exception to this general findingis MARTA, where periodic design reviews arebeing conducted by a team from DOT Transporta-tion Systems Center. This team includes humanfactors specialists, and their examination of pro-posed MARTA designs has led to several sugges-tions for integrating man more effectively into thesystem.

Proper attention to the role of the individual inATC systems can have substantial benefits for tran-sit operations. If automation is approached not as aquestion of how to replace the operator in the traincontrol system but as how to make best use of thishighly valuable human resource, the safety andefficiency of ATC systems can be greatly improved.Man is particularly valuable as an element of a real-

time control system because of his versatility, flex-ibility of response, and ability to deal with theunexpected or the unusual. To attain these advan-tages, however, man must be made a partner in thesystem. His job must not be treated as anafterthought or as the residue of functions thatequipment engineers have found technically oreconomically impractical to automate.

ISSUE 0-15: EFFECTS OF AUTOMATIONON EMPLOYEES AND PASSENGERS

What impacts does train control automationhave on transit system employees and onpassengers?

For employees, especially maintenanceworkers, ATC results in higher job qualifica-tions, more extensive training, and more de-manding performance requirements. Forpassengers, the effects are negligible except in-sofar as ATC influences the quality of service.

There have been no studies of the specific effectsof automation in rail rapid transit systems either foremployees or passengers, despite the obvious im-portance of these topics in the overall assessment ofthe social impacts of new technology. What follows,therefore, is based on anecdotal evidence and inter-views with transit system managers. The ap-plicability of these observations to transit systemsas a whole is hard to determine. The experience ofeach operating agency is somewhat unique in thatlabor conditions, workforce makeup, personnelpolicies, and operating history vary from site to site.New transit systems, like PATCO and BART, haveno previous experience with nonautomated opera-tion against which to judge the effects of ATC. Theinstallation of ATC equipment in older systems,such as MBTA or CTA, is both limited in scope andrelatively recent. For these reasons, comparisonsamong systems or within systems for before-and-after effects cannot be made. The comments offeredhere are therefore general in nature and confined tothose effects most frequently cited by system opera-tors and managers.

Operations Employees

A primary result of the automation of trainoperation functions is a general shift in the skill re-

127

quirements for trainmen. The motor skills, coor-dination, and knowledge of signals and rulesneeded to operate a train manually are still impor-tant qualifications, but they are no longer the soleconcerns. The role of ATO system monitoring andback-up places additional requirements on theoperator—knowledge of how the system operates,ability to interpret failure indices, skill in diagnostictechniques, and an understanding of how aidautomatic system operation without necessarilyassuming full manual control. Thus, the repertoryof operator performance tends to be larger insystems with ATO, and the modes of response morevaried.

The selection criteria for train operators do notappear to differ substantially for systems with orwithout ATO, and they are about the same for busoperators in those systems that operate both modesof transit.78 The general requirements are physicalfitness and the common standards of employability(checks of police record, retail credit, and previousemployment). Educational background (above acertain minimum level of schooling) and aptitudetests do not figure in the selection process, either formanual or automatic systems. Thus, ATC does notappear to alter the basic level of qualification forinitial employment as a train operator.

While employment qualifications are unaffectedby automation, there does seem to be a longer train-ing program for operators in systems with ATC.The longer program results not so much from aneed for more intensive training as from a need tocover a greater range of subjects. This is probably adirect consequence of the wider repertory of jobskills required of operators in systems with ATO.

Since manual train operation is not a regular partof the job, systems with ATO have found it neces-sary to provide opportunities for practice and to testoperators periodically to determine if manual skillshave been retained. There is no evidence that trainoperator performance standards are more exactingat one level of automation than at any other, exceptinsofar as systems with ATO call for a widervariety of job knowledge.

Train supervisory personnel appear to be verylittle affected by ATS. Selection criteria, training re-quirements, and job performance for dispatchers

713VVMATA, which now Opemtes a bus system and is prepar-

ing to start rapid tmnsit operations, is =king to recruit trainoperatora from its bus driver force.

and line supervisors are about the same for all therapid transit systems surveyed. The BART traincontrol room, because of the use of computers forsupervisory functions, has employees versed incomputer operation and maintenance-a class ofemployee not found in other transit systems. Forthese employees the skill, training, and perform-ance requirements are, of course, unique and,because of their special expertise, somewhat higherthan other types of supervisory employees.

Maintenance Employees

The major impact of ATC upon transit systememployees is for maintenance workers, Tradi-tionally, the signal maintainer was a person whohad good mechanical skills and a basic understand-ing of the theory and operation of electromechani-cal devices (especially relays). This worker tendedto be a generalist, in the sense that he was capableof dealing with all types of signal system failure andrepair, The installation of more advanced forms ofATC and the technological shift to solid-state logicand printed circuit boards has brought about achange in the type of maintenance employeeneeded and in the organization of the maintenanceforce. New and more specialized skills are required,and the organization has become more hierarchicaland segregated into specialty occupations. Transitvehicle maintenance has come to be more and morelike aircraft maintenance.

The electronic nature of ATC equipment hasmade the diagnosis and repair of malfunctions amore complex and demanding task. Typically, thistask is divided among maintenance specialists, withthe first-line maintenance worker responsible onlyfor identification of the fault and replacement of thedefective module as a whole. Isolation of the faultto the component level may not be the respon-sibility of the first-line worker. This part of themaintenance task may be assigned to a second levelof worker, who may repair or replace the failedcomponent or who may isolate the fault further andpass a particular element along to a third level ofmaintenance worker specializing in that type ofrepair.

An additional task assigned to maintenance per-sonnel in systems with ATC is that of configurationcontrol. During the period following the introduc-tion of new equipment or the opening of a newsystem, equipment modifications are made fre-quently. Because of the strong interdependency of

1 2 8

Track Gang Electronic Repairmen

Cab Signal Maintainer Wayside Equipment Maintainer

FIGURE 71.—Transit System Maintenance Workers

129

—.————.——-

components that characterizes most new andsophisticated transit system equipment (of whichATC equipment is only an example), it is becomingmore and more necessary to maintain extensive andaccurate records of exactly what equipment is in-stalled on a given car at a given time. The main-tainer must spend more time with service bulletinsand maintenance documents in order to keepabreast of configuration changes.

In the area of maintenance, new transit systemshave some special human factors problems that arenot shared by established systems. In an establishedtransit system there is already a maintenance forcein being and procedures and techniques for main-taining the equipment are familiar to all. The in-troduction of a new item, such as cab signals, dis-rupts the pattern somewhat but only for a small partof the maintenance force since the rest of the equip-ment is unchanged. In a new transit system, every-thing is new. The equipment itself may be a newdesign or, at least, new in its specific application.Workmen and supervisors are likely to be inex-perienced in maintaining transit equipment-of alltypes, not just ATC. Procedures are untested andunrefined by experience. The facilities themselvesare usually sized to handle normal workloads ratherthan the huge influx of failures and repairs that mayoccur during start-up. Manufacturers’ representa-tives may be working alongside the maintenancestaff making equipment modifications or assistingin debugging. The training system for preparingnew maintenance workers may not yet be function-ing smoothly. These conditions may result in an im-pairment of worker efficiency, quality controlproblems, and—if they persist--a lowering ofworker morale.

There are also long-term effects on the mainte-nance force produced by ATC. The size andorganization of the workforce, as noted earlier, aredifferent. Generally more workers are needed, withspecial skills, and with a more elaborate division ofresponsibilities. The qualifications for employmentas an ATC technician are usually higher and morespecialized than for other types of transit mainte-nance workers. The period of training, both inclassrooms and on the job, is often longer. The per-formance requirements on the job may also be morestringent, Existing transit systems that are convert-ing to some form of ATC have had difficulty infinding qualified personnel, and efforts to recruittrainees within the existing transportation or main-tenance forces have not always been successful.

Bringing in new personnel from the outside is analternative, but the training period may be longersince they are unfamiliar with transit equipment-adisadvantage that may be partly offset by the betterbasic skills typically found in personnel alreadyfami l i a r w i th e l ec t ron ic ma in t enance andspecifically recruited for that purpose. New transitsystems, of course, have little choice but to recruitand train an entirely new maintenance force sincethere is no existing labor surplus of ATC techni-cians, either locally or nationally, to draw on.

It should be noted that ATC generally leads to anupgrading of the maintenance force. Since ATC isan addition to all the other types of transit equip-ment, it increases, not decreases, the number of jobsavailable, The pay levels for this kind of work tendto be higher than for other types of transit mainte-nance; and, to the extent that ATC technicians arerecruited from within an existing workforce, itoffers employees opportunities for advancement.

Passengers

The transit passenger typically has very little in-terest in the technical details of the system—ATCor otherwise. One transit system manager ex-pressed it thus:

People use a mass transit system to get from apoint of origin to a point of destination, and theywant to do it quickly, reliably, comfortably andeconomically. The train is nothing more than apeople box, The system designers’ job is to createa system which will enable that people box totraverse the transit corridor rapidly and reliably,day after day after day. The passenger doesn’tcare—has no interest in knowing—whether thetrain is controlled by a master centralized com-puter , or local ized control—whether i t ispowered by AC or DC motors or by little squir-rels running around cages-whether it operateson standard gauge rails or extra wide rails—whether those rails are supported on timber crossties or concrete cross ties. The passenger doescare about being able to board his train every dayat a preestablished time, riding in a clean andcomfortable environment, arriving at his destina-tion without being ruffled either physically oremotionally, completing the trip as quickly as isreasonably possible, and accomplishing it all at afare which he considers to be reasonable.(Johnston, 1974)

130

The impact of ATC on passenger acceptance ofthe system would thus appear to be minimal, unlessthe ATC system is the specific cause of servicedelays-and publicly identified as such, Some tran-sit system managers expressed the view that publicconfidence in a highly automated system might belower than for a conventional system, especiallyduring the start-up period or following some otherperiod of operational difficulty, However, it wasalso believed that, once the public becomes ac-customed to the system and if performance isreasonably reliable, apprehension about automationwould subside. It is very difficult to gauge publicopinion in this matter for there have been nostudies directed to the topic of automation in transitoperations. Furthermore, public comment on newsystems, such as BART, tends to be in response tospecific events and often does not grasp the essen-tial technical issues.

There is a widely held view in the transit indus-try that a completely automated train controlsystem without an on-board operator is not a viableproposition. Passenger safety in emergency condi-tions demands the presence of a transit systememployee to control the situation, to evacuate thetrain, and to lead passengers to safety. TheAIRTRANS system has experienced problems in

this regard. Passengers in unattended vehiclesbecome apprehensive when the train stops some-where other than at a station, even though there isno real or apparent emergency. There have beencases of passengers leaving the train and walking onthe tracks, causing a shutdown of the system untilthey can be reboarded or led to a station. It is alsobelieved that passengers derive a sense of securityfrom the presence of an on-board operator, both as asource of aid in emergencies and as a protectionagainst personal attack or crime. Unmanned vehi-cles are also considered to present operationalproblems. Without an operator to control car doorclosure, the passengers may adversely affect head-ways and capacity because of the variability indwell time introduced by passenger-actuated doors,Systems with unmanned vehicles (and, to some ex-tent, those with one-man trains) have also foundthat passengers have difficulty in obtaining infor-mation about train routes and schedules. To accom-modate passengers, it has been necessary to installmore extensive” signing and public announcementdevices and, in the case of AIRTRANS, to hire addi-tional station employees to provide passenger infor-mation and assistance. The human factors ofsystem design and operat ion in relat ion topassengers is a matter that acquires increased im-portance as the level of train control automation in-creases and the level of vehicle manning declines,

7’2 -fif13 O - 7(1 - 1 rl 131

Chapter 6

THE PLANNING AND

DEVELOPMENT PROCESS

133

INTRODUCTION

The main purpose of a rail rapid transit system isto transport passengers with speed, safety, and de-pendability. The train control system provides theprotection (ATP), operational control (ATO),supervision (ATS), and communications necessaryto accomplish this purpose.

The older rapid transit systems, such as CTA,MBTA, and NYCTA, were designed to performmany train control functions manually. Until re-cently, the major uses of automation have been fortrain protection functions (ATP) and certain super-visory functions, such as dispatching. The develop-ment of new technology within the last decade or sohas made it possible to automate other train controlfunctions, and so the older rapid transit systems arenow in the process of converting to higher levels ofautomation, especially in the areas of train opera-tion and supervision.

Rail rapid transit systems built in recent years(PATCO and BART) and those now under con-struction are tending to make use of more extensiveautomation and more sophisticated train controlthan the older existing systems. Various forms ofadvanced ATC technology seem to figure in theplans of system designers from the very outset.Thus, it appears that the general trend in both exist-ing and future rail rapid transit is toward increasedautomation. In light of this, the process by whichtrain control systems are conceived, planned, pro-cured, and tested assumes great significance; and itis important to investigate the way in which theATC design evolves within the context of overallrapid transit system development.

The evolutionary cycle of ATC, like the totaltransit system of which it is part, has three majorphases: planning, development, and testing. Thesephases are generally sequential but there arenumerous interactions and iterative steps. Forsimplicity of discussion, however, the features andissues of each phase will be treated separately. Atthe end of this chapter is an examination of the sub-ject of research activities that support the overallplanning, development, and testing process.

The evolution of an ATC system can be lengthy,often as long as the evolution of the transit systemitself. Table 31 identifies the significant dates for 16systems—the five existing and three developmentalsystems considered in detail in this report and eightother systems for general reference. The CTA,

MBTA, CTS (Cleveland), and NYCTA programs in-volve addition of new ATC equipment or extensionof an existing line. For the others, the programspans the conception and development of the entiresystem. The times listed include the evolution ofgeneral train control system concepts and thedetailed engineering development.

The major issues associated with planning anddevelopment are examined in the order in whichthey generally occur in the system evolution proc-ess.

Planning (Concept Formulation and PreliminaryDesign)

The concept of the ATC system is usually for-mulated early in the overall transit system planningprocess, The major issues are concerned with theorigin of the ATC concept, the influences whichshape it, the selection of a desired level of automa-tion, and the criteria and techniques used to evalu-ate the concept and translate it into a preliminaryengineering design.

Development (Final Design and Procurement)

The final engineering design and procurementprocess may cover several years, during which theoriginal concept may undergo substantial change,The most significant issues relate to how theengineering design specifications are written, howcontractors are selected, how the developmentprocess is supervised and managed, and howemerging differences between concept and imple-mental ion are dealt with in the developmentprocess.

Testing

Testing is a continual process that begins as soonas specific items of ATC equipment are engineeredand, manufactured and ends when the entire systemis ready for revenue service. The issues in this areahave to do with the types of tests conducted, thetiming of the tests in relation to the developmentcycle, and the methods by which the ATC system isevaluated for serviceability and conformance tospecifications.

Research and Deve lopment (R&D)

R&D is a supportive activity that runs concur-rently with planning, development, and testing. Theissues to be examined include the types of R&Dbeing conducted, its application to the design of

135

TABLE 31.—Significant Dates in the Engineering Planning, Procurement, and Testing of ATC for Various Transit Systems

CTA CTS(Lake (Airport

St Exten.

Line) sion)

N Y C T A

(System

M B T A Im -

(Red prove.Dade Co. D/FW MARTA Line) NFTA ments)

Twin

Cities

SEA. Area

P A A C PATCO RTD T A C MTC W M A T A

————

1959 1958 1969 1967 1967 1960

Balti -more BART

1962 Before1965

1964 1965 1962 Before 1967 (f-l

1966

Planning b 1962

1967

1951

1953ATC

PreliminaryD e s i g n

S t a r t e d

ATC

PreliminaryDesignCompleted

ATC Pro-curement

Specifica-

tions

Issued

ATC Pro-

curementContractAwarded

ATC

BebuggingStarted

ATC First

UsedInPassengerService

Initial Plan-ing

(Years)

ATC Evolu-

tion( Years)

Total TimeSpan

( Y e a r s )

1962 1969 1969 1966 1966 1973 Before 1963 1970 1967 1969 1962

1963

1973 1966 1965 (1975) 1970 1974 1970 1973 1972 1967

(d)(1977) 1966 1966 (1978) 1971 1974 1966 1969 1971

1967 1966 1971 (1976) (1975) 1966 1969 1971

1973 1971 (1978) 1968 1972

(1981) 1972 1967 1966 (1984) 1974 (1977) 1971 (1981) (1978) 1969 1973 (1981) 1976

5 4 4 1 6 (f) 4 5

5 3+ (15) 5 (11) 5 (8) (f) (14+) 6

5

(14)

(19)

2

19

21

1 1 2 2

5+ (12) 14

6 (14) 165 3+ (20) 9 (15) 5+ (14)(f)

(18) 11

a Unless otherwise noterf, the dates Ilstwi are for the start of the actlvltv Dates enclosed In parenthtws ( ) are planned (iates All act]vltws are fnr new svstems except as notecfb ATC plannlng IS generally Influenced by a numlwr of system IIeclslons w this date IS the start of the nverall syswm planning“ Prellm\nary d~slgn IS cons] cierwi to In[.lu[ie snme conceptual work as well as ciemonstratlnn protects where applicable{i

Act]v)t y IS cllrrently In progresse Most transit systems are [.onstructwi In phdses The program duration listed IS for a single phasr or the first phdse of the programs Becauw early planrrlng of mult~ple prngrams IS usIIally comprehenmve

the t~me requlrwi for such planning WI]] generally Iw longer than required for d ~maller single-phase effOrtf At NYCTA the process of equipment replacement and updat!ng 1$ vmtually continuous

new systems, the use of test tracks, andneeds in the area of ATC technology.

major R&D

ISSUE D–1: DESIGN CONCEPTS

How do ATC design concepts originate, andby what criteria is the level of automationselected?

For new systems, ATC design conceptsemerge from policy and planning decisions aboutthe general transit system concept. Initial selec-tion of the level of automation tends to be in-fluenced more by social, economic, and politicalconsiderations than by engineering concerns. Inalready operating systems, where ATC is in-stalled to upgrade or extend service, engineeringconcerns–especially evolutionary compatibilitywith existing equipment-are predominant. Forboth new and old systems, the experience ofothers (particularly their mistakes) has an impor-tant influence.

Some preliminary notion of the type of train con-trol system desired is usually included in the state-ment of the basic transit system concept preparedby the policymaking body responsible for planningthe system. For all of the transit agencies investi-gated in this study, the policy and planningauthority is a commission or board of directors cre-ated by legislative act. The size and compositionvary. Some are elected; others are appointed. Themembers are usually not engineers and seldomhave technological backgrounds in the area of tran-sit operation and train control, but there is alwayseither a technical staff or an engineering consultantfirm to assist the board in planning activities. Some,particularly transit systems already in operation,have staffs of considerable technical competence.For example, the CTA and NYCTA staffs do all theengineering planning for new developments andoversee procurement and testing. In general,however, the local policy and planning agency aug-ments the technical capability of its staff by hiringconsultants who conduct studies to support plan-ning decisions and flesh out the basic design con-cept. In some cases, the consultant firm may also beresponsible for the subsequent engineeringdevelopment of the system.

The activities of the planning agency are in-fluenced by many factors: State and Federal legisla-tion, regulatory agency rules and decisions, UMTA

policy, economics, public opinion, local social con-cerns, labor relations, and political interests, toname a few. Technical, considerations often playonly a small part and may be overridden by theseother concerns. Specific examples from among thesystems investigated will help to illustrate thenature and diversity of the ways in which ATCdesign first takes shape.

The PATCO Lindenwold Line was planned andconstructed over an n-year period. It is not clearwhen the basic ATC design concept was formu-lated; but an engineering consul tant reportpublished in 1963, about midway between the timeof the initial decision to build the system and thetime the line was opened for service, recommendedthe use of ATP and ATO. The tone of the reportmakes it plain that the nature of the train controlsystem was still an open question 5 years after theplanning process started. The primary justificationadvanced by the consultant for ATP was safety, andfor ATO efficiency of operation.

In contrast, an ATC design concept for BARTwas established very early in the planning processand took over 20 years to evolve. Original planningstudies conducted by engineering consultants toBART in 1953 to 1956 advanced the general conceptof completely automatic operation at high speedand short headways. An onboard “attendant” wasenvisaged, not as an operator but as an aide topassengers, much like an airline stewardess. Theidea of building a glamorous “space age” systememploying the most advanced technology seems tohave been a dominant concern in BART from thevery beginning, Th i s app roach was c l ea r lymanifested in the ATC concept. The justificationmost often given was that advanced train controltechnology was necessary for the, high-speed, short-headway operation needed to attract patrons,

CTA, in planning the conversion to cab signal-ing, appears to have been most strongly influencedby operational and engineering factors, Cab signalswere seen by CTA as an improved method of assur-ing train separation and preventing overspeed, i.e.,as a way of enhancing safety. Compatibility withexisting signal equipment and other elements of thesystem was also a factor (as it is in MBTA wherecab signal conversion is now being implementedand in NYCTA where it is in the planning stage).Engineering and equipment concerns are also adominant concern in the planned expansion ofPATCO, where the existing ATC system dictatesthat the new lines have the same operational

137

— . .-

characteristics andbe integrated with

level of automation in order tothe present line.

Operational transit systems for airports (such asSea-Tac and AIRTRANS) feature automatic, crew-less train operation. These systems were plannedand built in a rather short time span (6 years forSea-Tac, 9 for AIRTRANS). The concept of un-manned vehicles was inherent in the nature ofthese systems from the beginning. It was felt by theplanning agencies and their consultants that fullyautomatic operation offered significant savings inlabor costs and was the only way to make thesystem economically viable.

There are sometimes general engineering deci-sions made during the planning process that maylimit the technology that can be employed for ATCequipment. For example, a number of transit plan-ning agencies have decided to employ only equip-ment already proven in use by other operating tran-sit systems. For WMATA, the schedule set by thepolicy makers did not permit extensive R&D andengineering studies before selecting a train controlconcept. Therefore, WMATA engineers decided tospecify an ATC system that could be realized withproven, existing hardware.

The formulation of the ATC system concept isalso strongly influenced by events in other transitsystems. The community of rail rapid transit agen-cies, consultants, and suppliers is a small fraternity.There is a continual exchange of informationamong the members and a high degree of mutualawareness of plans, problems, and operation ex-perience. Because the supply of qualified transitconsultants and engineers is limited, there alsotends to be a steady interchange of personnelamong transit properties, consultant firms, andequipment manufacturers. These forms of interac-tion assure that the experience of others will bereviewed during concept selection and preliminarydesign,

However, the review of others’ experience isoften rather narrowly focused. There is a tendencyto be swayed more by specific problems and inci-dents than by overall statistics and the general pat-tern of operations. “Avoiding others’ mistakes”seems to be a more dominant concern than emulat-ing their success. For instance, the problems en-countered by BART were in part responsible for themore conservative approach adopted by WMATAand Baltimore MTA. Atlanta’s planners also havechosen a train control system less sophisticated

than that originally proposed by their consultants(PBTB, who were responsible for BART), partly asa reaction to the experience in San Francisco. Cau-tion is a prudent course, but the rapid transit indus-try could also benefit if there were a more com-prehensive body of comparative performance datato help make decisions on an analytical, rather thana reactive, basis.

The salient points that emerge from an examina-tion of the initial planning process are that ATCdesign concepts originate (sometimes early, some-times late) in policy-level decisions about thegeneral nature of the system. The methodologyemployed to arrive at concept definition is often in-formal and influenced strongly by engineering con-sultant firms engaged to assist in planning thesystem. Except in the case of modernizing an exist-ing system, technical considerations of train controlsystem design seldom predominate. Route struc-ture, service characteristics, vehicle design, right-of-way acquisition, cost, and local sociopoliticalconcerns tend to be given greater importance at theearly stage of planning. The engineering aspects oftrain control are most often deferred to a latter stageof planning, when design specifications are to bewritten, As a result, the embryonic ATC design isusually not defined in detail until other parts of thesystem have taken shape, The preliminary ATCconcept thus tends to develop a life and perma-nence without being subjected to engineeringscrutiny and cost-benefit analysis to determine itsappropriateness for, and compatibility with, the restof the system,

There seems to be a crucial difference betweenexisting and new systems. The former give greaterweight to engineering concerns and specific opera-tional needs in defining an ATC concept. Newsystems tend to take a broader, more informal, andless technical approach, The engineering-orientedapproach offers the advantage of assuring a worka-ble ATC system tailored, although perhaps notoptimally, to specific local needs. But there is a dis-advantage. The scope of the ATC concept inupgrading an existing system tends to be limitedand constrained by what already exists. The bolder,“clean sheet of paper” approach employed by manynew systems results in a more technologically ad-vanced concept and greater coherence betweenATC and the system as a whole, but the practicalproblems of development and engineering may notalways be given sufficient attention,

138

I S S U E D – 2 : S Y S T E M DEVELOPMENT

How is the ATC system concept translatedinto preliminary and final functional design ?

Most system development work is done byengineering consul tants , except for largeestablished rapid transit systems where it is doneby the in-house staff. The methodology varies,but there is a trend toward a more systematic andsophisticated approach using simulation, systemanalysis, system assurance studies, and testtracks.

The first step in the development process forATC systems is preparation of a preliminary func-tional design, expressing the basic concept and itsunderlying policy decisions in engineering terms.The preliminary design defines performance re-quirements and organizes the ATC system withrespect to functional relationships among systemcomponents. At this stage, the ATC system is sepa-rated into its major subsystems (ATP, ATO, ATS,and communications), and the functions required ofeach are specified. Further analysis may separatethe system into carborne and wayside elements.The preliminary design also defines the interfacesbetween ATC and other parts of the transit system.

For most of the transit agencies investigated, thetechnical staff plays some role as engineering plan-ner in preliminary design. However, the extent ofstaff involvement varies widely. In establishedoperating agencies, such as CTA, CTS, andNYCTA, the engineering staff does almost all of thepreliminary design work. In new systems, wherethe technical staff may be quite small, especially inthe early planning phases, engineering consultantsare generally and extensively used. Heavy par-ticipation by consultants is also characteristic inestablished systems undergoing a major program ofnew construction or modernization. While the pro-portion of staff to consultant participation varies,there appears to be wide agreement among transitsystem managers that staff involvement should notfall below a certain minimum level, roughly 15 to 20percent of the design work. In this way, theauthority can maintain technical involvement inthe preliminary design process and exercise propercontrol over system evolution.

Several kinds of methodology may be employedin preliminary design. The specific methods differwidely from authority to authority, and it is difficult

to discern any common thread, beyond the generalbelief that technical studies are needed to gatherand analyze information about the performance ex-pected of the system. In the new systems now underdevelopment, there seems to be an increasingreliance on the so-called “systems approach”79 andthe use of techniques such as simulation, ridershipanalysis, function/task analysis, and cost/benefitstudies. Several agencies (BART, CTA, NYCTA,Sea-Tac, and PAAC) have also conducted studies attest tracks on their properties to gather informationneeded for preliminary design.

The application of system analysis techniquesdoes not appear, however, to extend very deeplyinto the design of the ATC system itself. There is atendency, for instance in cost/benefit studies, totreat ATC as a whole, without examining thechoices that may exist within the train controlsystem as to degree of automation or alternativemethods of achieving a given level of automation.One reason is the general lack of empirical data onthe performance of ATC systems, which precludesa precise formulation of potential benefits, A sec-ond reason is the overriding nature of the safetyfactor which strongly influences designers to auto-mate the train protection function, without regardfor the cost/benefit relationship of ATP to otherfunctional elements of ATO or ATS. Also, since theentire ATC package typically amounts to only 5percent or less of the total capital cost of the transitsystem, there is a belief that cost/benefit analysisshould be concentrated in areas where the payoffwill be greater.

Thus, the process of developing a preliminaryfunctional design of the ATC system still tends tobe more art than science, but there is a trend towarduse of more objective, quantitative, and systematictechniques. This is particularly evident at the points

7~he “sy.sterns approach, ” which derives mainly from aero-space technology, is a collective designation for techniques usedto solve complex problems in a methodical, objective, and oftenquantitative way. The systems approach involves a logical andreiterative analysis of the system into its constituent parts, eachrepetition leading to a greater degree of specificity, Othercharacteristics of the system approach include measurability ofparameters, constant recognition of subsystem interdependence,and parallel analysis of elements. The heart of the systems ap-proach is the *’System Engineering Cycle” which involves foursteps: (1) convert system requirements to functional require-ments, (2) convert functional requirements to specific detail re-quirements, (3) conduct analysis to optimize parameters, and (4)convert specific detail requirements into end products, (Grose,1970)

139

.-

of interface between ATC and other subsystems,where mutual influence and interdependence canbe reduced to quantitative expression and theparameters of performance can be manipulated.Even here, however, ATC system characteristicstend to be treated as dependent variables, i.e., thedriving concerns are other system characteristics, towhich the ATC system design must be accommo-dated.

System design is a continual and reiterativeprocess, preliminary functional design merging intofinal engineering design without any clear line ofdemarcation, The process culminates in the state-ment of specific equipment and performance re-quirements, suitable for incorporation in procure-ment specifications. Often, final design coincideswith the preparation of procurement specifications,and it is difficult to separate the two activities.However, for the purpose of this discussion, finaldesign is considered to include all activities neededto define the detailed technical requirements of theATC system, up to but not including the actualwriting of procurement specifications.

As in preliminary design, the final design is ex-ecuted either by the technical staff of the transitagency or by engineering consultants. Here, too, theolder and established agencies tend to rely more ontheir own personnel, and new agencies more onconsultants. Usually, a single consultant is hired forfinal design of the complete ATC system –carborne,wayside, and central control elements. Thisconsultant is often, but not always, the same firmthat carried out the preliminary functional design ofthe ATC system. Once reason for selecting a singleconsultant for the entire process is to assure con-tinuity and coherence of the ATC design as itdevelops, It is also considered advisable to have asingle consultant for all parts of the ATC system toensure integration of the design of vehicle andwayside equipment and their all-important inter-face.

Many of the factors that shape the preliminarydesign of the ATC system continue to have signifi-cant influence during the final design process. Non-technical factors still play a strong, but perhapsdiminishing, role as the system moves from plan-ning to engineering. The continuing influence ofnontechnical factors is not surprising since they areusually built into the design criteria and guidelinesthat emerge from preliminary design and are ap-plied to the final design. Still, as the system ap-

proaches the hardware stage, it is to be expectedthat purely engineering considerations should cometo the fore. Generally speaking, however, theprocess of generating detailed engineering require-ments from preliminary design criteria is basicallyan interpretive effort, with the experience and judg-ment of the designer playing the dominant part.However, there are two more formal designmethods that are being used increasingly in newt r a n s i t s y s t e m s . T h e y a r e s y s t e m s a f e t ymethodology and quantitative reliability, main-tainability, and availability analysis.

Most of the systems now being planned are in-cluding a formal system safety study, involvingdefinition of safety criteria, analysis of potentialsafety problems, and identification of ways to elimi-nate or minimize hazards, Some designers considerthis approach to safety superior to the traditionalmethods of “fail-safe” design. Others disagreesharply.80 It appears, however, that much of thecontroversy over the “fail-safe” and “systemsafety” methods is semantic; and it is premature todetermine whether the results of the two ap-proaches will differ, The important point is thatdesigners are turning, at least in the area of safety,to more systematic and quantitative methods ofanalysis,

Until recently, it has not been the practice in thetransit industry to specify safety requirements inquantitative form, i.e., as a numerical statement ofrisk or probability of occurrence. Many believe thatthe levels of safety which must be achieved are sohigh that it is difficult, if not impossible, to statemeaningful quantitative standards and to devise anacceptable and practical method of verifying thatthey have been met. This view is not universallyheld, and the topic is highly controversial .However, it does appear that future ATC specifica-tions will place strong emphasis on formal pro-cedures by which potential safety hazards can beidentified, evaluated, and reduced to “acceptable”levels. An effort is being made to put hazardanalysis on a quantitative basis, but much of thework is likely to remain qualitative and judgmental.(Again, this view is not shared by all in the transitindustry, ) Along with the emphasis on quantitativemethods, there is also a trend to define safety in asense that is broader than just train protection andto deal with the safety aspects of the total system.

1111%111 chapt(~r !I. p~igr tlh for (i (1 is(:llssion of this topi(; .

140

The second formal design method that is cominginto wider use in the transit industry is quantitativereliability, maintainability, and availability (RMA)analysis. A discussion of this design technique ispostponed to Issue D–4, where it is considered aspart of the general question of how these aspects ofsystem performance are written into procurementspecifications.

ISSUE D—3: PROCUREMENTSPECIFICATIONS

How are ATC design requirements specified,and is there a “best” way to write such specifica-tions ?

There are two basic approaches to writingthe design (equipment-specific)specifications—

approach and the functional (performance) ap-proach. Each has advantages and disadvantages.The only generalization to be made about the“best” way is that, whichever approach is used, itis of crucial importance to specify equipmentperformance standards and to define explicitlythe means of testing.

The final design of the ATC system is docu-mented in procurement specifications in terms ofrequired performance for ATC functions and/orequipment components. However, the procurementspecifications have a much broader scope than justa listing of required ATC system performance. Re-quirements for documentation, scheduling, installa-tion, management visibility and control, andvarious types of testing may be specified togetherwith numerous contractual and legal provisions.The procurement specifications include all of thedetailed information required for a prospective sup-plier to prepare a bid.

As a general rule, the organization that does thefinal design of the ATC system also prepares thetechnical portions of the procurement specificationfor that system. At times, another consultant writesthe procurement specifications in cooperation withthe final designers. In this way some additional ex-pert knowledge is incorporated into the specifica-tions.

The most common method of preparing procure-ment specifications is by drawing on availablespecifications for similar equipment, from prelimi-nary proposals submitted by equipment suppliers,or from experience gained through testing or use of

similar equipment. Often, a general incorporationof test and use experience is achieved by requiringthe use of “proven technology,” which means thatthe same or similar equipment must have been usedor tested successfully on an operating transitproperty in the United States.

There are two basic approaches to writing pro-curement specifications. Requirements can bestated in functional terms (performance specifica-tions) or in equipment-specific terms (designspecifications). The two are not mutually exclusive,and in practice something of each approach is used.Thus, implicit in even the most design-orientedspecification is the expectation that the equipmentshould perform in a certain way,

The design type of specification indicates, to agreater or lesser degree, the equipment or systemcomponents needed to perform individual func-tions. In the extreme case, design specifications callfor particular items, for which only a narrow rangeof substitutes, or none at all, are acceptable. Suchspecifications are often issued by transit agenciesthat have similar, satisfactory systems in operationand wish to assure compatibility of the new equip-ment with that already in place. Recent procure-ments of cab signaling equipment by CTA typifythis approach. Somewhat less restrictive is thedesign specification that calls for a type of equip-ment with stated characteristics but leaves the sup-plier some room for choice. The WMATA traincontrol system specification is an example of themodified design-oriented approach, which hassome of the features of a functional specification.

Functional (or performance) specificationsdefine what functions are to be accomplished butnot the way in which they are to be accomplished.For ATC systems, the BART specification comesclosest to the purely functional approach, TheDiablo test track was operated for the purpose ofdetermining the feasibility of new ATC concepts(not to select a system). At the end of the testingperiod a functional specification was written to ac-commodate any of the concepts successfullydemonstrated (and many others). For example, thebasic train separation system could have used radar,track circuits, or any other device that met thestated functional requirements,

Table 32 below is a rough classification of thetype of specification used by seven transit systemsin recent procurements. The development of the sixnewest systems (Baltimore, Dade County, MARTA,

141

————-——

NFTA, RTD, and Twin Cities) has not yet advancedto the point where ATC system specifications havebeen written.

TABLE 32.—Type of Specification Used in Recent ATC Pro-curements

TYPE OF SPECIFICATIONSYSTEM Design Functional Combined

AIRTRANS x

BART x

CTA x

NYCTA x

PATCO x

SEA-TAC x

WMATA x

The use of a design specification permits thebuyer to exercise a high degree of control over theequipment purchased. At the same time, however,it requires considerable experience and technicalcompetence on the part of the buyer to be sure thatwhat he specifies will perform as intended. There isalways the risk that individually procured sub-systems will not prove compatible, with the buyerhaving no recourse but to go through a process ofredesign or retrofit. If a testing procedure has beenestablished in the specification, product evaluationand acceptance is usually easier for the buyer whohas followed the design approach. To the extentthat design specifications are equipment-specific,they lock the buyer into a given technology and donot allow taking advantage of innovation, economy,or other improvements that the seller might other-wise be able to effect.

One of the major advantages of a functionalspecification is its independence from particularmeans of implementation. It gives the supplier greatlatitude when innovation is desired or when a widerange of hardware is acceptable. This approach ismost compatible with a new system being builtfrom the ground up or with an independent part ofan existing system. In effect, the functionalspecifications transfer some of the responsibility forsystem design from the procuring agency to theequipment supplier.

Functional specifications, because they are lessdetailed, may be somewhat easier to prepare thandesign specifications. On the other hand, it is some-

what harder to define the desired end product withprecision. The functional specification allows thesupplier to be creative, but it can also provide theopportunity for cutting corners. Litigation, as in thecase of the BART train control system contract, isalways a possibility if differing interpretations aretaken or if the method of testing system perform-ance is not well defineed. From the buyer’s stand-point, one difficulty with functional specificationsis that it may not be possible to determine if the pro-duct will meet performance requirements until thecomplete system is assembled,

The re i s no un ive r sa l ag r eemen t on t hesuperiority of either type of specification, Eithercan be employed successfully so long as the buyerrecognizes the shortcomings of the selected ap-proach and so long as the standards for an accepta-ble . product are clearly and fully defined. Theresults of the WMATA specifications, which com-bine a functional and a design approach, will beawaited with great interest to see if they offer acompromise solution to the problem of specifyingequipment requirements and characteristics.

It is of crucial importance that both the criteriaand methods of testing the equipment be made ex-plicit in the procurement specification, From a prac-tical standpoint, the design type of specificationmay offer some advantages over the functionalspecification in terms of the ability to define andmeasure reliability and maintainability-a problemthat lies at the heart of the difficulties encounteredby most new systems. Because of its importance,the topic of how RMA requirements are specified istreated as a separate issue immediately following.

ISSUE D--4: SPECIFICATION OFRELIABILITY, MAINTAINABILITY, AND

AVAILABILITYAre the methods of specifying reliability,

maintainability, and availability (RMA) adequ-ate to assure that ATC systems will give goodservice ?

This has been one of the most troublesomeareas of ATC system design and development.Transit agencies are becoming increasingly con-cerned with RMA problems, and an effort isbeing made to write specifications in more pre-cise and quantitative terms. In their present state,however, RMA specifications still fall short ofwhat the transit industry (both buyers andmanufacturers) consider satisfactory.

142

RMA specifications can be divided into twoclasses—those that state quantitative requirementsand those that do not. Before issuance of the BARTspecifications, most transit agencies followed anonquantitative approach to RMA specifications,and some still do. The BART specifications were apioneering effort to introduce in the transit industrythe quantitative methods used in the aerospace in-dustry for specifying RMA. This was a major in-novation at the time and, like nearly everything elseassociated with BART, controversial. However, allthe agencies planning new systems are now incor-porating some form of quantitative RMA require-ments in their specifications.

Historically, reliability and maintainability havebeen treated only in general terms in procurementspecifications by transit agencies. Some form ofwarranty was called for, but specific requirementsas to reliability (mean time between failure, orMTBF) or ease of repair (mean time to restore, orMTTR) were not stated. Certain transit agenciescontinue to follow this practice for a number ofreasons. In some cases, the procurement consists ofadditional equipment similar or identical to pastpurchases. Thus, the expected performance of theequipment is understood by buyer and seller to belike that already in use. Another reason has to dowith the size and nature of the transit industry.There are only a few buyers and even fewer sellers,all of whom have been in business for many years.Hence, the needs of the former and the capabilityand reputation of the latter are well known. In suchcircumstances, it is considered unnecessary to drawup elaborate and detailed statements of RMA re-quirements. The seller is familiar with the kind ofequipment now in use by a transit system, and thetransit agency knows that the seller must standbehind the product in order to remain in considera-tion as a source of supply. A third reason for takingthe nonquantitative approach, especially in smalltransit systems, is that the managing authority maynot feel it is cost-effective (or they may not be ableto get the funds) to prepare specifications that in-volve extensive engineering analysis, and perhapstesting.

The quantitative method of specifying RMA hasfound increasing favor in the transit industry fortwo basic reasons. First, the type of equipment nowbeing purchased, especially for ATC systems, ismuch more complex and technologically sophisti-cated, creating a need for the document thatgoverns the purchase of the equipment to become

increasingly detailed and precise. Second, the num-ber of suppliers has increased and now includesfirms without a long and established record in thearea of train control equipment manufacture andinstallation. Starting with BART and continuingwith WMATA, MBTA, and a number of newsystems being planned, transit agencies are turningto a quantitative approach.81 Still, a decade after theBART initiative, the specification of RMA require-ments remains a developing art,

There are significant differences in how quan-titative RMA requirements are written, dependingupon whether the procurement document is adesign or a functional specification. In a functionalspecification, the buyer defines generic types offailures, their consequences, and required systemperformance. The seller is (in theory) free to con-figure the system in any way seen fit so long as thefunctional requirements are met and the systemperforms as expected. In a design specification, thebuyer develops a specific equipment configuration,evaluates the consequences of failure of each com-ponent (equipment items not functions), anddefines the component performance requirements.The seller must then meet the performance require-ments on an item-by-item basis. Thus, the sellermay well have no responsibility for the perform-ance of the total system, but only for the parts as setforth in the procurement specification. In effect, thefunctional specification transfers much of theresponsibility for detailed system design to theequipment supplier, whe rea s w i th a de s ignspecification this responsibility is retained by thepurchaser.

With regard to RMA, the difference betweendesign and functional specifications centers aroundthe definition of failure. In design specifications thedefinition is reasonably clear-cut and precise.Failure means that a given component does not re-spond to a given input or fails to make a particularoutput within stated tolerances. In a functionalspecification, failure is defined not in terms ofspec i f i c equ ipmen t pe r fo rmance , bu t moregenerally as the inability of the system (or sub-system) to perform certain functions. Some func-tional specifications (such as those prepared forBART and Sea-Tac) also identify the consequencesof failure that are of concern.

MMARTA, Dade County, Denver RTD, NFTA, PAAC, andTwin Cities are all contemplating the use of quantitative RMAspecifications.

143

. .—

A problem of interpretation can thus arise inevaluating equipment procured under a functionalspecification. Some failures and their consequencesare defined; but others are not, even though thesame piece of equipment may be involved. Whatthen is a failure? And what particular equipmentmalfunctions are to be counted in determining thereliability of the purchased equipment? There is adisagreement, and litigation in progress, betweenB A R T a n d t h e A T C e q u i p m e n t s u p p l i e r(Westinghouse Electric Corporation) as to the in-tent of the specification on these very points.

The WMATA train control system procurementspecification, written with the BART experience inmind, attempts to deal more clearly with the defini-tion of failure. In the WMATA specification, failureis defined as “any malfunction or fault within anequipment which prevents that equipment fromperforming its function in accordance with thespecification. ” Thus, it appears that WMATA RMArequirements pertain to all equipment failureswithout regard to the effect on train operation.However, the specification does not clearly indicatewhat modes of operation are to be counted and howequipment operating time is to be reckoned incalculating MTBF, In some systems, ATC units arelocated at each end of the train and actually controlonly half the time. If a failure occurs in a unit notinvolved in train operation at the time of malfunc-tion, is this to be counted as failure? And if so, howmany hours has it been operating? All the time thatthe car has been in revenue service, or only that partof the time that the ATC unit has been used to con-trol the train?

Without belaboring the example, it is clear thatthe transit industry still has not reached a full anduniversally accepted understanding of how tospecify and test equipment reliability. A recentstatement by a representative of an equipmentmanufacturer (King, 1975) highlights the continuingproblem.

Success and failure of transit equipment andsystems must be defined in relation to their mis-sion. Indeed, the term “mission” itself probablyrequires redefinition. Many industry specifica-tions in recent years have not agreed on suchpoints as whether a transit vehicle completes itsmission at the end of one trip or the end of a fullday, or when that day ends, or whether the vehi-cle must be available during all peak serviceperiods. If the function of transit equipment is

carrying passengers, has a mission failed if anequipment outage occurs during nonrevenueservice? These are some of the fundamentalquestions which must be answered to define tra-ditiona1 reliabi1ity in a manner acceptable totransit industry application.

One of the significant problems affecting theability of the transit industry to draw up meaningfulRMA specifications is the lack of a data basedescribing the performance now being achieved inthe industry. Individual manufacturers have someinformation, as do individual transit systems, butthere is no uniform method of reporting and noavailable industry-wide data base.

This need has been recognized by transit agen-cies and equipment manufacturers; and, throughtheir industry organization (the American PublicTransit Association), an effort is underway to dealwith the problem. APTA task group, known asRAM (for Reliability, Availability, and Main-tainability), has been assigned the responsibility ofdeveloping recommendations for a standardizeddata collection and reporting procedure. Theproblem of making these data generally available,free from local transit system bias and manufac-turers’ proprietary concern, is still unsolved,

ISSUE D–5: ‘ EQUIPMENT SUPPLIERS

What firms supply ATC equipment? Is theretransfer of ATC technology between automatedsmall vehicles and rail rapid transit systems?

Historical ly, two U.S. f i rms—GRS andUS&S--have supplied most of the ATC equip-ment to the rapid transit industry. In recentyears, several new firms, supplying either specialproduct lines or control equipment for smallvehicle sytems, have entered the market. Themajor transfer of ATC technology is from railrapid transit to small vehicle systems, but not thereverse.

The suppliers of ATC equipment to the rail rapidtransit industry fall into two distinct groups: thosethat provide a broad line of services and equipmentand those that have limited lines or specialtyproducts. There are many firms in the latter catego-ry, but the former includes four companies, GeneralRailway Signal Company (GRS) and Union Switchand Signal Division of the Westinghouse Air Brake

144

Company (US&S)82 are old, established firms thathave a long history in the signals and communica-tion business and have dominated the market. Re-cently, two new suppliers have entered the com-peti t ion. Westinghouse Electr ic Corporat ion(WELCO) supplied the ATC system for BART,where they were low bidder against GRS and US&S.Transcontrol is furnishing the ATC system for theSan Francisco MUNI light rail system and for theToronto Transit Commission in Canada,

There are many more suppliers of ATC equip-ment for small, automated-vehicle, fixed-guidewaysystems. In addition to GRS, US&S, WELCO, andTransControl, the list includes Philco-Ford, TTI(now TTD) and Varo Monocab.

The number of firms supplying small-vehicleATC systems, and the organizational relationshipsamong them, change from year to year. Some dropout of the market, new ones enter, and others formjoint ventures or acquire each other. It is a marketwhere there are many more companies offeringsystems than have actually received contracts forinstallations, Further, the resulting contracts areusually rather small. The complete “Satellite Tran-sit System” installation (guideway, vehicles, andcontrols) at the Seattle-Tacoma airport was about$7 million, while the AIRTRANS system at theDallas-Fort Worth airport was about $3 I million.The ATC portions of these systems were about 7 to12 percent of the total contract prices.83

To date, transfer of technology between conven-tional rapid transit systems and the new small vehi-cle systems has been in one direction—from theconventional to the new systems. Reverse transfer,and entry of small vehicle system developers intothe conventional rail rapid transit market, has notoccurred, perhaps due to the much larger size of thecontracts and capital commitments required tocompete in the conventional rail rapid transitmarket, or perhaps due to the failure of AGT sup-pliers to develop workable systems for rail rapidtransit application.

While some foreign-made ATC equipment isutilized in the United States, the market is not really

8ZUnlon Switch and Signal is also referred to by the acronymof its parent firm, WABCO.

@aIn relative terms, this proportion is somewhat greater thanthe 3 to 5 percent of total contract price that is typical for railrapid transit systems. The absolute dollar amounts, however, arequite small.

receptive to foreign incursions. There are severalreasons. Some procurement specifications excludeforeign suppliers by requiring prior transit servicein the United States or by including restrictions onforeign-made components. Also, U.S. transit agen-cies tend to doubt that foreign suppliers would beable to provide continuous long-term service.Finally, there are some major differences betweenU.S. and foreign ATC technologytechniques.

ISSUE D–6: CONTRACTOR

and engineering

SELECTION

How ore contractors for ATC design andengineering selected?

The lowest technically qualified bidder isusually selected. Competitive bidding and awardto the low bidder is required by law in manyStates.

Usually, two or more suppliers will compete forthe opportunity to design, build, and install ATCsystem hardware and software in response to thetechnical specifications describing required systemcharacteristics. Ultimately, responsibility for selec-tion of the supplier rests with the directors of thetransit authority. Most frequently, the directors relyon their technical staff for evaluation of the pro-posals and for monitoring the work of the selectedcontractor. This procedure was followed at CTA,CTS, MBTA, NYCTA, and PATCO. However, atBART, the general engineering consultant (Parsons,Br incke rho f f -Tudor -Bech t e l ) was de l ega t edauthority for some of the contractor selection andmanagement. Interviews with personnel at newsystems in the planning or early constructionphases (MARTA, RTD, WMATA, Balt imore,NFTA, and the Twin Cities) indicate that theseagencies will also utilize consultants to assist incontractor selection and management,

The increasing involvement of consultants incontractor selection and management for new railrapid and small vehicle systems reflects the increas-ing complexity of new rail rapid and small vehiclesystems. The design and development of suchsystems is often beyond the capability of the limitedstaff maintained by most transit agencies. It shouldbe noted, however, that consultants may havesomewhat different motivation and may use some-what different evaluation criteria than the transitauthority,

145

Contractor selec tion is relatively simple when The opportun“off-the-shelf” equipment is to be used and thecompetition reduces to a matter of price amongprospective suppliers, all of proven capability.Often, however, the available equipment does notsatisfy all of the specifications and requirements.Contractor selection then involves identifyingqualified suppliers, publishing an invitation for bid,evaluating the bids received from the prospectivesuppliers, and awarding a contract. Table 33 sum-marizes the contractor selection approaches used bytransit authorities in several recent procurements.

Several of the transit authorities require that aprospective ATC system supplier be a manufac-turer of equipment proven in use on operating tran-sit systems in the United States. If the ATC systemat BART is considered to be proven, this restricts thelist of qualified ATC equipment suppliers to justthree companies: GRS, US&S, and WELCO.However, technical personnel at some authoritiesdo not accept the BART ATC system as proven.Thus, only GRS and US&S are presently consideredqualified by these authorities. The list could beenlarged by including Transcontrol if Canadian in-stallations were accepted.

ity for a new company to becomequalified as a supplier of ATC equipment is offeredby several authorities, who will permit the com-pany to install and demonstrate ATC equipment ata test track location on the authority’s property. Iftesting proves that the equipment has desirable per-formance features together with acceptable safety,quali ty, rel iabi l i ty, and maintainabil i ty, theauthority’s technical staff may approve this com-pany’s qualifications to bid for the next ATC equip-ment procurement. The prospective supplier mustbear the expense of the demonstration equipment,installation, maintenance, and testing in this pre-qualification program.

Prior to 1969, the Dallas/Fort Worth AirportBoard conducted an investigation of possible sup-pliers of an automated system. As a result of this in-vestigation a Varo/LTV/GRS team and Dashaveyorwere selected as the two (and only) qualified candi-dates. These two submitted preliminary engineer-ing reports in October 1969. In 1970, Varo/LTV/GRSand Dashaveyor received technical study grants fordemonstration of their systems at the plant. Initialbidding for AIRTRANS took place in March 1971,with Varo/LTV/GRS and Dashaveyor being the

TABLE 33.—Contractor Selection Approaches

Transit Bidder Evaluation ContractSystem Qualification Process Award

BART (d) (b, c) WELCO

CTA (b) GRS, US&S

CTS (d) (b) GRS

D/FW (e) (b, c) GRS

MARTA (d) (b, c)

MBTA (d) (b) GRS, US&S

NYCTS (a, d) (b) GRS, US&S[f)

PATCO (d) (b, c) us&s

SEA-TAC (b, c) WELCO

WMATA (g) (b) GRS

Demonstration at test track.Low bid.Proposed performance.Manufacturer of proven equipment.Demonstration at plant was an original requirement.

At a second bidding there was no such prequalification.R-44 and R-46 procurements.Preliminary proposals.

146

only bidders allowed. One bid was rejected as toohigh, and the other was rejected as not responsive tothe specifications. In May 1971, a second biddingtook place with four bidders: Bendix/Dashaveyor,VSD-LTV, WABCO Monorai l Division, andWELCO. VSD-LTV was selected as the supplier.The subcontract for the train control system wasawarded to GRS by VSD-LTV.

The “invitation to bid” requests a cost quotationfor supplying the ATC system and services definedin the procurement specifications. The solicitationmay also require submission of a technical proposalthat describes how the bidder intends to satisfy therequirements of the procurement specifications. Inaddition to the technical requirements, provisionsfor documentation, program planning, managementvisibility and control, quality control, acceptanceand systems assurance testing, and the many otherfactors specified as important to the procurementmust be taken into account by the prospective sup-plier in preparing his bid. Experience shows that itis very difficult to add or increase a requirementonce an “invitation for bid” has been published andthe prospective suppliers’ responses have beenreceived.

As a general rule , competitive bidding isemployed by the transit authorities; and, in mostcases, competitive bidding is required by State lawor local ordinance. Usua l l y , howeve r , t heauthorities reserve the right to reject all bids andhave a new solicitation. This study has disclosed noinstances where a sole-source solicitation had beenemployed.

The established transit agencies select an ATCequipment contractor from previously qualifiedsuppliers on the basis of the lowest price. Otheragencies employ a single-step process where tech-nical capability and cost are weighed together.WMATA was unique in that they used a two-stepprocess in which the responsiveness of prospectivecontractors’ proposals to the procurement specifica-tions in a prebid solicitation was used to make aselection of qualified bidders. Subsequent selectionof the winning contractor from the two qualifiedbidders was based solely on cost.

To date, cost estimates and award to the low bid-der have been based solely upon the capital costs ofsystem development and construction. Life-cyclecosting, which would require cost competitionbased upon both the capital and operating costs, isan alternative costing method that has not been

used but may find increasing favor as energy andeconomic conditions cause a shift in values.

Once a contract has been awarded, data onprogram status and control over program directionavailable to the transit authority management arelimited to that specified by the contract. Therefore,it is important that the contract provide the meansfor monitoring the contractor’s progress and for ex-erting some directive control over contractor ac-tivities.

Management control is achieved in many waysranging from a resident engineer at the contractor’splant to formal design status reviews, RMA predic-tions, progress reports, and other such techniques.Traditionally, management control of an ATCsystem contract has been achieved by assigning sig-nal engineers from the authority’s staff the task ofmonitoring the work of the ATC contractor. Theseengineers are expected to know the status of thecontractor’s program at all times throughout thecontract, and, in particular, to be aware of anyproblems and the work being done to solve them.They also direct contractor progress by exercisingapproval of designs proposed by the contractor,

Maintaining management control has become in-creasingly difficult as ATC systems have grownmore complex. BART, PAAC, and WMATA ATCsystem procurement specifications included provi-sions for system assurance programs, periodicdesign reviews, and other modern managementtechniques. Several transit authorities expect to hireseparate consultants to plan, specify, and monitorthe system assurance programs for their ATC pro-

14772-683 0- 76 - 11

— .

curements. These consultants will report directly tothe transit authority technical staff.

One important method for achieving manage-ment control is independent review of the ATCmanufacturer’s design. This review may be con-ducted by the transit authority engineering staff orby engineering consultants. The manufacturer is re-quired to correct all the deficiencies identified.Besides providing an independent evaluation of themanufacturer’s design, this procedure also educatesthe reviewer on the details of the design. This par-ticularly is important in new systems where thestaff may not have lengthy transit experience. Avariation of this approach is being used at MARTA.Periodic reviews of the MARTA train controlsystem design are being held under the auspices ofUMTA, with the DOT Transportation SystemsCenter serving as a technical consultant.

Established transit properties such as CTA andNYCTA have traditionally required the manufac-turer to continue to correct equipment deficienciesuntil the equipment performance is acceptable tothe chief engineer. Management control by theseauthorities succeeds, in part, because of the limitedmarket for ATC equipment. If an ATC equipmentmanufacturer wishes to remain in business, he mustnecessarily satisfy his customers, and these two arethe largest in the country. The major change inmethods of management control for the new ATCsystem procurements is the introduction of require-ments for detailed program planning by the contrac-tor. The increased management involvement per-mits control action to be taken immediately when adeviation from the program plan is noted. Thismakes it possible for management to avoid potentialproblems rather than waiting until they occur andrequire drastic action to correct.

Upon completion of the manufacturing process,the ATC equipment is delivered to the transitauthority, installed, and tested. Test procedures aredescribed in the next issue.

ISSUE D-8: TESTING

How are ATC systems tested? What kinds oftests are conducted, for what purposes, andwhen in the development cycle?

There are three categories of ATC systemtesting,system

148

each beginning ata different stage in thelife cycle and satisfying different needs.

Engineering testing occurs early in the develop-ment cycle and provides data for detailed systemdesign and modification Assurance testing is per-formed to evaluate how well the equipmentmeets procurement specifications. Acceptancetestiing is performed when the whole ATCsystem has been installed and debugged and maybe performed on significant subsystems beforetheir ‘integration into the total system. Accept-ance testing is the final demonstration that thesystem meets specification. There is room for im-provement in several areas-test planning, docu-mentation, and dissemination of results.

Testing serves a number of important functionsin the development process. It provides the datanecessary to support ATC design. It serves to iden-tify actual or potential problems during manufac-ture and installation. It is the means to verify thatthe resulting system meets specified requirements.

There are three basic types of testing: (1)engineering testing, (z) assurance testing, and (3)acceptance testing. Each is initiated at differenttimes in the system life cycle, and each satisfiesdifferent needs, but they are not mutually ex-clusive. They frequently overlap in time, and dataobtained in one type of testing may be useful for thepurposes of another. Although all three types oftests are initiated prior to opening of the system,they may extend well into the period of revenueservice.

The results of testing are of primary interest tothe transit agency installing or modifying an ATCsystem and to its system contractors. The testresults may also be of value to other authoritieswho are planning a similar system. Careful plan-ning of tests, description of test procedures, anddocumentation of results is essential to maximizethe value of testing.

Of particular interest for this report is the ade-quacy of the testing process in terms of planning,procedures, and documentation of results. Also ofinterest is responsibility for testing and evaluationof test results. Finally, the degree to which testresults for one system are utilized at others plan-ning similar systems deserves exploration.

Engineering Testing

Engineering testing begins early in systemdevelopment and includes tests of components and

subsystems to verify that they perform as expected.There are also tests undertaken to diagnose thecause of a problem and assist in its solution. Thissecond category of tests is called “debugging.”

Engineering tests are generally performed by theATC system contractor to support equipmentdesign and manufacture. Results are not alwaysdocumented and are generally not submitted for-mally to the transit authority, A representative ofthe authority may be in residence at the supplier’splant and may monitor engineering test results.NYCTA, for example, follows this procedure.

Because engineering testing occurs early insystem development and there is higher order test-ing later on, it is probably not necessary to havemore formal documentation and wider dissemina-tion of engineering test results than is presently thecustom. Furthermore, manufacturers frequentlyconsider the results of these tests to be proprietary.

Assurance Testing

Assurance test ing includes inspect ion andquality control during production and tests to en-sure that the equipment meets procurementspecifications.

In general, the procurement specifications in-clude provisions for the quality control program.Unfortunately, quality control programs are not al-ways adequate, For example, the BART ATCsystem procurement specifications provided forsuch a program, but strong and effective qualitycontrol was really not achieved. An effectivequality control test program must include not only agood inspection and test program but managementprocedures to follow up and correct deficiencies.

Besides quality control, tests are conducted todemonstrate that equipment meets specificationsfor performance, safety, reliability, maintainability,and availability. Such tests are performed on in-dividual components at the factory or as they are in-stalled, then on subsystems, and eventually on thewhole ATC system. Failure of the equipment toperform according to specifications leads to diag-nostic testing to isolate faults and correct them—another type of debugging. Ideally, these testswould be completed and all deficiencies correctedbefore revenue operation. However, the length oftime required for some kinds of assurance tests(notably rel iabi l i ty) and pressures to beginpassenger service often dictate that operations start

before the tests are completed. Some transitauthorities recognize this necessity by indicating inthe procurement specifications those assurancetests that must be completed before revenue serviceand those that will be accomplished during revenueoperation.

It is important to note that statistically significanttests to demonstrate ATP safety probably cannot beconducted, The required levels of safety are so highthat a valid quantitative test for safety would takeyears or even decades to complete, even if acceler-ated testing methods were employed. As a result,assurance of ATC safety is accomplished by a com-bination of analysis and testing. The analyticalwork is done to identify possible design or engineer-ing defects that could produce an unsafe condition.Testing then concentrates on these areas. While itmay not be able to produce statistically significantresults, test data of this sort can lend credibility toengineering judgments made about safety.

Acceptance Testing

Acceptance testing is the final set of tests on thecompleted ATC system to demonstrate that thesystem meets all procurement specifications, Ac-ceptance testing is specified in detail as part of theATC system contract and usually consists of an’ in-tegrated series of tests which take place overmonths or years. Acceptance testing tends to con-centrate first on safety features, then on perform-ance, and finally reliability and maintainability,Formal tests of the personnel subsystem and man-machine integration are seldom, if ever, conducted.Problems in this area are detected and corrected asthey arise in the course of other testing or opera-tions. The ATC system is accepted by the procuringagency when i t has been demonstrated thatspecification requirements and contractual accept-ance provisions in the contract have been met.

The planning, conduct, and communication oftest results are basic to all three categories of test-ing, The adequacy of documentation of plans, testprocedures, and results was reviewed during thisstudy in order to evaluate the testing process. Thegeneral conclusion is that documentation of testplans has been less than adequate,

From interviews with representatives of transitsystems now being planned, and from examinationof procurement specifications, it is apparent thatthere will be increased emphasis on formal docu-mentation of test plans in the future, For example,

149

the PAAC ATC system procurement specificationsrequire the contractor to prepare and submitvarious test plans appropriate to the differentcategories of testing. As another example, MTARapid Transit Development Division (Baltimore)expects to hire a reliability, maintainability, andsystem safety consultant who will be required toplan a comprehensive and integrated program forthe entire transit system, including the varioussystem assurance tests pertinent to RMA andsafety. This consultant will work with both thegeneral engineering consultant and the ATC systemdesign subcontractor.

Confidence in test results is determined to a largedegree by the detail to which testing procedures aredocumented. Careful attention to details such as ac-curacy, precision of measurement, and control ofthe test environment is important. In some cases, itis difficult to assess the quality of testing that hasbeen conducted in existing transit systems becausedocumentation is lacking or inadequate.

For testing to be of maximum value, the resultsmust be communicated to interested parties. Withina single organization, this may be accomplished in-formally by oral report or internal memoranda.However, in an integrated test program, more for-mal reporting procedures are necessary to assurethat the test results are properly disseminated. As intest planning and performance, there is room forimprovement in the dissemination of results, par-ticularly outside of the transit agency.

R&D may be defined as discovery of newknowledge and its development for use in practicalapplication. R&D must be distinguished from ap-plications engineering which refers to the solutionof specific technical problems. With this distinction

in mind, the following summarizes the organiza-tions which might be expected to perform R&D inATC and their involvement in such activity.

R&D Programs

Operating transit agencies perform very littleATC R&D. Fiscal realities of the operating environ-ment do not support such activity. Operating agen-cies do conduct ATC applications engineering.

Agencies planning new rail rapid systems andtheir subcontractors perform R&D in the course ofsystem development--chiefly design and develop-ment of new hardware, test track demonstrations ofnew concepts, and basic analytical work. Fundsmay be provided for such purposes by the FederalGovernment as part of technical study programsand capital grants. Transit agencies sometimes usetheir own funds to support such work.

The American Public Transi t Associat ion(APTA) is the principal rail rapid transit industryassociation. Some of its committees are active inareas related to ATC, principally safety andreliability. Such work is paper-and-pencil studiesand is supported by member organizations. TheTransit Development Corporation is an industry-organized R & D corporat ion. No programsspecifically related to ATC have been undertaken,

Some R&D in ATC reliability and small vehiclesystems is done by manufacturers. This work issupported primarily by private investment, Therehas been some private investment in test trackdemonstration programs. (See Issue D-10, p. 151.)Most industry work in ATC for rail rapid transit isapplications engineering.

Educational research organizations, such as theUniversity of Minnesota, Northwestern University,Aerospace Corporation, and Applied PhysicsLaboratory, have funded contributions to thel i terature for small-vehicle, f ixed-guidewaysystems. They have not made substantial privatecontributions to rail rapid transit R&D for ATC.

The Federal Government is the principal sourceof R&D funds. Major Federal support to assist test-ing and demonstration of ATC equipment for con-ventional rail rapid systems was given in themid-1960’s in conjunction with the BART andTransit Expressway test tracks. (See Issue D-10, p.151.)

Recent Federal programs have generally beenassociated with support of major vehicle or system

150

concept development rather than ATC as such.These programs include the State-of-the-Art Car(SOAC), the current Advanced Concept Train(ACT I), the TRANSPO ’72 demonstrations, theStandard Light Rail Vehicle, PRT activities atMorgantown, West Virginia, and the now-canceledDual-Mode Program,

In small vehicle technology, a new projectdirected toward the development of a high perform-ance PRT (HPPRT) system has major ATC ele-ments. Also, the Applied Physics Laboratory (APL)of Johns Hopkins University has been providingmore or less continuous support to UMTA in PRTtechnology. Most of the APL work has focused onanalytical studies of operational and reliabilityproblems associated with PRT systems. APL hasalso provided general technical support to UMTA,notably as a technical monitor (with MITRE) of theTRANSPO ’72 PRT demonstrations.

Recent and current work in system assurance hasbeen closely allied with ATC technology and thequestion of manned versus unmanned vehicles. AnUMTA-funded ongoing program in these areas isbeing conducted by the Transportation SystemsCenter (TSC). One product of this work was areport entitled “Safety and Automatic Train Con-trol for Rail Rapid Transit Systems, ” published inJuly 1974. It is expected that the results of the TSCinvestigation of system assurance and the questionof manned/unmanned systems will be available in1976.

Except for the APL work, there has been littlesupport for the development of analytical toolsneeded to evaluate ATC (and other) problemsassociated with advanced technology systems. Thissituation now appears to be changing. A part of thenow-canceled Dual-Mode project was to have in-volved development of the analytical tools neces-sary to evaluate such general concerns as opera-tional strategies and reliability. Such a requirementis included in the later phases of the recently initi-ated HPPRT program.

There are indications that a more programmaticapproach to ATC technology for small vehicles willbe initiated. UMTA is currently developing anAutomated Guideway Technology (AGT) programwhich will deal with many system and subsystemproblems on a generic rather than project-specificbasis. If there are any significant contributions torail rapid transit system of these programs, they arelikely to fall in the area of the development of

methodology and analytical tools. Equipment re-quirements for AGT and rail rapid transit are sodifferent that contributions to rail rapid transithardware technology are unlikely. However, betteranalytical tools would be an important contribution.

Application of R&D

The application of the results of R&D varies ac-cording to the sponsoring organization, Privatelysupported R&D, such as is done by manufacturers,is generally proprietary and not fully available tothe industry. Unfortunately, this is where most ofthe expertise resides,

The results of federally supported research andthat conducted by educational institutions generallyfinds its way into the literature, Much of this workis more theoretical then practical in outlook.Further, such work is often concerned with auto-mated small-vehicle technology rather than moreconventional rapid transit. The increasing involve-ment of the Federal Government in rail rapid transitmay change the situation.

Transi t agencies planning new systems ormodifying old ones generally exchange informa-tion, on a personal basis, with their counterparts atother transit agencies, This helps to compensate forthe lack of research literature and the withholdingof proprietary data held by manufacturers.

ISSUE D-10: TEST TRACKS

What role do test tracks play in ATC R&D?Who operates and funds test tracks?

Test tracks are not built solely for ATC studiesbut to serve several objectives, and their valueshould be judged accordingly. For developmentof ATC, test tracks are used for R&D, demonstra-tion of conceptual feasibility, and hardware testand evaluation. By permitting scientific andengineering work in the absence of constraintsimposed by revenue service, test tracks are vitalto advances in transit technology. Some testtracks have short life spans. Others are more orless perrnanent facilities. They are operated andfunded by the transit agencies, manufacturers,and the Federal Government.

As used here, a test track is a facility built ex-pressly for the purpose of engineering and scientific

151

studies, and not revenue trackage that may be usedfor test purposes. Thus, the Morgantown project isnot a test track. The TRANSPO ’72 exhibition,while perhaps better classed as a demonstration, isincluded because of the post-TRANSPO testprogram. Test track programs discussed below arecategorized by the three types of organizationswhich operate them: transit agencies, manufac-turers. and the Federal Government.

Transit Agencies

BART Diablo Test Track.—The purpose of thistrack was to demonstrate the conceptual feasibilityof alternative subsystems for BART—not, as com-monly thought, to select hardware to be procured.The results of the program were used as a basis forwriting functional specifications for BART equip-ment.

The 41/2-mile test track was located betweenConcord and Walnut Creek, California. It was oper-ated in the mid- to late- 1960’s, at a total programcost of about $12 million. The Federal Governmentsupplied about two-thirds of the funds, and BARTthe remainder. Most suppliers participating in theprogram are believed to have invested substantialfunds of their own.

ATC was 1 of 11 different system elementsstudied at the track. Because the purpose was con-cept demonstration using prototype hardware,reliability and maintainability studies were not partof the ATC test program. Four ATC systems weredemonstrated, Suppliers were General Electric,General Railway Signal, Westinghouse Air Brake,and Westinghouse Electric.84 The results of the for-mal tests were that all four systems met the generalrequirements for BART ATC, with no one systemsignificantly better.

After final ATC specifications were prepared byBART, the winning contractor, WestinghouseElectric, was selected on the basis of low bid.Because the WELCO system was developed inresponse to new specifications and designed to beprice-competitive, it is not surprising that it differedfrom any demonstrated. This system was not subse-quently tested on the Diablo track before finalsystemwide installation. Whether such testingwould have avoided some of the la ter ATCproblems encountered in BART depends upon the

BqThe Philco Corporation also tested portions of an ATCsystem later, after the completion of the formal test program.

type of tests which might have been performed andthe criticality of the analysis of results, rather thanthe particular track used.

PAAC Transit Expressway Program Transit Ex-pressway.—This program, conducted by the PortAuthority of Allegheny County, ran from June 1963to November 1971 at South Park, 11 miles fromdowntown Pittsburgh. The objective was to designand develop a new technology—namely a fullyautomated system of medium-size, light weight,self-propelled vehicles which could be operatedsingly or in trains of 10 or more vehicles. The workwas done in two phases at a cost of $7.4 million.Two-thirds of the funds were provided by theFederal Government; and the remainder was pro-vided by Allegheny County, the State of Penn-sylvania, and Westinghouse Electric.

As the first fully automated transit system, sig-nificant development work was done on ATC. TheATC system underwent major changes between thefirst and second phases of the program. The finalsystem is comparable to BART, with the exceptionof the t ra in detect ion equipment which wasspecifically designed to detect the rubber-tiredvehicles planned for the system.

The importance and value of this program lies inthe many innovations demonstrated there and lateri n c o r p o r a t e d i n t o s y s t e m s n o w o p e r a t i o n a lelsewhere. The ATC technology has been used byWestinghouse Electric for the Seattle-Tacoma andTampa airport systems, for BART, and for the SaoPaulo METRO in Brazil. PAAC used the project todevelop procurement specifications for TERL, aprogram recently defeated by the voters.

Manufacturers’ Test Tracks

Manufacturers’ test tracks have been built pri-mari ly for work on automated small-vehiclesystems. These tracks are used either to developnew systems, to check equipment prior to delivery,or both. Federal funds may be used, as was the caseof the Dashaveyor and Varo test tracks which wereused for feasibility studies conducted by these com-panies for AIRTRANS at the Dallas-Fort Worth air-port. Some company test tracks that have been usedfor ATC development or checkout are:

● Dashaveyor, Pomona, Calif.

● Varo Monocab, Garland, Tex.

● WABCO Monorail Division, Cape May, N.J.

152

. TTD, Denver, Colo.

● Bendix, Ann Arbor, Mich.

● Alden, Milford, Mass.

Federal Government

TRANSPO ‘72.—Four automated small-vehiclesystems were demonstrated at TRANSPO ’72 andlater evaluated in a test program conducted be-tween August and November 1972. Federal fundsamounting to about $7 million were provided forthe demonstration and test program. There was alsosubstantial private investment. The exact amount isunknown, but it is thought to be of the same orderas the Federal contribution. The systems demon-strated and their manufacturers were:

. Dashaveyor System—Bendix Corporation

. ACT System —Ford Motor Company

● Monocab System—Rohr Industries

. TTI System-Otis

The systems were developed under tight timeconstraints with limited funds. This led to somecompromises in the ATC system design. The post-TRANSPO test program showed that some of theATC equipment had undesirable control charac-teristics, including long delay times and speedoscillation. It was concluded that the basic cause ofthese problems was the prototype nature of theequipment.

Apart from its value as a public demonstration ofn e w t e c h n o l o g y , t he ma jo r bene f i t o f t heT R A N S P O ’72 program was the increasedcapability in small-vehicle technology gained by thefour participating manufacturers, Because of basicdifferences in philosophy and operating charac-teristics between automated small-vehicle systemsand rail rapid transit and because of the lessstringent demands placed on a system in an exhibi-tion (in comparison to a revenue operation), theTRANSPO ’72 program had limited value in im-proving ATC systems for general transit industryapplication.

Pueblo Colorado Test Facility.—DOT’s HighSpeed Ground Transportation Center at Pueblo,Colo,, became operational in 1973, Managed by theFRA, the Center can test several types of groundtransportation systems. Both advanced systems andrail technology programs are conducted, Theformer programs include the Tracked Levitated

Research Vehicle (TLRV), the Tracked Air CushionResearch Vehicle (TACRV), and the Linear Induc-tion Motor Research Vehicle (LIMRV), For railtechnology programs, the Center includes 20 milesof conventional railroad trackage, used for studyingtrain dynamics under a variety of track and gradeconfigurations, a 9.1-mile oval rail transit trackwith a third rail for testing electrically powered roll-ing stock, and a Rail Dynamics Laboratory forsimulator testing of full-scale railroad and rail tran-sit vehicles. As a part of the now-canceled Dual-Mode Program, it was planned to build two guide-way loops at the site, each 2 miles in circumference.

Probably the most significant rail transit activityat Pueblo was the testing of the State-of-the-ArtCar (SOAC) in 1973. There was little ATC relatedwork associated with this R&D activity, and theATC provisions at Pueblo are all but nonexistent.There are several reasons for this. DOT has beenusing the facility for other purposes. Limitedfacilities are available. (For example, there are noprovisions for inserting signals into the rails,) Thesite is very remote from both operating propertiesand equipment suppliers. Most transit agencies feelit is essential to conduct final ATC developmentwork in the actual operating environment (at-mospheric, electrical, etc.) where the equipmentwill be run. Unless there are specific federallyfunded programs requiring that the work be con-ducted at Pueblo, it seems unlikely that significantamounts of ATC research for rail rapid transit willbe conducted there.

The MITRE Corporation (1971) conducted asurvey of rail rapid transit agencies and equipmentmanufacturers to identify problems that should beaddressed in a federally funded research program.Of the 11 top priority areas indicated by this survey,none had any direct relationship to ATC. Theresults must be accepted with some caution because

153

.: .’ - t

FIGURE 72 DOT Test Track, Pueblo, Colorado

154

none of the industrial firms surveyed were ATCequipment manufacturers and because the intent ofthe study was to identify problems for investigationat the DOT Pueblo test site. (As indicated earlier inIssue D-10, p. 151, the Pueblo test facility is notsuited for investigation of ATC problems.) Still, thesurvey does suggest that ATC is not viewed as amajor R&D problem by a significant part of the tran-sit industry.

During visits to transit agencies made by BattelleColumbus Laboratories as technical consultants inthis assessment, comments and suggestions weresolicited on R&D needs in rail rapid transit tech-nology, particularly those associated with ATC.Here again, the results indicate that ATC is clearlynot a major concern.

Operating transit agencies felt that the majorR&D needs were:

● Improvement of chopper control, multiplexingof train lines,85 and a.c. traction motors;

● Documentation of slip-slide tests for use inofficial and expert testimony in damage andinjury suits;

● Clarification of the trade-off values associatedwith such technical matters as analog vs.digital signals, control signal frequencies andmodulation rates, types of station stops, chop-pers vs. cam controllers, and the use of p-wire;

● Review of the availability and allocation ofradio frequencies for both voice and datatransmission by transit systems;

● D e v e l o p m e n t o f a d a t a b a s e a n dclearinghouse for rel iabi l i ty and main-tainability information for the benefit of tran-sit systems and manufacturers.

Transit systems in the planning and constructionstages had a differing set of priorities:

● Investigation of electromagnetic interferenceproblems;

● Improvement in the rel iabi l i ty of ATCsystems and related equipment;

. Study of techniques for, and the value of,regenerative braking;

. Establishment of a data bank on the safety,

●

●

reliability, and maintainability experience ofoperating transit systems;

Maintenance training programs to ensure thatnew and sophisticated transit equipment (in-cluding but not limited to ATC) can be pro-perly cared for;

Studies of collisions and crash resistance, par-ticularly for small-vehicle systems.

Since one of the main purposes of this tech-nology assessment was to weigh the need for R&Din the area of automatic train control, this topic wasgiven special attention. In addition to review of theliterature and collection of opinion within the tran-sit industry through the interviews cited above, thematter of research needs and priorities was madethe subject of a separate investigation by the OTATransportation Program staff and the OTA UrbanMass Transit Advisory Panel. This investigationdrew especially on the experience of individualpanel members and of various transit systemmanagers, equipment manufacturers, technicalconsultants, and DOT officials. The findings of thisinvestigation, as they apply to rail rapid transit, arepresented below.86

At the outset, it should be noted that there is noneed for a significant R&D effort to make major ad-vances or innovations in ATC technology for railrapid transit systems. The basic technology is suffi-ciently developed for present and near-term futurepurposes. What is needed now is research anddevelopment to refine the existing technology andto improve performance at reduced cost. The majorelements of such a program are discussed below.Figure 73 is a matrix, categorizing the importance ofthese R&D efforts against the estimated relativecost to carry them out.

Reliability and Maintainability

There are several aspects of reliability and main-tainability in which further work is needed.

Equipment Reliability and Maintainability

There is a major need to develop more reliableand maintainable equipment. This applies notjust to ATC but other types of rail rapid transitequipment.

aSThe underlined items are those directly or indirectly re-lated to ATC.

aeR&D needs for automated small-vehicle systems are ex-plored in a separate OTA report, Automated Guideway Transit:An Assessment of PRT and Other New Systems, June 1975.

155

RESEARCH AREAS

RMA Analytic Techniques

Equipment RMA

RMA Data Bank

RMA Standards

Safety Methodology

Technology Transfer

Handicapped Requirements

Standardization

●

FIGURE 73. ATC Research and Development Priorities and Relative Cost

Techniques for RMA Analysis

Improved and more quantitative methods areneeded to evaluate total system performance interms of rel iabil i ty, maintainabil i ty, andavailability. Component performance measuresexist. Total system performance measures donot. Total system measures would permit betterallocation of reliability requirements among sub-systems, better understandingtrade-offs, and better utilizationnance work force.

RMA Standards and Guidelines

An effort is needed to establishment standards and to clarify

of reliabilityof the mainte-

realistic equip-manufacturers’

responsibilities in the area of RMA. The stand-ards must be high enough to assure reasonableavailability of equipment but not so high as tomake the equipment unnecessarily costly.

156

Reliability and Maintainability Data

A pool of data from testing and operational ex-perience pertaining to equipment reliability andmaintainability y would be of great value to transitsystem planners, research groups, and manufac-turers. At present, there is no uniform way ofrecording and reporting such information, and noclearinghouse for collecting and disseminating itwithin the transit industry.

Safety

The safety levels of the rail rapid transit industryare high and exceed nearly all other forms of publicand private transportation. Still, there is a need forresearch in two aspects of safety.

Train Detection

The much publicized train detection problems ofBART (which are probably no more severe than

those experienced in other transit systems) haveunderscored the need for clarification of thestandard for train detection and the need for auniform method to test the performance of traindetection systems.

Safety Methodology

Controversy over system safety versus fail-safeprinciples abounds in the transit industry, Thereis also debate over how safety is to be measuredand how safe is safe enough. Research is neededto develop an objective and quantified methodfor evaluating the safety aspects of rail rapidtransit system performance.

Man-Machine Relationships

Function Allocation

There is great variability among transit systemsin the duties assigned to the human operator. Sig-nificant errors were made in the original designof the BART system because of the highlypassive role assigned to the train attendant. Theman-machine interface needs to be carefullystudied to determine the optimum role of thehuman operator in automated systems and to en-sure that provision is made for the operator to in-teract effectively with the system in abnormal oremergency situations. The role of personnelassigned in a supervisory capacity needs to besimilarly examined.

Cost-Benefit of Automation

Research is needed to determine the relative ad-vantages of manual and automated methods ofoperat ion with respect to energy savings,variability of trip time, equipment utilization,system capacity, and manpower costs. Such datawould be of value not only in the design of newsystems but also in the .modernization of oldones.

Application of Technology

Even though ATC is a rather mature and welldeveloped technology, there remain some problemsof practical application. Three areas are in need ofspecial attention.

Standardization

There are a number of technical and economicbenefits to be gained from reducing the diversityof ATC equipment now in use or planned for in-stallation in rail rapid systems. These advantagesmust be scrutinized and evaluated against thedisadvantages of inhibiting innovation and im-peding improvement that standardization mightbring.

Technology Transfer Within the Industry

There is a general shortage of persons with ex-perience in ATC system design, manufacture,and operation at all levels in the industry. Thisshortage is most keenly felt by agencies planningand building’ new systems. Research is needed todevise more effective methods for sharing infor-mation, exchange of experienced personnel, andtraining of new personnel.

Requirements for the Handicapped

Under the stimulus of the Federal Government,there is an increasing concern in the transit in-dustry with the transportation needs of the han-dicapped. As a part of the investigation of thegeneral social costs and benefits of providing railrapid transit service for the physically, visually,and auditorily impaired, there is a need to con-sider the specific influence of ATC. Among thematters of interest are acceleration and decelera-tion limits and their effects on system capacityand trip time, passenger assistance on trains or instations with a low level of manning, and thesafety of the handicapped and others in emergen-cy situations.

157

Chapter 7

POLICY AND

INSTITUTIONAL FACTORS

t,..

INTRODUCTION

Rail rapid transit is a public entity. Transitsystems are built with public funds to serve publictransportation needs. The public is involved in theconception and planning of the system-both inpublic referenda to approve the building of thesystem and in citizen participation programs duringthe planning process. The planning and operatingagencies, themselves, are quasi-public bodies,whose directors are responsible to a State ormunicipal government or to a local electorate. Theoperation of the transit system, especially in recentyears, may be subsidized with some combination ofFederal, State, and local funds. As a consequence,the form and operation of a rail rapid transit systemare strongly influenced by public policy and institu-tional factors,

Three forms of influence can be distinguished:legislative, regulatory, and institutional

Legislative influence is manifested through thecontent, authority, and impact of laws enacted atthe Federal, State, and local levels of government.Generally, such laws serve one of two purposes:regulation or promotion. In the earliest days ofpublic transit, and continuing somewhat afterWorld War II, the intent was primarily regulatory.Laws were enacted to control the private firms thatprovided public transportation and to ensure thatthe public interest was protected.

Since the middle of this century, the purpose oflegislation pertaining to public transit has shifted tothat of promotion and subsidy. This shift was coin-cident with, and occasioned by, the precipitousdecline of the transit industry to the point that itwas threatened with extinction. As a result, most ofthe recent legislation has been aimed at promoting aresurgence of public transportation. These lawsauthorize the expenditure of public funds (often inlarge amounts) to design, build, and operate transitsystems. These laws also establish and supportresearch programs to advance the state of tech-nology and to broaden its application. While noneof this legislation has dealt specifically withautomatic train control technology, this aspect oftransit system design and operation has benefitedfrom the general increase of financial support forrail rapid transit.

Regulation, although no longer the predominantpurpose of legislation, is still a major concern at alllevels of government. The oversight and control of

transit systems is an important function of Federaland State agencies, especially in the area of safety.Local governments tend to place more emphasis onthe regulation of fares and levels of service. Thepattern of regulatory legislation is far from static;and, like promotional legislation, it appears to beextending--especially at the Federal level—morewidely and deeply into the area of system opera-tion.

Institutional factors are manifested primarilythrough actions of the transit industry, labor unions,and—to some extent—the public at large. While notso clearly defined or so easy to isolate as legislationand regulation, these institutional factors also serveto shape the course of transit system developmentand operation.

The purpose of this chapter is to examine theissues raised by ATC in the area of public policyand institutions. In some cases, these ATC issuesare not wholly distinguishable from the generalcontext of rail rapid transit system developmentand operation. These larger, systemwide topics willnot be treated, however, except as background tothe particular aspects of ATC or the reciprocaleffects that policy and institutions have on traincontrol system technology and its application.87

SUMMARY OF EXISTINGLEGISLATION

Most of the legislation relating to rail rapid tran-sit is of recent origin, and none contains specificprovisions for the promotion and regulation of ATCper se. Nevertheless, this legislation (especiallyFederal laws) does have an indirect effect uponATC design and development through the generalsupport provided to rail rapid transit technology.The following is a summary of the Federal lawswith sections germane to ATC.

Urban Mass Transportation Act of 1964 (PL88-365)

In general terms, the Act of 1964 provided threeforms of financial support:

87For an examination of the more genera] policy issues per-taining to transit system planning and development, see theOTA report, An Assessment of Community Planning jor MassTransit, November 1975 (Report Nos. OTA-T-16 through OTA-T-27).

161

●

●

●

Grants or loans to assist State or local agenciesin acquisition, construction, or improvementof transit facilities and equipment88

Grants to State or local agencies for planning,engineering, and design studies related tomass transit;

Grants for research (including new tech-nology) and training.

Funds in all three categories have been used fordevelopment and acquisition of ATC systems.Capital grants have been made both for the purposeof upgrading existing ATC equipment (e.g., the re-cent cab signal installation programs in CTA andMBTA) and for planning and constructing newsystems with advanced forms of train control (e.g.,WMATA and MARTA). Funds available under the1964 Act have also been used to support severalATC-related research and development activities,both within DOT and by outside R&D organiza-tions.

The 1964 Act also contains two specific sectionsthat have an influence on decisions related to traincontrol system automation:

●

●

Transit employees adversely affected by anyfederally assisted project must receive specialconsideration, including protection of rightsand benefits;

Transit systems must afford accommodationto the special needs of the elderly and handi-capped.

Protection of individual workers (not specificjobs) is contained in section 13(c) of the Act. whichalso requires that clearance for a grant be obtainedfrom the Department of Labor. The Act allows theelimination of jobs, but only as workers presentlyholding those jobs retire or vacate the positions forother reasons. Thus, economic benefits of work-force reduction through automation of an existingtransit system may be deferred for a number ofyears until retraining, transfer, or attrition can ac-count for the displaced workers. Alternatively,direct compensation can be paid to affectedworkers, eliminating the jobs earlier but at anearlier cost. As noted previously, however, it ap-pears that few employees are actually put out of

880rigina11y, the 1964 Act provided for two-thirds Federalfunding, with one-third State and local matching. In 1973, theAct was amended to increase the Federal share to 80 percent.

162

work by increased automation of existing systems,New systems do, in fact, have smaller train crews,but this work force reduction is largely offset by theincreased need for more and higher skilled workersto maintain the more sophisticated and complexATC equipment89.

The Urban Mass Transportation Act of 1964, andits amendments, directs that consideration be givento the means of providing service to, and assuringthe safety of, the elderly and handicapped. This hasraised problems that are not yet fully resolved with-in the transit industry. The chief concerns related toATC are control of door operation and emergencyevacuation of vehicles in automated systems with-out an onboard operator. There is also uncertaintyabout how accommodation of the elderly and han-dicapped will affect the service offered to otherpassengers in normal operations and their safety inemergencies.

Department of Transportation Act of 1966 (PL89-670)

This Act created administrative and supervisorybodies of the Federal Government that now have amajor influence on transit system development andoperation as a whole, and ATC in particular. TheAct established both the Urban Mass Transporta-tion Administration (UMTA) and the NationalTransportation Safety Board (NTSB), UMTA is theprincipal DOT organization by which grants andFederal assistance to transit development are ad-ministered. NTSB is charged, inter alia, with over-seeing the safety of transit systems and with acci-dent investigation.90

Federal Railroad Safety Act of 1970 (PL 91-458)

This Act placed the safety of rail rapid transitsystems within the purview of the Federal RailroadAdministration (FRA). To date, however, FRA hasnot actively pursued this interest, apparentlybecause of preoccupation with problems of intercityand commuter railroads, As discussed below inIssue P-z, there is evidence within recent months of

W)st;(> ]ssl If, [).1 Z, l)f>g i n n i ng on pilg(~ I I ~, for a fl I rth[~ r tr(~a t -ment of this point.

wThe NTSB investigation of the BART Fremont accident,entitled “Safety Methodology in Rail Rapid Transit SystemDevelopment,” August 1973, and an earlier report, “SpecialStudy of Rail Rapid Transit Safety,” June 1971, raised severalimportant questions about the advantages and disadvantages ofATC.

more active involvement of FRA in rail rapid tran-sit.

National Mass Transportation Assistance Act of1974 (PL 93-503)

This law is, in effect, a significant amendmentand extension of the Urban Mass TransportationAct of 1964. Its major provisions include allocationof additional funds for urban mass transportationprograms and—for the first time—makes Federalfunds available on a fifty-fifty basis for operatingexpenses. Under the 1964 Act, Federal support wasavailable only for capital expenditures.

Some of the sections of the 1974 Act specificallyrelate to ATC are:

● Section 5 (n) The provisions of section 13(c)of the Act of 1964 are made applicable to allassistance under the formula grant program.

● Section 107 The Secretary must investigateunsafe conditions in facilities, equipment, andoperations funded under the Act of 1974,which result in serious safety hazards. If un-safe conditions are found, he may withholdassistance until appropriate actions are taken.

It is still somewhat early to assess the generaleffects of this Act on transit system development orits specific impact on ATC. Opinion on these sub-jects is mixed within the transit industry and in theFederal Government, and evidence from transitsystem operation is still too fragmentary to indicatetrends.

State, Regional, and Local Legislation

Before 1964, when the Federal Governmentbecame involved in capital grants to mass transit, fi-nancial support was almost exclusively the concernof State and local governments. Such support, whengiven, was usually for publicly owned systems. Pri-vate transit operators, while subject to variousforms of State and local regulation, typicallyreceived no support from public funds and werealmost wholly dependent upon the fare box forrevenues.

As transit ridership declined and operationsbecame less and less profitable, private operatorsoften severely curtailed services. Eventually, manyfound it impossible to continue. It became neces-sary for public bodies to assume control in order to

prevent the total loss of these transit systems to thecommunity.

At the State, regional, and local levels, and occa-sionally by interstate agreement, legislation hasbeen enacted to set up various public or quasi-public agencies for operation of public transitsystems. These organizations take a variety offorms. Some are purely operating authorities.Others also have planning responsibilities. Mostcontrol all modes of transit in their area of jurisdic-tion. A few (such as BART and PATCO) operateonly a rail rapid transit system.

Many States have formed Departments ofTransportation for the purpose of coordinatingmass transit activities on a statewide basis, In largepart, State DOT efforts are concentrated on obtain-ing a larger share of Federal funds or increasing theeligibility of local agencies to participate in Federalprograms. There has also been considerable supportfor mass transit at the State level in the form ofdirect subsidies, special taxing plans, and publicassistance programs such as transportation ofschoolchildren and the elderly,

It is difficult to generalize about these State andlocal legislative structures except to indicate thatthe concern is primarily on the public serviceaspects of the system as a whole. Technologicalcharacteristics, chiefly as they relate to safety, doreceive attention in those States which have apublic utilities commission established to regulatetransit system operation, In some States, however,the transit agency itself is charged with regulatingits operation.

ISSUE P—1: IMPACT OF EXISTING. ‘ LEGISLATION

The Urban Mass Transportation Act of 1964 andits amendments has been of enormous help to therail rapid transit industry for the planning and con-

163

struction of new systems and for the modernizationof equipment and facilities in existing systems. Thebenefits of this law stem not only from the largeamount of money made available by the FederalGovernment for capital grants but also from the-in-centives offered to State and local governments toparticipate in capital acquisition and improvementprograms by providing matching funds. During thedecade since enactment of the 1964 law, major im-provements have been made in the New York,Boston, and Chicago rail rapid transit systems, andw o r k o n n e w s y s t e m s h a s b e e n s t a r t e d i nWashington, D.C., Atlanta, and Baltimore.

The recently enacted National Mass Transporta-tion Assistance Act of 1974 continues the policy ofFederal support for mass transit and, for the firsttime, extends Federal assistance for operating costs.Under the 1964 Act Federal support was providedonly for capital improvements and acquisition(two-thirds Federal funds, one-third matching Stateand local funds). The 1974 Act authorizes UMTA toprovide capital grants (on an 80-20 basis) andoperating aid (on a 50-50 basis). Table 34, a sum-mary of the UMTA budget for Fiscal Year 1976, in-dicates the magnitude and distribution of theFederal Government’s assistance program for masstransit.

TABLE 34. UMTA Budget for Fiscal Year 1976

AMOUNTUMTA PROGRAMS (millions of dollars)

Capital Grants 1,100.0Operating Aid 500.0(Carryover) 150.0Technical Studies 38.7

system improvement and growth fostered byFederal Government assistance,

There is a widely held view in the transit indus-try that the 1964 Act may have had the effect of en-couraging the development and use of automatedtrain control systems. Because the Act providedgrants only for capital improvements or acquisitionand not for operating assistance, planners may havebeen induced to concentrate their resources oncapital-intensive features such as automatic traincontrol (which would be eligible for Federal assist-ance) in the hope of thereby reducing later operat-ing costs (for which Federal assistance funds werenot available).

The argument is plausible, but it does not seemto be supported by events. First, the amount ofmoney for train control systems provided by the1964 Act has been relatively small, probably not

. more than 2 to 5 percent of the total capital as-sistance program, and it is doubtful that such anamount could have had the imputed effect, Second,the ATC projects that have been undertaken in thisperiod and supported by Federal funding have beenjustified on grounds other than potential manpowersavings through automation, At CTA and MBTA,for example, the justification for cab signal installa-tion was safety of operation not labor saving, Itshould also be noted that the two most automatedsystems placed in service during the time the 1964Act was in force (PATCO in 1969, and BART in1972) were planned and built without expectationof Federal assistance.91 Further, the new systems inAtlanta and Baltimore, for which preliminary plan-ning and design took place between 1964 and 1974,employ lower levels of automation than the BART

R&D and Demonstration Grants 53.4 system. If the 1964 Act had had the influence pur-Managerial Training 0.6 ported by some persons in the transit industry, justUniversity Research 2.0 the opposite would have been expected, i.e.,Administrative Expenses 12.5

MARTA and Baltimore MTA would have a degreeTotal 1,857.2 of train control automation equal to or surpassing

that of BART. Thus, it seems unlikely that FederalGovernment policy, as expressed in the 1964 Act,

There is no evidence that either the 1964 Act or has tended to foster automation.

the 1974 Act has had a specific impact on ATC tech-Federal Railroad Administration (FRA).—Thenology or its application in existing and planned

transit systems. The provisions of these laws areFRA, which has long had jurisdiction over the

quite general, and there is no explicit or implied safety of interstate and commuter railways, has in-terpreted the Federal Railroad Safety Act of 1970 tosupport provided for ATC in particular. While there

have been ATC programs undertaken with funds 91BART did receive some Federal assistance in the lattermade available under the 1964 Act, and some pro- stages of development and construction, but this was long afterposed with funds from the 1974 Act, they represent the commitment had been made to a highly automated form ofno more than a part of the general pattern of transit train control.

164

confer upon it (through delegation by the Secretaryof Transportation) authority for safety regulation ofall transit systems using rail technology. To date,however, FRA has not actually exercised thisauthority over urban transit systems except to in-stitute a standardized procedure for accident report-ing and to announce proposed rulemaking withregard to train protection systems and the safety ofdoor operation.

With regard to train protection, FRA is consider-ing the possibility that cab signals and onboardautomatic stopping devices should be required forall rail rapid transit systems. This requirementwould apply only to new systems, and some excep-tions would be granted to existing systems thathave a heavy investment in wayside signals and tripstops. The concern of the FRA with door operationcenters on how to prevent accidents in whichpassengers are caught or struck by doors. Prelimi-nary hearings on door safety have been held and theviews of the transit industry have been solicited.FRA has not yet decided the approach to be taken,but their stated intention is to regulate the force andmanner of door closure and the safety interlocksbetween door operation and train motion.92

FRA’s sphere of authority is confined to thesafety of equipment already in use. They are notable to exert direct control over the design processfor new systems, However, the FRA can wield in-direct control since they could shut down-or pre-vent the startup of—any new system not meetingthe safety regulations in force for operatingsystems.

Urban Mass Transportation Administration(UMTA).—At the present time UMTA does notperform a regulatory function, in the commonly ac-cepted sense of the term. However, some form ofregulatory authority does appear to be implicitwithin the general responsibility of UMTA to over-see and administer funding for the development ofnew systems. Certain of UMTA’s requirements fortransit system development programs verge onregulation—for example, the requirement that tran-sit districts requesting capital grants for new systemconduct studies of transportation system alterna-

g21n passing, it should be noted that the FRA’s concern withthe safety of door closure did not arise from rail rapid transit in-cidents but from operating experience on commuter railroadsregulated by the FRA.

tives and trade-offs. Also, the Safety Division of theUMTA Office of Transit Management has proposedinitiation of a comprehensive “system safety”program, which might later be broadened to cover“system assurance, ” Under a system assuranceprogram, the concerns of safety would be integratedwith those of reliability and maintainability. If thisis done, the domain of regulation would be ex-panded to include all aspects that contribute to safe,efficient, and reliable transit system operation.While local transit agencies might not be requiredto conduct such programs by UMTA regulation, thecontrol of grant funds exercised by UMTA wouldhave considerable mandatory force. In fact, severaltransit agencies have already instituted systemassurance programs in anticipation that it mightbecome a future requirement for obtaining UMTAgrants.

UMTA may soon begin investigating transit acci-dents as a regular activity. Section 107 of the Act of1974 requires the Secretary of Transportation to in-vestigate serious safety hazards in systems whoseconstruction or operation is financed with FederalGovernment funds, and UMTA is a logical choicewithin DOT as the agency to carry this out.However, it is intended that such investigationswould be conducted only after a serious accident orincident had occurred and not as a routine before-the-fact activity.

In the past 2 years, UMTA has made use of theDOT Transportation Systems Center (TSC) to carryout in-depth investigations of ATC in two new

BART and the PAAC skybus. Thetransit systems—former was an investigation of a newly openedsystem which was in the midst of controversy overthe safety of the ATC system. The latter was an in-vestigation of the proposed ATC system for a tran-sit system still in the preliminary design stage. Ofparticular interest there was PAAC’s intent to oper-ate rubber-tired vehicles with no onboard person-nel. At the present time, TSC is also assistingUMTA in design reviews of the MARTA transitsystem now being built in Atlanta. Among the areasof concern to the TSC participants are the safetyaspects of the ATC system design and the effective-ness of the integration of man and machine func-tions.

Some members of the transit industry have ex-pressed concern about these TSC activities. Theyfear that TSC might gradually assume a regulatory

165

function for rail rapid transit--especially by settingdown rules, standards, and guidelines that mightbecome the basis for a de facto form of regulation.UMTA’s position is that, while TSC may continueto provide technical assistance, there is no intent toassign any sort of regulatory role to TSC.

N a t i o n a l T r a n s p o r t a t i o n S a f e t y B o a r d(NTSB).—The Department of Transportation Actof 1966 created the NTSB which, while not pre-cisely a regulatory body, has had influence on tran-sit system safety in general and ATC in particular.The NTSB is empowered to investigate rail rapidtransit system accidents (as well as accidents inother types of transportation) and to make recom-mendations to the Secretary of Transportation con-cerning procedures and equipment that affect thesafety of operation. NTSB has made a number of ac-cident investigations and special studies and hasproduced several significant recommendations.

One report, entitled “Safety Methodology in RailRapid Transit System Development” (August 1973)has engendered strong controversy because it in-cluded a recommendation for “abandonment of thefail-safe concept.” The NTSB view, which isdiametrically opposed to traditional railroad andrapid transit practice, has brought adverse commentfrom all segments of the transit industry. The reportwent on to recommend that, as a replacement (orperhaps more correctly a supplement) for fail-safe,the industry adopt “an organized approach to ac-complishing rapid transit system safety through theapplication of current safety management andengineering concepts. ” Without entering into themerits of the NTSB argument, this report can becited as a major impetus for the system safety andsystem assurance programs now being consideredby UMTA.

The role of NTSB appears to be expanding. TheTransportation Safety Act of 1974 contains provi-sions which require NTSB to conduct a muchbroader program of accident investigations thanthat set forth in the Department of TransportationAct of 1966 that established NTSB. It is estimatedthat this will involve over 700 additional accidentinvestigations per year in rail transportation alone.Whi l e r a i l r ap id t r ans i t i s no t men t ionedspecifically, NTSB will be expected to investigateall fatal railroad accidents, all accidents involvingpassenger trains, and all rail transportation acci-dents resulting in substantial property damage.

Regulation of rail rapid transit systems is carriedout at three levels of government: Federal, State,and local. In some cases, the transit system operat-ing authority may also be self-regulatory, Until re-cently the concern of regulatory bodies at all levelshas been essentially limited to the area of safety.Since, in the traditional view, safety involves pre-vention of collisions and derailments, regulatory in-terest has centered almost exclusively on ATP sub-systems and equipment, Now that automation hasbeen extended into train operation and supervision,the scope of regulatory agency concern is broaden-ing to include all aspects of safety and to deal withsafety on a system-wide basis. Aside from safety,other aspects of system operation (with possible ex-ception of fare structure and level of service) havereceived little or no attention from regulatory agen-cies.

Federal Regulation

Three Federal Government agencies have partialjurisdiction over safety matters. These agencies andtheir areas of responsibility are described briefly onthe following pages.

Under the 1974 Act, NTSB still is not vested withany rule-making authority or power to establish re-quirements that specific safety-related actions orremedies be effected, As before, the primary role ofNTSB will be to investigate and make recommenda-tions to the Secretary of Transportation, who willretain the authority to accept and enforce therecommendations as seen fit,

An important point with regard to NTSB recom-mendations, which is not always recognized by the

166

public, is that NTSB does not attempt to evaluatethe economic or technical aspects of its recommen-dations. The sole concern of NTSB is to maximizethe safety of a transportation system, The respon-sibility for evaluating feasibility and cost-benefit isleft to the appropriate regulatory agency and thelocal authorities who must ultimately decide on acourse of action.

State Regulation

In many States, regulatory bodies were created inthe 1900-10 period to oversee transit operation andprotect the public from the monopoly power of pri-vate owners. As such, these State agencies werealmost exclusively concerned with economicregulation. With the shift of local transit systems topublic or quasi-public ownership and operation inthe 1940 and 1950 decades, these agencies were leftwith vestigial responsibilities, and some ceased toexist. Few of these State agencies, then or now,have been active in safety regulation. As a practicalmatter, then, most local transit authorities are self-regulated in the areas of both economics and safety,

During this study it was found that many transitauthorities considered themselves to be essentiallyself-regulated, but perhaps subject to requirementsimposed by such agencies as the State DOT orPublic Utilities Commission, the State legislature,or even a regional planning commission of one sortor another. Transit systems serving areas which in-clude the State capital appear to receive substantialattention from the State legislature, although notnecessarily in the form of regulation. An example isthe State of Georgia Legislature which created theMARTA Technical Overview Committee. Thiscommittee is empowered to Iook into any or allaspects of the MARTA system. Also, the State ofMinnesota Legislature has taken an active interestin the activities of the Twin Cities MTC and, in1973, directed that a special study of PRT alterna-tives be performed because they were not com-pletely satisfied with the results of previous studies.

In some States, the public utilities commission(PUC) has had an active role in the regulation of railrapid safety, often with specific interest in the ATCsystem. Two notable examples are in California andMassachusetts.

In 1967, the California PUC issued specific re-quirements dealing with ATC. Their coverage wassomewhat general but they specifically addressed

the subject of ATC. These requirements, a result ofsection 29047 of the California Public Utilities Code,state that BART shall be subject to safety regula-tions of the PUC and that the commission shall in-spect BART facilities for safety of operations andshall enforce the provisions of the section.

The Massachusetts PUC has taken an active in-terest in the ATC system installed on the MBTA’sSouth Shore extension, and has ordered that fullyautomatic operations be restricted until such timeas the PUC is satisfied that no potentially unsafeconditions exist.

Industry Self-Regulation

Regulation of a transit system by an externalagency is not an easy matter, It requires establishingan organization, staffed by technically competentand experienced personnel, to write standards,review plans and designs, and conduct tests and in-spections. Even if the necessary personnel could befound, it might not be practical at the State or locallevel to create such an agency. Typically, a Statecontains only one rail rapid transit system; and toestablish a special authority to oversee a singleoperating agency might be a governmental ex-travagance.

For this reason, most publicly owned transitagencies are self-regulated, both for safety andeconomic matters such as fares and level of service.As public or quasi-public bodies, they respond tothe influences of the political system by which theyare created and to the economic constraints im-posed by the use of public funds.

The opinion within the transit industry is thatself-regulation is a workable solution. The excellentsafety record of rail rapid transit is cited as proofthat a self-regulating body can manage its affairs ina responsible manner, with the public interest as aforemost concern. The opponents of self-regulation,while not questioning the integrity and sense ofresponsibility of the local transit system officials,point out the inherent danger of vesting a singleagency with the authority to conduct transit opera-tions and oversee the results. Both sides of the argu-ment have merit, and one of the basic issues in thearea of public policy for rail rapid transit is to find aproper balance between external regulation by aState or Federal agency (or some combinationthereof) and responsible management by the localoperating authority.

167

ISSUE P-3: ACCEPTANCE TESTING AND

QUALIFICATION

What part is played by regulatory bodies inthe testing and qualification of ATC systems?

Except for the public utilities commissions incertain States, regulatory agencies are seldom in-volved in testing and qualifying a syatem for ini-tial service. Up to now, the Federal Government”has not taken an active role in this area.

Before a transit system is placed in service, eachof its major components and finally the system as awhole must be subjected to acceptance andqualification tests. Customarily, this testing is car-ried out by the engineering staff of the operatingauthority, often with the assistance of technicalconsultants and manufacturers’ representatives.The State regulatory agency (typically a publicutilities commission) may observe some part of thetests and may receive the results for review, but theState agency usually does not take an active role intesting and rarely, if ever, conducts independenttests to verify that the system performs according tostandards. Regulatory agencies of the FederalGovernment (FRA, UMTA, and NTSB) are not in-volved at all in acceptance and qualification testing,and they do not perceive that they have a legislativeor organizational mandate to do so.

Thus, with regard to testing and qualification,local transit authorities tend to follow the pattern ofself-regulation. The reasons are primarily those ofpractical necessity and not explicit Governmentpolicy. Automatic train control equipment, likemost other components of a modern transit system,is complex and technologically sophisticated. For alocal or State agency to conduct tests of this equip-ment would require a staff of technicians trained inthe use of sophisticated instrumentation and ex-perienced in train control system operation. In viewof the general shortage of such qualified personnelin the transit industry, State agencies find them-selves in a position where they must compete withthe transit authority, manufacturers, and consultingfirms for the few persons available. Further, Stateagencies may be at a competitive disadvantagebecause they cannot offer the salaries, prestige, oropportunities for advancement that are available inan operating transit organization, a manufacturer,or an engineering consultant firm.

Scarcity of technically qualified personnel is notthe only reason. The program of testing necessaryfor a local or State agency to qualify a system forservice is virtually the same as the test program nor-mally pursued by the operating authority itself inassuring that the equipment performs according tospecification and manufacturers’ warranty. Becauseof this, public agencies have been reluctant toestablish separate organizations to engage in effortsthat would largely duplicate those of the operatingsystem they are charged with regulating, especiallysince there may be only one transit system withinthe entire State. And even if the regulatory agencywere willing to do so, it might be difficult to con-vince the State government and legislature thatsuch would be an effective and necessary use ofpublic funds.

Most State agencies have found that the practicalcourse is to monitor the tests conducted by the localoperating agency and to review the findings toassure conformance with established standards,either those of the State agency or those of the tran-sit industry generally, In some instances, the Stateagency has entered into a cooperative arrangementwith the local transit authority, whereby certaintests are conducted by the local authority on behalfof the State or whereby State standards have beenadopted by the local agency.

In passing, it should be noted that the primary,and almost sole, concern of State regulatory bodiesis the safety of the system, specifically the designfeatures that prevent collisions and derailments.The broader aspects of passenger safety and opera-tional concerns such as reliability and availabilityare almost never matters of regulatory action,

The history of the transit industry has shownthat, because of their size and complexity, newsystems are almost never completed by theirscheduled opening date, As deadlines are missed,public impatience and political and economicpressure mount. Because acceptance and qualifica-tion testing is usually one of the last items on theschedule, the local operating agency is stronglytempted to shorten the test program or to defer apart of it until after the system has opened for serv-ice. The State regulatory body, influenced by thesame pressures, may find it necessary to acquiesce,

To assure that acceptance and qualification test-ing is not slighted in these circumstances, somehave suggested that the Federal Government(through either UMTA or FRA) should require a

168

certain amount of testing before a new system is putinto service, It is argued that only the FederalGovernment has the authority and the resistance tolocal pressure required to ensure that the interest ofpublic safety is not compromised by expediency,There is also an economic justification advanced.The Federal Government, having provided as muchas 80 percent of the funds for development and con-struction, has the major interest in the new systemand should assure that full value has been receivedfor the investment of public moneys, A third, andpurely practical, reason for Federal Government in-volvement is-that only at the national level would itbe feasible to assemble and maintain a technicalorganization capable of carrying out such tests.

There are strong counterarguments. As a matterof policy, it is debatable that the Federal Govern-ment should enter into an area where State agenciesand local self-regulation have traditionally ruledand where the general adequacy of such regulationhas been demonstrated. Further, it may not be cor-rect to view financial support of local transit systemdevelopment as an investment by the FederalGovernment. Rather, it may be an instance ofrevenue sharing without the Federal Governmentacquiring proprietary interest. On practical grounds,the imposition of Federal-level requirements fortesting may add unnecessary delay to the accept-ance and qualification process because of the needto submit test plans to a Federal Government agen-cy for approval, to have the test results reviewed,and to obtain additional authorizations to open thesystem for service. There is also the possibility that,if disputes arise between the local transit authorityand the Federal Government agency, the accept-ance and qualification process might be evenfurther protracted.

The unresolved issue of responsibility for accep-tance and qualification testing is part of the largerquestion of how and by whom should regulation oftransit systems be accomplished, The question isnot, of course, confined to the subject of ATC; it ap-plies to all aspects of transit systems developmentand operation, Still, the matter of acceptance andqualification come most sharply into focus in thearea of train control systems because of the vitalpart played by ATC in passenger safety. There is aclear and present need to assure, by some combina-tion of local, State, and Federal regulation andsupervision, that technology is used wisely in thepublic service.

ISSUE P-4: STANDARDIZATION

What effects would standardization have o n

ATC?

There would be both positive and negativeeffects. The benefits of a uniform technology liein the areas of improved system assurance andreduced research and development costs. Themajor disadvantages are restrictions on innova-tion and limited freedom of choice in systemdesign.

Few fields of technological endeavor run the fullcycle from experimentation to mature developmentwithout the introduction of standardization. Atsome point, the establishment of design and per-formance standards becomes desirable to checkprol i ferat ion of design var ia t ions, to reducedevelopment costs, to limit technological risk, andto assure that best use is made of existing tech-nology. The real issue is not whether to standardize,but when and to what degree. If standards are im-posed too early or too rigidly, innovation and tech-nological improvement may be stifled. If too late,the variety of designs may be so great that thestandards become meaningless, and there may beeconomic hardship for those who own or manufac-ture equipment that lies outside the prospectivestandard.

At this time, the matter of standardization ofATC is an open question. Some argue that it wouldbe healthy for the transit industry and the generalpublic whose tax moneys are used to support thedevelopment and installation of new ATC systems.Others contend that it would be unwise to standard-ize now at a time of great experimentation and in-novation because many promising avenues of im-provement might be cut off. The following is a briefexamination of three areas where standardizationmight be most applicable,

Procurement Specifications

As ATC equipment has become more complexand sophisticated, the specifications governing thedesign and procurement of this equipment havegrown more detailed and explicit with regard tosystem performance and contractor responsibiIities.At the present time, however, each transit agencyprocuring a new system or upgrading existing in-stallations writes a more or less unique specifica-

169

— — . . . .

tion, tailored to local needs and conditioned by theirindividual experience with ATC,

There would be an advantage for all if therewere a uniform set of terms, conditions, and pro-cedures for the transit industry. This might take theform of a model specification, establishing a stand-ard terminology and defining basic methods forverifying compliance. A model of this sort need notbe a completely rigid document; there should beroom for variation to accommodate local needs andconcerns. Further, the specification would notestablish uniform performance requirements; thesewould be left to local decision. But it would stand-ardize the phrasing of these requirements and setforth a universal method for acceptance. andqualification testing.

This approach would offer several advantages. Itwould assure that a well-thought-out documentwas available to the planners and directors of newsystems for guidance in an area where they mightbe lacking in experience, It would help assure thatthe best of past experience and current practice isincorporated in new systems. There would also beadvantages for ATC equipment manufacturerssince a standard set of requirements and procedureswould permit contractors to know exactly what isexpected of them and would provide continuity andregularity from one procurement to the next. Forthe public, standardization might lead to benefitssuch as reduced engineering and development costsand elimination of some subsequent operationalproblems.

Against these advantages must be set three majordisadvantages. A detailed specification is of ques-tionable value for simple procurements; it mightresult in overly elaborate and unnecessarily expen-sive provisions without materially enhancing thequality and performance of the equipment. It maynot be possible, at the present state of technologyand specification writing, to produce a documentwith sufficient generality to cover all situations andstill exercise meaningful control over the details ofdesign and performance. Finally, there is somequestion whether the methodology of systemassurance is sufficiently well developed and preciseto permit its application to ATC systems.

ATC Characteristics

There is a tendency for the planners anddevelopers of ATC systems (and transit systemsgenerally) to design to their unique goals and re-

quirements. In some cases, this is justified; non-standard solutions are needed to meet special localproblems and conditions. In other cases, however, itis not clear that the additional benefits of a special-purpose design justify the increased costs. Increasedstandardization of ATC system equipment offersthe promise of substantial economic and opera-tional benefits. On the other hand, there is the riskthat standardization could inhibit innovation andtechnological progress. The major arguments forand against standardization of ATC equipmentcharacteristics are enumerated below.93

The standardization of ATC equipment charac-teristics could produce several positive effects. Itwould tend to reduce the variety of designs and theproliferation of special-purpose equipment. Itwould help assure use of the best of proven tech-nology in new systems. Commonality of equipmentwould make it easier to obtain and stock spare parts(an important consideration for small transitsystems), Standardization could lead to some sav-ings in equipment engineering and acquisitioncosts, and perhaps substantial reduction of debug-ging costs (which are higher for new designs thanfor already proven equipment).

There are some possible negative effects ofstandardization, There is such a wide range of tech-nology now in use in existing systems that it wouldbe difficult to establish a common core of ATCequipment characteristics, There is no one type ofdesign that is clearly superior to others or that is ap-plicable to the broad range of conditions that existin transit systems, Freezing design characteristics at.this time, when there are some promising innova-tions just coming on the scene, may minimize theopportunity for technological progress. The deci-sion to select a particular system or systems asstandard might work a hardship on those who useor manufacture “nonstandard” equipment andmight adversely affect industry competition.

Test Procedures for Train Detection

The basic and proven method of train detectionis the electrical track circuit. While the track circuitis highly effective and reliable in most circum-stances, there is a long history of difficulty with

QaNOte that the area of standardization is ATC equipmentcharacteristics, not components or specific designs. Componentand subsystem design could be improved and refined (for exam-ple, to increase reliability) while still retaining the same funda-mental characteristics.

170

electrical train detection on little used track whererust and film may build up and inhibit rail-to-railshunting by the train wheels and axle. The ex-perience in BART has brought this problem to theforefront of attention in the transit industry, but it,has existed in the background for years in othertransit systems. Many in the transit industry believethat is a need for a redefinition of the performancestandards for train detection circuits and for im-proved testing methods.

Standardization of test criteria and procedureswould have the primary advantage of providing auniform and objective way to verify the perform-ance of train detection circuits and would therebyassure that effective train protection is achieved.This would have also a secondary benefit, in that apotential source of misunderstanding (and litiga-tion) between the buyers and sellers of ATP equip-ment would be largely eliminated. However, thereare some offsetting disadvantages. The problem oftrain detection is so complex and influenced by somany extraneous variables that it may not be possi-ble to develop a single, universally applicable stand-ard and testing method. Even if such a standard andtest could be devised, it might prove to be overlyconservat ive and could lead to excessivelycomplex equipment or unnecessarily redundantmechanisms.

ISSUE P-5: SAFETY ASSURANCE

Is action by the Federal Government needed toensure the safety of ATC design and operation?

Federal action may be required to establishsafety standards, methods of measurement, andtesting procedures. Many in the transit industrybelieve, however, that such could be broughtabout internally by the process of self-regulation.

As noted in the discussion of Issue P–z, Regula-tion, most transit systems (both operating andplanned) are essentially self-regulating in mattersof safety. While many members of the transit indus-try recognize the need for improvement in safetystandards and methodology, they believe that thesafety record of rail rapid transit demonstrates theeffectiveness of self-regulation and that direct ac-tion by the Federal Government is not required,They also argue that local self-control, while

perhaps not an ideal method, is preferable to inter-vention by a Federal agency because the localofficials are much closer to the needs and problemsand more likely to be responsive to the concerns ofthe public in the area served. This position is notstrictly a “hands-off” policy. Many local transitagency officials feel that the Federal Governmentcould be of substantial help in the matter of safetyassurance, but primarily in a supportive and adviso-ry capacity and not in the role of a direct regulator.

There are, of course, counterarguments. The in-creasing complexity of transit systems (and theATC equipment that controls train movement) hasgreatly magnified the difficulty of insuring that allelements are safe and reliable throughout the life ofthe system. The task of safety assurance may thushave grown beyond the capability of a local operat-ing agency to deal with it systematically and effec-tively. Perhaps only an organization at the nationallevel could command the resources and have theauthority to cope with the problem. Perhaps also,only a national organization could be expected todevelop a sufficiently uniform and impartial set ofstandards to ensure that safety matters are handledequitably and consistently throughout all the transitsystems in the country. If a transit system is con-sidered not as a local public utility but as part of thenational transportation resources, then Federalregulation can be further justified on the groundsthat the Federal Government is the only body capa-ble of overseeing the service of national interests.

If external regulation of rail rapid transit safety isdeemed necessary, there are three principal mattersthat need to be addressed: safety standards,methods of measurement, and testing procedures.

Safety Standards

How are the elements of an ATC system to per-form under normal and abnormal conditions? Whatare the requisite fail-safe characteristics of ATCsystems? What level of protection must be providedfor passengers, train crew, and equipment in theevent of failure or malfunction? And finally, whatdegree of risk must exist before a system or situa-tion is considered unsafe?

Methods of Measurement

There is a need for common definitions andmethods of measurement. It would be of little valueto standardize ATC systems and to develop ageneral ATC system specification without also

171

defining what characteristics are to be measured cisco area), but it was greatly out of proportion toand how such measurement is to be accomplished. the degree of actual injury and damage. Since then

the public concern over safety seems to have sub-Test Procedures sided to an insignificant level and revives only mo-

The procedural aspects of testing need to begiven attention, Uniform procedures would helpassure that testing gives valid results and that noimportant aspect of system performance is over-looked. Uniform procedures would also helpguarantee consistency of treatment and evaluationfor all transit systems in the country.

mentarily in response to some new safety incidentor publicity surrounding the ongoing engineeringtests of the BART system. Transit operatingofficials in other cities such as New York, Chicago,and Boston remarked during the course of this studythat the same pattern of public concern for safety ismanifested there in response to accidents andmishaps.

Closely allied to safety in the public mind is thematter of security from criminal acts while ridingtrains or waiting in stations. Public concern does

There is almost no information available on theattitudes of the riding public toward transit systems.Judging from newspaper coverage and individualexpressions of public opinion through the newsmedia, the public tends to take a transit system forgranted until some specific problem occurs. When itdoes, public reaction is likely to be negative andnarrow in focus, centering around the incident itselfand ways to prevent recurrence, Public concern isseldom of long duration and recedes as the normalpa t t e rn o f t r ans i t ope ra t i ons i s once moreestablished and memory of the incident is eclipsedby other interests.

Transit system operators believe that the publicis primarily concerned about personal safety whileriding the trains and about security from robberyand crimes upon their person, Again, however, thelevel of safety (i.e., the number and frequency of ac-cidents and injuries) is such that public concernabout personal risk comes to the fore only when amishap occurs. The case of BART is a classic exam-ple. Before the Fremont accident, there appeared tobean unspoken acceptance of the safety of the ATCsystem, The public reaction to the accident wasprompt and widespread (even outside the San Fran-

of assistance when needed. That is, the public doesnot take a stand against ATC because it wouldreduce the level of manning of the trains (andperhaps the stations). Rather, the concern is withthe measures that may be employed to compensatefor the absence of crewmembers.

An interesting demonstration of the public’sviews took place in Denver, where a system ofsmall unmanned vehicles was proposed, Duringpublic hearings, numerous questions were raisedabout how muggings and assaults could be pre-vented or discouraged, what form of monitoringwould be used, and what actions would be takenafter detection of a crime. The suggestion that vehi-cles could be monitored by central control “listen-ing in” by two-way radio was considered by someas a form of eavesdropping, and therefore unac-ceptable.

There seems to be a general feeling that transitsystems should be safer than the general urban en-vironment. Crime rates in transit systems aregenerally lower than in the city at large, and yet thefear of criminal acts seems to be higher in subwaysthan on the streets. Paradoxically, efforts by transitauthorities to increase the security of patrons issometimes a two-edged sword, The presence oftransit police may be reassuring, but it may alsogive the impression that the transit system propertyis so dangerous that extensive policing is necessary.NYCTA is a case in point, This system has a verylarge transit police force that actively patrols trainsand stations, and yet public concern over “crime inthe subways” is perennially high.

172

With regard to the dependability of service, thepublic does not usually distinguish the role of ATCfrom that of other elements of the transit system.Either the trains run on time or they don’t. If thereare delays or habitual disruptions of service, thepublic is most likely to lay the blame on manage-ment in general rather than any particular compo-nent of the system. Also, it seems that the publicdoes not regard lack of dependability as so serious amatter as safety. Nevertheless, the public does castits negative vote. With the instant dependability ofthe automobile ever present, public dissatisfactionwith transit service usually takes the form ofpatronage diversion from public to private transpor-tation. Transit system managers, on the other hand,regard dependability as virtually coequal to safetyas a way of attracting public patronage. It is perhapsfor this reason that transit system publicity tends to

stress the speed, convenience, and dependability ofmass transit in their advertising to attract riders.

The public attitude toward cost is most diffuseand hard to isolate. If the individual citizen is amember of the fraction of the population thatpatronizes rail rapid transit, he pays the fare butprobably does not think about how the costs are dis-tributed. For the rest of the public the costs of con-structing or supporting a transit system (or anyspecific part such as ATC) is indirect, ill-defined,and probably unnoticed. Where there is public reac-tion to the cost of a transit system, it usually is ingeneral terms and in connection with a publicreferendum on the issue of transit system develop-ment bonds or taxation. On such occasions, the costof ATC specifically is submerged in the total cost ofthe system.

173

. . ‘, .. .

APPENDICES

—

Appendix A

TRAIN CONTROL SYSTEM FUNCTIONS94

Train control is the process by which the move-ment of rail rapid transit vehicles is regulated andsupervised to assure safety and efficient operationof the system. Such control can be effected bymanual means, by automatic devices, or by somecombination thereof. The description of train con-trol system operation presented below is cast infunctional terms that apply equally to manual orautomatic forms of control. Because automation isthe central issue of this report, the discussion alsoincludes an examination of the relative merits ofman and machine components. A discussion of thetechnology of automatic train control is presentedlater, in appendix B.

The train control system is comprised of ele-ments that perform four major types of functions:

● Train protection

. Train Operation

● Train Supervision

● Communicat ions

TRAIN PROTECTION

The sole purpose of the train protection system isto assure the safety of vehicle movement by pre-venting collisions and derailments. Traditionally,the train protection system is functionally separateand distinct from other elements of the train controlsystem; and it is designed so as to protect not onlyagainst failure of other system elements but againstfailure of its own elements as well.

Before taking up specific train protection func-tions, it is necessary to consider the general concept

y~This appendix is based on material originally prepared byBattelle Columbus Laboratories in support of the OTA study.TfIe editors gratefully acknowledge the contribution of BattelleColumbus Laboratories but accept full responsihilit y for the ver-sion presented here and for any alteration of content that mayhave been introduced in condensing and editing the material forpublication,

of train protection and its role in the overall schemeof system operation.

Figure A-1 is a conceptual representation of trainprotection functions in a typical rail rapid transitsystem. The indicated functions might be per-formed by men, machine, or both; and they mightbe performed by elements in locations other thanthose shown in the diagram, Since the purpose ofthe illustration is simply to indicate functional rela-tionships, these different forms of implementationcan be ignored for the moment.

The train supervision system may generate a re-quest for the movement of trains or switches. Someof the requested moves may be unsafe. It is theresponsibility of train protection system to insurethat only safe moves are carried out. In order to dothis, it is necessary to know the status of switches,the location of trains, and the allowable speeds fortrains as affected by track limitations and the pres-ence of other trains in the system.

Wayside logic typically performs the route in-terlocking function by processing information onthe desired action, the location of trains, the statusof switches, and the allowable speed of trains. Out-put of this logic consists of safe commands to moveswitches and safe speed commands issued to trains.

Excluding certain track and train surveillancefunctions, the primary concern of the onboard trainprotection system is the restrictive control of thepropulsion and braking system. Essentially, trainprotection enables or inhibits the performance ofcertain propulsion or braking actions. Train protec-tion determines actual train speed and compares itto the commanded safe speed. If the train is exceed-ing the commanded speed by a predefine amount,action to override all other less restrictive propul-sion and braking functions is initiated. Similarly,the status of critical elements on board the train aswell as certain conditions on the track are used todetermine what action should be taken regardingthe propulsion and braking system.

177

178

Train and Track Surveillance

Train and track surveillance involves monitoringthe train, the track, and areas immediately adjacentto the track for safety-related conditions. It can alsoinvolve monitoring adjacent tracks and trainsoperating on them. passenger security, thoughclearly a safety-related matter, is not usually con-sidered a part of the train and track surveillancefunction, Door monitoring and control, anothersafety-related function, is considered here to be atrain operation function. 95 Train and track sur-veillance are essentially human roles in all operat-ing rail transit systems today, but the amount ofhuman involvement varies widely.

Train surveillance is concerned with monitoringthe status of the train and its passengers. Onboardoperators (motormen and conductors) traditionallyperform this role. The primary advantage of thehuman in such a role is his ability to comprehendand interpret many diverse types of inputs. Exceptat PATCO and on the MBTA Green Line, the opera-tor of a train is typically confined in a space that isphysically removed from the passengers. His pri-mary role in train surveillance is monitoring thestatus of the equipment. Conductors, if present,have more freedom of movement and can some-times go to the scene of a possible problem to deter-mine its nature and severity.

Passengers may provide some train surveillancefunctions. In systems where they can communicatewith employees, they may report onboard condi-tions. Two-way communications systems are pro-vided at Sea-Tac and AIRTRANS, where the vehi-cles are unmanned. It is likely that passengers couldbecome more involved in train surveillance in auto-mated small-vehicle systems.

Onboard operators provide the track surveillancefunction at all operating rail transit properties. Inthe closed environment of Sea-Tac, essentially notrack surveillance is performed. At AIRTRANS,another special environment, only minimal tracksurveillance by roving employees is provided.

Under ideal conditions, little, if any, track sur-veillance would be required. Humans external tothe system cause most of the problems requiringtrack surveillance. Trespassing, vandalism, andsuicide attempts are three of the most commonly

gssome transit engineers consider door control to be a trainprotection function because of its relationship to safety.

cited factors which make some form of track sur-veillance necessary. Here again, the human is un-surpassed in the ability to identify and deal with avery broad range of track surveillance problems.

While a human can act to prevent some acci-dents, he cannot prevent all of them, partly becausehe simply cannot stop the train in time. If one wereto assure an instantaneous response and brake ap-plication along with a rather high braking rate of 3mphps, it can be calculated that the minimum stop-ping distance from 60 mph for a typical train is 880feet, and 220 feet at 30 mph. Clearly, there are manysituations in which the potential hazard is either notvisible at this distance or is created within the stop-ping distance of the train, (Suicide attempts are theclassic example here.)

Damage assessment is a track and train sur-veillance function which can be performed byhumans. When something has happened, a humancan assess the damage to track or train and deter-mine if it is safe to proceed.

Train Separation

The function of train separation is to maintainphysical separation between following trains sothat they are not in danger of colliding with eachother. In the simplest manual system, train separa-tion can be provided by the operator who drives thetrain much as a person drives an automobile. Hemust know the maximum safe speed with which hecan approach curves and other places of limitedvisibility, and he relies upon seeing the train aheadand taking appropriate action to prevent a rear-endcollision.

Figure A-2 illustrates the basic principles in-volved in automating the train separation function,The dashed line indicates the theoretical speed-dis-tance relationship that a following train couldmaintain and still be able to come to a stop beforereaching the end of the train stopped ahead.

When a block-type detection and speed com-mand system is used, the location of the train with-in a block is not known accurately, Therefore, thetrain must be assumed to be in the most hazardouslocation, i.e., at the rear of the block. In the exampIeshown, the train is almost out of block BC but mustbe assumed to be at the location shown in dottedlines. The shift of the theoretical speed-distanceprofile can be seen to be essentially one block long.In general, the shorter the blocks, the closer the at-tainable spacing between trains.

179

—-

FIGURE A–2,—Train Separation in a Conventional Block Signaling System

Many arrangements of speed commands arepossible. For example, if the commands were onlystop or go, all signals shown (except perhaps theones at A and E) would indicate stop. It is fre-quently desirable for operational reasons to have atrain approach closer than the safe stopping dis-tance from full speed. In such a case, an intermedi-ate speed command would be provided, as shown inthe block DE. A train traveling in this block wouldreceive an intermediate speed command followedby a zero speed command from point D. One canreadily see that the shorter the length of blocks andthe greater the number of speed commands, themore closely a train can follow the theoreticalspeed-distance profile. Obviously, the system foraccomplishing this becomes more complex and ex-pensive as block length is shortened.

It should be noted that a number of factors usedin the actual design of a system are not shown inthis simple illustration. There is some systemresponse time involved before braking is initiated.Grades may reduce the actual effectiveness of thebrakes. No safety factors are included. Somesystems provide additional blocks between trains,All these tend to further increase the spacing be-tween trains.

The difference between intermittent and con-tinuous speed command transmission can be seenhere. Suppose the train is moving and, in a brief in-crement of time clears block BC. Block CD will im-

mediately indicate the intermediate speed andblock DE will go to full speed. If a following trainwere just a few feet into block DE when the speedcommand in that block changed, the operator wouldhave no way of knowing it if the speed commandwere transmitted by a visual signal located at E. Hewould have no indication of a change in signalstatus until the signal at D became visible. By con-trast, the continuous, or cab, signal would im-mediately indicate to the operator that he could in-crease his speed. An additional advantage of cabsignals is that a train can move into block DC to stopwhereas, with wayside signals and trip stops, thetrain would stop at D, or even farther back.

All rail transit systems to date have beendesigned on the assumption that a leading train iseither stopped or will stop instantaneously. It can beseen that trains could follow one another muchmore closely if a less stringent assumption could bemade about the stopping of a leading train. This isprecisely what is proposed in many short-headwayPRT systems where position, velocity, and ac-celeration of a lead vehicle would all be consideredin establishing headways. Some studies of PRTheadway have shown, however, that removing thebrick-wall concept reduces minimum achievableheadway by only a small amount.

So far as is known, there are no rail rapid transitsystems planning to abandon the traditional trainseparation philosophy. The major differences in

180

train separation practice are associated with theway in which train separation commands are en-forced, Intermittent or wayside signaling systemsprovide train separation by use of trip stops.96 Con-tinuous or cab signaling systems enforce trainseparation using onboard equipment. A safe speedcommand is transmitted to the train and equipmenton board insures that the operator initiates action toslow the train as appropriate.

Humans play a role in train separation under atleast some situations. In some locations, operatorsare permitted to approach trains ahead of themmore closely than would be permitted by the signalindication, This is always done under strict pro-cedural controls. It maybe done at highly congestedstations or in emergency situations. Practicallyspeaking, the maneuver is identical in nature todriving one bus up behind another bus which hasstopped.

Route Interlocking

Route interlocking is the process by which trainsare prevented from making conflicting moves, i.e.,moves that would be unsafe. Typical conflictingmoves are those which would cause a train to col-lide with another train, to go off the end of thetrack, or to run through an open draw bridge. Routeinterlocking involves monitoring the presence andposition of the trains in a system and the positionsof the track switches. The information from themonitors is processed by logic, usually the frontcontacts of vital relays, and used to inhibit or permitthe movement of the trains. As an example, when atrain is dispatched from one location to another andthe trip involves passage of the train through one ormore track switches or crossovers, the route in-terlocking allows the train to proceed through theswitches and crossovers only when it is safe to do soand prevents other trains from entering the routeuntil the first train has safely passed.

Information on the presence and location of thetrains is obtained from the train detection system,

96A trip stop IS a mechanical device which is located near Orbetween the running rails; it is associated with a wayside signal.When the wayside signal indicates that a train is to stop, the tripstop is positioned so that it will apply the brakes of any trainwhich attempts to pass the wayside signaL When the waysidesignal indicates that a train can proceed, the trip stop ismechanically positioned so that it does not affect the brakes of apassing train. Thus, the trip stop enforces a stop command pre-sented by the wayside signal.

as described earlier. Information on the status andthe position of the track switches is monitored bythe route interlocking. Before a route is alined for atrain, it must be determined that the proposed routewill not be in conflict with an existing route foranother train. If no conflict exists, then the ap-propriate track switches must be positioned andtheir positions verified. Then each switch must beimmobilized and locked until the passage of thetrain has been verified. These precautions arenecessary to insure that the switch positions are notchanged after the route has been alined and to in-sure that a switch is not moved under a train. Eitherof these events would be unsafe.

Route interlocking is an essential part of the trainprotection for all but the most simple transitsystems. As system complexity increases, route in-terlocking assumes greater importance. Early routeinterlocking functions were often accomplishedthrough the use of complex mechanical deviceswhich prevented establishing potentially hazardousswitch positions. Some such equipment is still inuse. New installations are all equipped with electri-cal or electronic logic which, may also permitremote actuation of switches and signals.

Overspeed Protection

Overspeed is the condition where the actual trainspeed is greater than the intended or commandedspeed. Overspeed can be dangerous because thetrain may derail if it goes too fast around a curve orthe train may have a collision because it is going toofast to stop within the available distance. It is theresponsibility of other train protection system ele-ments to determine the allowable safe speed and toassure that the commanded speed does not exceedthe allowable safe speed.

Basically, overspeed protection has two inputsand one output. The inputs are commanded speedand actual speed. A signal enabling or inhibiting thepropulsion and braking system is the output. Acomparator, either a man or machine, compares ac-tual speed with commanded speed and determinesif propulsion power can be applied or if a brake ap-plication is required. The overspeed protectionfunction can be accomplished on board the vehicleor through the use of wayside equipment.

All transit systems that use cab signaling alsohave automatic overspeed protection. In order to dothis, it is essential that the onboard speed measuringand comparing device have a virtually zero prob-

181

— . —. —

ability of failure in an unsafe mode, Even thoughsingle tachometers traditionally have been regardedas “fail-safe,” redundant tachometers are some-times used. The outputs are compared, and if dis-agreement exists, a failure is assumed to have oc-curred and the overspeed protection logic treats thisas an overspeed condition. A more frequently usedapproach, though, is a fail-safe speed measurementsystem not requiring redundancy that reducesreliability y.

It is not uncommon for a cab signal display to failwithout failure of the overspeed protection system.Under such conditions, the operator can run thetrain safely but may have difficulty in maintainingthe desired speed without exceeding the overspeedlimit, The audible warning devices that are nor-mally provided permit the operator to run withoutcab signals without receiving a penalty brake ap-plication.

OverSpeed can be detected and controlled fromthe wayside. Through the use of timing circuits andknown lengths of track, it is possible to determine ifthe average speed of a train over a certain distanceis equal to or less than the allowable safe speed, Ifthe measuring distance is short, an essentially con-tinuous overspeed protection can be provided, If themeasuring interval is long (say tens or hundreds offeet), only an average measure can be obtained, sothe instantaneous value could exceed the intendedlimit,

TRAIN OPERATION

Train operation consists of three major func-tions:

Velocity Regulation

Station Stopping

Door Control and Train Starting

In the traditional concept of signaling, train opera-tion is not considered a safety-related aspect of traincontrol. However, there are some safety aspects oftrain operation. If abrupt starts, stops, and changesare made, passengers may be thrown down and in-jured. If door control is assumed to mean themonitor ing of the s ta tus of the t ra in doors ,passenger safety is also involved,

There is some disagreement among train controlengineers concerning the functional relationshipsamong train operation, train protection, propulsion,and braking systems, Some consider control of jerk,

slip-slide, and flare-out as train operation functions;others consider them to be propulsion and brakingfunctions. Some consider door control a part of trainprotection (because of its relation to safety); othersplace door control within the province of trainoperation. These subjects will be touched uponbriefly below.

Velocity Regulation

Overspeed protection is designed to prevent atrain from going too fast. Velocity regulation is con-cerned only with controlling the speed of the trainin response to operational needs. Velocity regula-tion systems are “nonvital,” that is, they are not es-sential to the safety of the system.

Velocity regulation may be accomplished by aman or machine, When a man acts as the controller,he simply compares the actual speed with desiredspeed and tries to minimize the difference betweenthe two. The desired speed may be presented in theform of wayside or onboard displays. Actual speedis determined from speedometers on board thevehicle, A human controller uses a handle of somesort to control speed much as an auto driver uses anaccelerator pedal. In such a system, the hand on thehandle is the interface between the train controland propulsion and braking systems.

A machine controller performs exactly the samefunctions as a man, but the control signal is pro-vided in the form of an electrical signal to a con-troller in the propulsion and braking equipment.Most ATO systems to date have provided this signalin a combined digital and analog form. The systemdesigned for WMATA uses a completely digital in-terface,

For operational reasons, it is sometimes desirableto modify the speed of the train. This is usuallycalled performance level modification, Here, a com-mand (verbal, visual, or electrical) is transmitted tothe train telling it to run at a selected fraction of thecommanded s a f e speed . Pe r fo rmance l eve lmodification is not a safety-related function.

Performance level modification is normally ac-complished on board the vehicle. In manualsystems, an operator may receive a verbal or visualinstruction to operate the train at reduced speed. Insome systems, notably NYCTA, performance levelmodification is not normally used, Trains are heldat stations rather than operated at slower speedsbetween stations, Both BART and WMATA pro-

182

vide transmitters in stations and at other criticallocations to send performance level requests to thetrain. Performance level modification is thus ac-complished automatically without operator inter-vention. Baltimore is considering a performancelevel modification system in which a visualwayside display is provided to an onboard operatorwho then manually sets the desired performancelevel for the next segment of the trip.

There is a general trend toward the use of auto-mated velocity regulation. The two newest systemsin operation (BART and PATCO) employ auto-mated speed regulation, NYCTA is planning to in-stall it (and other features) over a long period oftime with the objective of eliminating the conduc-tor, MBTA has installed velocity regulation on thenew portions of the Red Line. CTA, however, optednot to use automated velocity regulation on its newcab signaling installation.

Station Stopping

Station stopping involves bringing the train torest at a selected location along a station platformunder some form of programed control. It is nottechnically a safety-related function. Both humansand machines perform the station stopping func-tion.

In manually operated systems, the operator nor-mally uses some reference mark as an indicator ofthe point where he should initiate braking. Thismark may be any wayside object, possibly a markerplaced for the specific purpose of braking reference,A skilled human can ordinarily stop a train withinan accuracy of a few feet. Variability in trainweight, performance characteristics, and track con-ditions affect the human’s ability to stop a train pre-cisely. The required deceleration rate also affectshis performance. The higher the average rate, thegreater the variability in result.

The degree of sophistication of automatedprogram stop equipment is a function of the ac-curacy required. PATCO utilizes two “triggers”spaced some distance away from the station asreference points for programed stopping, The firsttrigger initiates maximum-rate deceleration. A sec-ond trigger, roughly at the end of the platform,causes the ATO package to measure the train’sspeed and adjust the deceleration rate accordingly,A manually set switch in the train cab is used todefine train length so that the braking action will

cause the train to be centered on the platformregardless of its length. Under adverse weather con-ditions, the operator makes the stop manually, ini-tiating deceleration at a point marked by a yellowpole on the wayside.

Where both station and train doors are used, it isnecessary to aline the train with the doors within anaccuracy of a few inches. Both Sea-Tac andAIRTRANS have such a system. Information ontrain weight, instantaneous position, speed, anddeceleration may be processed by an onboard com-puter to achieve precision stopping. At BART, along wayside antenna provides the position signalsnecessary for the onboard program stop computer.Other needed information is derived and processedon board.

Door Control and Starting

Some engineers do not consider door control andstarting to be train protection functions since theopening and closing of doors present no hazards tothe train. Clearly, however, the safety of passengersis affected by door operation, so it is common to in-terlock door functions with train protection, a prac-tice that leads some engineers to regard door controlas a part of train protection.

Three basic pieces of information are requiredfor the control of door opening, It is necessary todetermine that the train is in a proper location fordoors to be opened. If there are doors on both sidesof the train, the proper side must be identified ateach station, Assurance that the train is stopped andwill not move is required. (Clearly some of these re-quirements must be overridden in emergency situa-tions.)

On starting, four conditions should exist before atrain leaves a station. The doors should be closedand locked. No passengers should be caught in thedoors. It should be time for the train to depart. Thetrain protection system should indicate that it issafe to move the train.

In manual systems, most of the door control,monitoring, and starting functions are performed byhumans. When a conductor is on board, control andmonitoring of doors is ordinarily his most importantassignment. Lights are usually provided to indicatethe status of all doors. (It is worth noting that a 10-car train may have as many as 40 doors, each withtwo leaves, on each side of the train. Thus, 160 doorleaves must be monitored during the movement of

183

the train through the system.) Because there is notruly foolproof practical door, all U.S. rail rapidtransit systems have onboard personnel to act as aback-up to insure that door closure is not initiatedwhen passengers are boarding and leaving and toverify that no one is caught when the doors areclosed and locked.

Where trains are unmanned, a more complexform of door control is required, In special environ-ments such as the Sea-Tac and AIRTRANSsystems, door systems much like those of elevatorshave been used. The platforms are enclosed anddoors on both train and platform must be closed andlocked before the train can move. Doors withpressure-sensitive edges are used to prevent possi-ble entrapment of passengers in a closing door. Alldoor functions are interlocked with the train protec-tion system.

Jerk Limiting

Jerk is defined as the rate of change of accelera-tion. Control of jerk contributes to a smooth rideand from a rider standpoint, a somewhat safer one.Customarily, jerk limiting is a function of the pro-pulsion not the train control system.

Jerk limiting applied during stopping is some-times termed “flare-out control.” It is identical tothe maneuver that a skilled automobile driver per-forms just as the car is coming to a stop. By easingoff on the brake, the transition from deceleration tofull stop is smoothed out. Because there are safetyimplications to releasing the brakes, flare-out con-trol is usually designated to be either a part of thetrain protection system or to be interlocked with it.

In a manually operated transit system the flare-out function is performed by the operator much as itis performed by an automobile driver. The smooth-ness with which the function is performed dependsto a great extent upon the skill of the operator. In anautomatic train control system, the flare-out func-tion can be performed automatically by sensing thespeed of the time when the train velocity becomesless than some predetermined amount. It is neces-ary that this reduction in braking effort be allowedto persist only for a short period of time. Otherwise,the braking system of the train could be disabled.Accordingly, flare-out is controlled by a timer sothat the reduction of braking effort can persist onlyfor a few seconds. During normal operation thesefew seconds are sufficient to bring the train to acomplete halt, and the brakes then are re-applied. It

is essential that the design and the implementationof the flare-out system be such that a failure cannotpermanently withhold braking action. Figure A-3 isa schematic diagram of a typical automatic flare-outcontrol system.

Slip-Slide

Slip refers to the slipping of wheels during theapplication of power. Slide is concerned withwheels sliding when brakes are applied. Slip or slideoccurs when the tractive effort of the train exceedsthe adhesion capability of the wheels and rails. Ex-cessive slip, which occurs during acceleration, candamage the propulsion equipment, wheels, andrails, Slide, which occurs during deceleration, candamage the wheels and rails; a wheel locked duringbraking can be ground flat on the bottom if it isdragged very far, with possible damage to the crownof the rail as well. In addition, one or more slidingwheels during braking can increase the distanceneeded to stop a train because the coefficient offriction between a sliding wheel and rail is lowerthan that between a rolling wheel and rail.

Slip-slide control is traditionally considered partof the propulsion and braking system, but it has im-portant relationships with the train control system.For example, correction of sliding during brakingcan be obtained only by reducing the braking effort,which has obvious safety implications. Eitherthrough the design of the braking system or in con-junc t ion w i th t he t r a in p ro t ec t i on sys t em,assurances must be provided that operation (or mal-function) of the slip-slide control does not perma-nently prevent application of the brakes when abrake application is required.

TRAIN SUPERVISION

In general terms, train operation functions areconcerned with the movement of individual trains.Train protection acts as a restraint to prevent acci-dents for individual trains or between trains. Bynature, these two groups of train control functionsare tactical and localized, in that they deal withshort-range concerns for specific elements or placesin the transit system. In contrast, train supervisioncomprises a group of functions concerned with theoverall regulation of traffic and the operation of thetransit system as a whole. Thus, train supervisionfunctions are strategic, systemwide, and more long-range.

184

The

FIGURE A–3.-Conceptual Diagram of Service Brake Flare-Out Control

functions of train supervision are:

Schedule Design and Implementation

Route Assignment and Control

Dispatching

Performance Monitoring

Performance Modification

Alarms and Malfunction Recording

Recordkeeping

Schedule Design and Implementation

In most rail rapid transit systems, the functionsof schedule design and implementation are not con-nected on a real-time basis. Train schedules areevolved to meet the transit system’s objectives,whether they be minimization of operating cost,maximization of service, utilization of equipment,or whatever. Most train scheduling in such situa-tions is performed manually, with perhaps occa-sional assistance from a computer,

Train schedules do not change frequently. Oncea basic service pattern has been established, it mayremain unchanged for months or years. The pri-mary changes may be the addition of special trainsto provide extra service to special events. This typeof extra service is usually provided in off peak hoursand presents no major train control problem. Pro-viding special crews is likely to be the most difficultproblem here.

Where major changes of schedule or operationalprocedure are contemplated, it may be necessary toutilize computer simulations. NYCTA has beenusing such simulations for about a decade for ex-amining complex scheduling and routing problems.Simulations may also be used in the planning ofsystems. Where systems are computer controlled,provision may be made to use the computer forsimulations of possible operational changes.

Operational implementation of the schedule isgenerally focused in some central control facility.This facility may be simple or elaborate. Hierarchi-cal control structures may be utilized, The primary

185

functions of the control center are (1) receipt anddisplay of information on the status of the system,(z) decisionmaking regarding action to be taken,and (3) issuing commands for action. It should bekept in mind that supervisory functions and deci-sions may affect the safety of passengers, but trainsupervision cannot override train protection con-siderations.

Most control centers have functions beyond traincontrol alone. It is common to monitor and controlthe electric power systems and other critical ele-ments such as pumps and blowers in these facilities,Monitoring of station platforms, fare collectionareas, or parking lots may also be carried out bycentral control facility, Passenger service com-munications may be provided, as well as some ser-vice to the news media or the general public.

Route Assignment and Control

The supervisory system selects, assigns, and con-trols the routes to be followed by trains. Under nor-mal circumstances, the routes of a conventionaltransit system are fixed, Major delays must be in-volved before train rerouting is done. In a lineartwo-track system, the most ordinary form of rerout-ing is concerned with operating the system overonly one track until a problem on the other track iscleared. Except for the systems which use com-puters, alternating the direction of traffic flow is ac-complished under the control of humans at a con-trol center or tower. BART provides special com-puter programs which can generate the necessarycommands for single-track operation. WMATAplans a similar approach, This approach presumablycan lead to more efficient operation, both in termsof increased flow through the system and in thefreeing of the controllers to make other decisionsduring such emergencies.

In a few transit systems, there are opportunitiesto route conventional trains from one line toanother in case of major service disruptions,NYCTA, for example, can reroute trains on some ofthe main Manhattan lines. Additionally, the four-track (two local, two express) arrangement of por-tions of the system permits interchange of trainsbetween some tracks on the same route-always ata loss in overall performance.

Dispatching

Train dispatching is concerned with the makeupof train sets and the timing of their departure from

selected points in the system. In conventional railtransit systems, a written schedule is used to indi-cate the anticipated system needs for the day, bothby train size and time of departure. Dispatchingusually takes place from terminals or yards.

Most train dispatching in conventional systems isaccomplished through the use of preprogrammed dis-patch machines at terminals and entry points.These machines simply provide a visual indicationthat it is time for the train to depart. In a shortsystem such as PATCO, there is normally nofurther supervisory control of the motion of thetrain through the system. Operators are providedwith a timetable, and verbal communications areused if any problem arises.

Modification of the dispatching routine may beaccomplished under computer control in systemssuch as BART and WMATA. Manual or verbaloverride is used at all operating transit propertiesexcep t BART. Mod i f i ca t i on o f d i spa t ch ingschedules is required to compensate for variousdelays on the line.

Performance Monitoring

Train performance monitoring is closely alliedwith train dispatching and route assignment. Essen-tially, the purpose of this function is to smooth outirregularities in the flow of traffic. Methods of per-forming this function range from the very simple tothe very sophisticated.

Basically, there are two approaches to perform-ance monitoring and control, They may be ac-complished on an intermittent or a continuousbasis. In all conventional systems in operation andbeing planned, performance monitoring is done onan intermittent basis, Train running times betweenstations or terminals are measured and any controlactions deemed necessary are taken. There is nocontinuous monitoring of the performance of thetrain while it is running. (Verbal communicationscan ordinarily be used to provide an indication ofserious performance problems as soon as they oc-cur.)

If it is desirable to modify the performance(speed, running time, acceleration) of a train, com-mands for the performance modification are usuallytransmitted at selected wayside locations (typicallystations). Continuous performance modificationcommands are not provided. Again, verbal com-

186

munications may be used to transmit a command atany time.

Performance Modification

There are two basic ways of modifying train per-formance. Trains may be held at specific stations toprovide more uniform spacing. Either in conjunc-tion with this or as an alternative, the actual run-ning time of the train speed or acceleration rate canbe changed.

In the systems which have the least amount ofautomation, performance monitoring and control isessentially accomplished by humans. Supervisoryand onboard personnel monitor the state of thesystem. Information on significant perturbationsmay come in through model board displays or voicecommunications. Required performance modifica-tions may be indicated by voice transmission orreductions in speed commands.

One step upward in automation is the addition ofdispatching lights at certain stations. By use of suchlights, trains can be held in these stations to attemptto smooth out the flow of traffic.

At the highest levels of automation are thesystems which use computers to adjust perform-ance requirements continually so as to provideschedule adherence and/or uniform flow of trainsthrough the system. Both BART and WMATA havefacilities to monitor the performance of all trains inthe system and to compare the actual and desiredstatus of the system, Through rather elaborate con-trol procedures, computers then issue commands tomodify individual train performance in a way suchthat the system objectives are met.

Views vary on the value of automating the func-tion of train performance modification. At BARTand WMATA not only are the desired performancelevels calculated automatically, they are transmit-ted and implemented automatical ly as well .Baltimore is considering an evolutionary systemwhich could eventually incorporate a computer.Initial thoughts are that performance level com-mands would be displayed in stations and onboardoperators would set switches to establish perfor-mance levels for the ATO equipment. NYCTA,which plans eventual conversion to automatic trainoperation, does not now contemplate the use ofperformance level controls as such. It is planned tocontrol train spacing by dispatching at terminalsand selected stations as well as by verbal control.

Alarms and Malfunction Recording

Aside from the annunciation of delays in trainmotion, supervisory systems can be used to indicateother problems in carborne equipment. Fire, low airpressure, lights out, air-conditioner failure, motorfailure, and many other things are potential candi-dates for malfunction alarm. In traditional mannedsystems, information on the status of onboardequipment is not transmitted to the waysideautomatically. Annunciation of malfunctions isprovided by displays in the operator’s cab. The in-formation may be relayed immediately by voicetransmission if the problem is serious. Minorproblems may be reported at the end of a run or theend of the day.

In unmanned systems, there is a greater need forannunciation of malfunctions. Both the Sea-Tacand AIRTRANS systems have malfunction annun-ciation systems with displays in the central controlfacility. A hierarchy of malfunction conditions isdefined, and each group is treated in accordancewith the seriousness of the event involved.

It should be noted that the annunciation systemfor train supervision may be a subset of a largersystem which deals with the status of many types ofequipment throughout the system. This may in-clude pumps, blowers, electrical power distributionequipment, and so forth.

Recordkeeping

Supervisory equipment or personnel keeprecords for individual vehicles and the overall tran-sit system. By means of train and car identificationequipment, information is provided on the ac-cumulation of car miles and used to schedule main-tenance activity. If computational capabilities are apart of the supervisory system, additional functionsmay be performed. The computer may be used togenerate work orders or schedules for routine main-tenance. Spare parts ordering may be handled. Man-power utilization and payroll data may be proc-essed. Reliability and maintainability data may bederived. Special management reports also may begenerated.

Where malfunction communication equipmentis used, it speeds diagnosis of system faults. Ingeneral, it appears that in-service diagnosis andrepair of individual failures is significantly less im-portant than maintaining operation of the system asa whole. Where practical, attempts are made to con-

187

tinue to operate trains wenough to get them out

th failed equipment longof service. If this is not

possible, pushing a train or modular replacement ofelements is attempted, The use of significantamounts of diagnostic equipment appears more ap-propriate in maintenance shops than on in-serviceequipment,

COMMUNICATIONS

Communication functions are implicit in allother types of train control system activity, andvarious forms of communications (both verbal anddata) have been mentioned in connection with thedescription of train protection, operation, andsupervision functions. Table A–1 is a summary ofthe major types of communication performed byand within the train control system. Each is dis-cussed below, with emphasis on those that have notbeen treated previously.

Train Protection

Train protection communications are tradi-tionally separate from all others. Special precau-tions are taken to insure that signals from other cir-cuits or systems do not mix with train protectionsignals.

It appears that future ATP systems will all relyheavily on electrical or electronic transmissionsystems. Voice communications are a part of ATPbut, in general, play a minor role. Voice com-munications are largely used to transmit informa-tion regarding visual verification of a safety situa-tion or procedural instructions related to emergencyoperation.

Operational Control and System Status

These two functions are discussed togetherbecause system status provides the feedback infor-mation for operational control. Essentially, theseare the communications involved in the train super-vision function. Most routine status information istransmitted by electrical means and displayedvisually. In automated systems, much of the statusinformation is processed directly by computers.Visual displays may be provided routinely or on acall-up basis.

Status informationmotion of vehicles. Itplatforms, availabilityditions, and any of a

includes more than just thealso concerns conditions onof trains in yards, track con-hundred other things. Mal-

function alarms transmitted from carborne orwayside equipment provide status information oncertain equipment. The more highly automated thesystem, the greater the need for equipment statusinformation, but this does not necessarily mean in-creased automation of the means for transmittingequipment status information. If additional infor-mation is to be communicated, it can also be givenover a voice channel by the onboard attendant.

Operational commands may be transmitted byalmost any means, ranging from a printed timetableto electronic devices. There is a trend toward use ofelectrical or electronic devices for signal transmis-sion in the new systems being planned, but thereare some specific exceptions. NYCTA, for example,plans to operate a hierarchical supervisory systemin which major decisions will be made in a com-mand center and relayed by voice or teletypewriterto dispersed towers for execution, Baltimore is con-sidering visual transmission of performance level

TABLE A–1.—Primary Means of Communicating Information Related to Train Control

FunctionPrimary Means of Transmission (T’) and Display (D)

Visual Voice Electric Signal Written

Train Protection T, D — T, D —

Operational Control I D T T, D —

System Status I D T T, D —

Emergency Communication I — T — —

Passenger Service — T — —

Maintenance Information — T T T, D

Business Operations — — T T, D

188

—

adjustments using displays on the wayside in sta-tions. Voice communication with all trains appearsto be an essential element in all new systems and inupgrading old ones, Emergency operational com-mands may thus be relayed by voice communica-tion.

Emergency Communications

Emergency communications are the only typethat may involve contact with outside agencies oremployees not on trains or reachable by telephone.For this and other reasons, at least one space radiosystem is provided in all transit systems to allowcommunication with roving employees.

It is felt by most transit system managers that theability to communicate with passengers, preferablytwo-way but at least one-way, is extremely impor-tant in controlling emergency situations. Assuringthe passenger that his plight is known and help is onthe way is believed to have considerable psy-chological value.

Emergency communications may also involvedealing with police and fire departments as well asother organizations of the civil government, Thesecommunications may be handled either by radio ortelephone.

The human role in emergency communicationsis very important for the simple reason that thenature of emergencies is such that unexpectedevents occur. Because humans respond to a verywide range of situations, it seems unlikely that theemergency communication role of the human canbe replaced.

Passenger Service

Train control equipment or personnel act to pro-vide information to the passengers. In mostsystems, onboard operators or station personnelprovide information on station identity and traindestination. ATS equipment is used to performsome of these functions in highly automated

systems. At BART, for example, special destinationsigns indicate the imminent arrival of trains, the ap-proximate location at which the trains will stop, andthe destination of the trains. These particular signsare also used to display commercial messages andthereby produce revenue for the system, BothAIRTRANS and Sea-Tac ut i l ize prerecordedmessages in the trains to provide information topassengers. AIRTRANS has both TV displays andlighted signs to display route information at the sta-tions,

Maintenance Information

Elements of the train control system maybe usedto provide information needed for scheduled orunscheduled maintenance. Train and car identifica-tion systems can be used to provide information onaccumulated car miles. In highly automatedsystems, malfunction detectors and annunciatorstransmit malfunction information either directly tothe maintenance facility or through central controlto the maintenance facility. Voice communicationsrelating to maintenance problems may be chan-neled through central control or handled directly.

Communication of maintenance information re-lated to inventory control may also be handled overthe ATC communication system, especially if acentral computer is used. This maybe accomplishedover commercial telephone lines, or special datatransmission links. Except for fully automatedsystems, there does not appear to be a trend towardsignificant increases in ATC communications formaintenance information transmission.

Business Operations

Basic data available from the train control systemmay be used in planning business operationsregarding workforce allocation, expansion plans,procurement policies, vehicle utilization, and so on,This information is generally presented in the formof tabulated reports which may be computer print-outs or periodic manual summaries of system per-formance parameters.

189

Appendix B

AUTOMATIC TRAIN CONTROL TECHNOLOGY”

train controland engineer-

The technology of automaticembraces many kinds of equipmenting techniques. All aspects of this technology can-not be adequately presented in a brief appendixsuch as this. Therefore, the discussion is confined totwo major elements of train control technology: thetrack circuit and methods for speed command andcontrol. The technology forms the basis for almostall automation of train protection, train operation,and train supervision functions. Of the two, thetrack circuit is the more basic. It was the first to bedeveloped, and it underlies the operation of speedcommand and control systems. It is the fundamen-tal method of train detection and, while there hasbeen experimentation with other methods over theyears, none has proven to be as effective and relia-ble as this electrical technique for determining thepresence and location of transit vehicles. From thisbasic positional information, signal systems areoperated, train protection is accomplished, trainoperation is controlled, and supervisory functionsare carried out.

TRACK CIRCUITS

The track circuit is an electrical circuit which in-cludes a length of running rails (or special rails) andpermits detection of the presence of a train. A trackcircuit may also be used to communicate com-mands, instructions, or indications between thewayside and a train. Track circuits provide informa-tion on the location of the trains, and this informa-tion is used to command train speeds so that thetrains operate safely. For instance, if a train at-tempts to approach too close to the rear of anothertrain, information on the locations of the two trains,provided by the track circuits, is used to command a

slowdown or stop of the following train beforethere is danger of a rear-end collision.

The basic d.c. track circuit was invented by Dr.William Robinson and first used in a railway ap-plication is 1872. Although the equipment and tech-nology have changed considerably in their detailsince that time, the basic principle has remained thesame. An electrical signal of some kind is impressedbetween the two running rails, and the presence ofa train is detected by the electrical connection be-tween the two running rails provided by the wheelsand axles of the train (wheel-to-rail shunting).98

Before proceeding to a discussion of the varioustypes of track circuits, it must first be consideredhow track circuits are used in the operation of atransit system. A track circuit provides informationon whether a train occupies a given length of track(a block). The occupancy information for a particu-lar block and for contiguous upstream of blocks isused to control the operation of all trains within thegiven area, For instance, when a train is detected ina block, that occupancy information is used to causea zero-speed command for the block immediatelybehind the train. Depending upon the block lengths,the line speeds involved, and the number of availa-ble speed commands, the second block behind thetrain may have a command speed between zero andfull line speed. The third block behind the train mayhave a commanded speed greater than or equal tothe second block, and so on. In all cases, the blocksbehind a train are signaled so that a train entering ablock has sufficient braking distance to leave theblock at a speed not greater than the commandedspeed. In the case of a zero-speed command, the

QTThiS appendix is based on material originally prepared byBattelle Columbus Laboratories in support of the OTA study.The editors gratefully acknowledge the contribution of BattelleColumbus Laboratories but accept full responsibility for the ver-sion presented here and for any alteration of content that mayhave been introduced in condensing and editing the material forpublication,

~In tran5it systems with rubber-tired vehicles, special railsare mounted beside the guideway and brushes or shoes on thevehicle contact these rails as the vehicle moves along. Thespecial rails replace the running rails of a conventional steel-wheel, steel-rail system, and the brushes or shoes replace thewheels and axles of a train in the operation of the track circuits,This is only a difference in detail; the principle is the same as inconventional track circuits.

191

train must beof the block.

— — — — —

able to stop before it comes to the end

D.C. Track Circuits

In all track circuits an electrical signal of somekind is impressed between the running rails, andthe presence of a train is detected by the electricalconnection that the wheels and axles of the trainmake between the two running rails. In d.c. trackcircuits, the electrical signal is direct current,usually supplied by batteries, The detector for theelectrical signal is a relay.

Figure B-1 shows a simple d.c. track circuit. Thetrack circuit consists of a block or length of trackwhich is defined at each end by insulated joints inthe running rails. The insulated joints provideelectrical insulation between a given track circuitand the abutting tracks which comprise other trackcircuits. The signal source, in this case a battery, isconnected to the rails at one end of the track circuitwhile the receiver (a relay) is connected to the otherend, When no train is present, the track circuit issaid to be unoccupied, and the direct current sup-plied by the battery is transmitted by the runningrails to the relay and energizes it or “picks it up. ”When the relay is energized, the upper set of relaycontacts is connected causing the green signal lightto be turned on. When a train enters the track cir-cuit its wheels and axles connect the two runningrails together, shorting the battery and therebyreducing the current through the relay. This causesthe relay to “drop,” as shown by the dashed line inthe diagram. This action connects the bottom set ofrelay contacts, turning off the green signal light and

..

turning on the red light to indicate that the block isoccupied by a train. The resistor in series with thebattery protects the battery by limiting the currentthe battery must provide when a train is present,

The terms “pick up” and “drop” refer to the posi-tion of the special “fail-safe” relays used for traindetection. These relays are constructed fromspecifications approved by the Association ofAmerican Railroads and are designed so that theirnormally open “front” contacts will be closed onlywhen sufficient electrical energy is being suppliedto the coil, One or both of the normally open contactmembers are made of carbon or carbon impregnatedwith silver, which cannot be welded. The relays usegravity rather than spring return and are mountedvertically so that the relay armature, to which thecontacts are attached, is returned to the droppedposition when the current through the coil isreduced below some critical value. The failure rateof these relays for the mode in which the normallyopen contacts would be closed with no power ap-plied to the relay coil is so low that for all practicalpurposes it is considered to be zero.

The track circuit shown here has been simplifiedfor the purpose of illustration. In actual practice, therelay would have several sets of contacts connectedin combination with the contacts from other relaysin nearby track circuits to form logic circuits for thecontrol of the signaling devices (the red and greenlights). Even in the simple form shown in FigureB-1, however, it can be seen that the breaking of anyconductor or the loss of power in the circuit willcause either a red signal or no signal at all to be dis-played. A red or “dark” signal is always to be in-

1

192

terpreted as a command to stop. If, for instance, thegreen light burned out or the relay coil open-cir-cuited so that the relay could not be “picked up, ” itwould be impossible to have a “green” signal. Inthat case a train would be required to stop when itarrived at that signal. To put it another way, all sig-naling systems are designed so that a green signal(meaning proceed) is presented only when the trackcircuits provide positive information that it is safeto do SO.

The double-railinterference whenthe return for d.c.

d.c. track circuit is susceptible tothe running rails are also used aselectric propulsion current. For

this reason, d.c. circuits are not used in rail rapidtransit. 99 Single-rail d.c. track circuits could be used,but in fact all modern rail rapid transit systems usesome form of a.c. track circuit.

P o w e r - F r e q u e n c y A . C . T r a c k C i r c u i t s

The power-frequency a .c . t rack circui t isenergized by an alternating electrical current with afrequency in the range of 50 to 150 hertz.100 Exceptfor the type of current and apparatus used, the a.c.track circuit is similar in operation to the d.c. trackcircuit described above.

~he principal modern application of the double-rail d.c.track circuit is in railroads with diesel-powered locomotives,

l~his type of circuit is often called simply an a.c. track cir-cuit.

Figure B-2 shows a simple power-frequency a.c.track circuit. As with the d.c. circuit, the a.c. trackcircuit consists of a block or length of track which isdefined at each end by insulated joints in one orboth of the running rails. Figure B-2 shows a doublerail circuit with insulating joints in both rails. Thea.c. signal source (usually a transformer) is con-nected to the rails at one end of the track circuitwhile the receiver (a relay) is connected to the otherend, In addition to the signal source and thereceiver, the a.c. track circuit contains a pair of im-pedance bonds at each pair of insulated joints. Animpedance bond is a center-tapped inductancewhich is connected across the rails on both sides ofthe insulated joints. The center taps of the pair ofimpedance bonds are connected together as shown.The purpose of the impedance bonds is to providecontinuity between the track circuits for the d.c.propulsion power and to distribute the propulsioncurrent between the two running rails. The im-pedance bonds do this while still maintaining arelatively high impedance at the signaling frequen-cies between the two rails and between adjacenttrack circuits.

When no train is present, the alternating currentsupplied by the transformer at the left side of thediagram in Figure B-2 is transmitted by the runningrails to the relay and “picks it up. ” The energizedrelay turns on the green signal light, exactly as in ad.c. track circuit. The wheels and axles of a train en-

FIGURE B–2.—Simple Power-Frequency A.C. Track Circuit

193

— . —.

tering the track circuit connect the two runningrails together; and the current through the relay isreduced, causing the relay to “drop.” This connectsthe bottom set or relay contacts, turning off thegreen light and turning on the red light to show thatthe block is occupied. The resistor in series with thetransformer (at the left in the diagram) protects thetransformer by l imit ing the current that thetransformer must provide when a train is present.

High-Frequency A.C. Track Circuits

Some a.c. track circuits use a current that alter-nates at a frequency in the range of hundreds orthousands of hertz. Because this frequency rangecorresponds roughly to the spectrum of audiblesound, such circuits are sometimes called audiofre-quency track circuits.

High-frequency a.c. track circuits eliminate theneed for insulated joints in the running rails.Because insulated joints are expensive to install andto maintain, eliminating them leads to a significantcost reduction. Eliminating insulated joints alsoallows the track circuit to operate with the con-tinuous welded rails being used in some newer in-stallations.

Figure B-3 shows a simple high-frequency a.c.track circuit. Since no insulated joints are used inthe running rails, the ends of the block establishedby special transformers are connected to the rails.

The transformer winding attached to the rails isusually a single turn of heavy copper bar stock. Thetransformer core is often a toroid. The othertransformer winding is tuned to resonate at theoperating frequency by a capacitor. The transmitteris the a.c. signal source and provides electricalenergy at the operating frequency in the audiofre-quency range. The receiver in this case is notsimply a relay, as with the d.c. and power-frequen-cy a.c. track circuits, but an electronic circuit whichresponds to the electrical signal provided by thetransmitter. The receiver may be used to actuate arelay which performs functions like those in the d.c.or power-frequency a.c. track circuits. Thus, whenno train is present the high-frequency a.c. potentialsupplied by the transmitter is connected to the run-ning rails by the transformer and transmitted alongthe running rails to the other transformer and itsassociated receiver. When the receiver detects thehigh-frequency a.c. signal, the relay is energizedand the green signal light is turned on. When a trainenters the block, the circuit behaves much as itwould with the a.c. or power-frequency a.c. trackcircuits, That is, the train wheels and axles connectthe two running rails together, and the current tothe receiving transformer and i ts associatedreceiver causes the track circuit relay to drop, turn-ing off the green light and turning on the red light.

The circuit illustrated in Figure B-3 is highlysimplified. In practice, it is necessary to accommo-

194

—

date the adjacentRather than install

—.

track circuitstwo separate

on either side,transformers for

each track circuit, a second resonant winding can beincluded in each transformer or a heavy primarywinding can be passed through more than onetransformer core. Thus, a single transformer assem-bly is used at the boundary between adjacent trackcircuits and serves each. Although part of the sametransformer assembly, the resonant windings areeffectively isolated from each other because theyare tuned to and operate on different frequencies.

Figure B-4 shows another type of high-frequencya.c. track circuit. No insulated joints are used in therunning rails. The ends of the block or track circuitare established by shunts which are heavy coppercables or bars attached to the rails, The transmitteris the a.c. signal source and supplies electricalenergy to a loop which is placed between the railsas shown. The loop is the primary of a transformerof which the rails and the shunts form the second-ary, At the other end of the track circuit, a pickupdirects the high-frequency a.c. energy to thereceiver, which in turn actuates the track circuitrelay. When no train is in the track circuit, the high-frequency a.c. potential supplied by the transmitteris directed into the loop and thence into the runningrails to the pickup associated with the track shuntshown at the right portion of the diagram, When thereceiver detects the high-frequency a.c. signal, the

relay is energized and the green light is turned on.When a train enters the track circuit, the wheelsand axles connect the two running rails together,and the current to the receiver is reduced. Thereduced current to the receiver causes the track cir-cuit relay to drop, turning off the green light andturning on the red light, as in other types of trackcircuits. The relay in a practical circuit would haveseveral sets of contacts which would be connectedin combination with the contacts from relays innearby track circuits to form logic circuits for thecontrol of the signaling devices.

In practice a transmitting loop and a pickup areassociated with each track shunt. Adjacent trackcircuits are operated on different frequencies, andthe receivers have frequency selectivity so theyonly respond to their intended frequencies. This isthe type of track circuit used in the BART system.

C h e c k - I n / C h e c k - O u t C i r c u i t s

All of the track circuits described up to this pointoperate on the closed-circuit principle. Any disrup-tion of the circuit by a train passing along the railsor by power or component failure, “opens” the cir-cuit and causes a red (stop) indication to be dis-played by the signal system.

An alternate approach to track circuit design uses“check-in/check-out” logic, Simply stated, this cir-

A.C. Track Circuit

195

— -

cuit is based on the principle that once a train isdetected or “checked in” to a block, it is assumed tobe there until it is “checked out” by being detectedin an adjacent block. The presence of a train may bedetected only intermittently at the time when it en-ters a new block. This is in contrast to the conven-tional track circuits described above in which thepresence of a train is detected continuously. Insome check-in/check-out systems the first and thelast cars of a train are checked in and out of theblocks as the train moves through the system, Atransmitter of some kind located on the train can beused with a receiver at a fixed wayside location toindicate that a train has entered the block associatedwith the wayside receiver, In some cases, twotransmitters are used, one at the head end of a trainand the other at the rear. When the head-endtransmitter enters a new block and is checked in,the block remains in the occupied condition untilthe rear-end transmitter also indicates that the rearend of the train has entered the new block, At thattime, the train is checked out of the block behind.

Check-in/check-out has some operational disad-vantages, For instance, consider the effects of atemporary loss of power to the signal system. Withconventional track circuits, the loss of signal powerwill cause all track circuits to indicate occupancy,but when the signal power is restored, the true oc-cupancy situation is again indicated. With a check-in/check-out system, the loss of signal power maydestroy the “memory” circui ts charged with“remembering” that a train has entered a block. Thememory often consists of electrical relays which areenergized (or deenergized) to indicate the presenceof a train in a block. The loss of electrical power candestroy the information stored in such a memory.(A memory whose information can be lost by a lossof electrical power is termed “volatile.”) Thus,when the signal power is restored, the informationon track circuits which are occupied may have beenlost. In this case the identity and location of eachtrain in the affected portion of the system must beestablished before the entire transit system can beoperated again safely. In a small transit system theidentification and location of each train may not bedifficult to establish. However, in a large, complexsystem even a short-term interruption of a portionof the system can create a bottleneck which makesit very difficult to restore the system to full opera-tion. Thus, check-in/check-out systems do not findapplication as the primary train detection system inrail rapid transit systems.

A special case of a check-in/check-out system isthe SOR (sequential occupancy release) system re-cently installed at BART, which uses the check-in/check-out principle as a logical back-up to the pri-mary train detection system which uses high-fre-quency a.c. track circuits. The purpose of SOR is toprotect against the loss of train detection in theevent the primary system fails and to prevent serv-ice interruptions due to false occupancy indications,

The SOR system provides a latch such that an oc-cupied track circuit continues to indicate occupancyuntil it is reset by the detected occupancy of the sec-ond downstream track circuit. Thus, with the loss ofshunt or failure to detect the presence of a train, thelatched-up track circuit still indicates occupancyand prevents a following train from colliding withthe rear of the leading train. A series of computersis used in the SOR system, and the logic is such thatthe computers can recognize false occupancies, i.e.,a track circuit which shows occupancy without aprior occupancy of the preceding track circuit isconsidered by the computer to be falsely occupied,

SPEED COMMAND AND CONTROL

In considering how the speed of transit vehiclesis controlled by automatic devices, it is important tounderstand the principle of closed-loop controlbefore proceeding to a discussion of the means bywhich speed commands are t ransmit ted andreceived, A closed-loop control system is one inwhich some feedback of information on the statusof the system (or its response to command inputs) isused to modify the control of the system. As aminimum, feedback verifies that the command wasreceived, Feedback may also be used to modulatesubsequent command inputs so as to smooth out ir-regularities of response, to make increasingly moreprecise adjustments of the state of the system, or tocompensate for external perturbations. Thus, thebasic purpose of closed-loop control is to assurecontinuity of control by confirming that commandinputs have been received and that the commandedstate of the system has, in fact, been achieved.

The alternative to closed-loop control is open-loop control, where commands are transmittedfrom the controlling to the controlled element with-out any feedback or acknowledgment that the com-mand signal has been received and interpreted pro-perly. The traditional wayside signaling of rail rapid

196

transit is an open-loop system. So, too, is a manuallyoperated train with cab signals, although theautomatic overspeed and stop enforcing mechan-isms of cab signals represent the beginning of aclosed-loop system. Systems with ATO are trueclosed-loop systems. Feedback is used to monitorthe response to propulsion and braking commandsand to regulate the performance of the system on acontinuous, real-time basis. Thus, a closed-loopsystem, in contrast to an open-loop system, ischaracterized by continuous control and self-adjust-ing commands conditioned by observation ofsystem response.

The technology for controlling the speed of tran-sit vehicles is based on the track circuit. The signalsused for train detection can also be used for thetransmission of speed commands to wayside signal-ing devices and to the trains. Two general methodsare used for the transmission of such commands. Inone method, the track circuit signal is turned on andoff at a specific rate, which is interpreted as a speedcommand. This rate modulation scheme is called acoded track circuit. The second method is calledbinary message coding. With either method, equip-ment on the wayside or on the train senses the sig-nals in the rails and decodes the speed command.

Coded Track Circuits

This technique is applicable to either d.c. or a.c.track circuits. The track circuit signal is switched onand off (modulated) at a rate which is related to thespeed command. The switching rates are in therange from about 50 to 500 times per minute. In ad.c. track circuit, the direct current applied to therunning rails at one end of the track circuit is simplyturned on and off at the desired rate. Waysideequipment at the other end of the track circuitreceives and decodes the signals. A code-followingtrack relay is used in the track circuit and codescontinuously when the circuit is not occupied. Therelay is energized when the current is allowed toflow and is deenergized or “drops” when the cur-rent stops. The decoding equipment is actuated bythe contacts of the code-following relay. When agiven code (rate of transmission) is received, a par-ticular relay in the decoding equipment is energizedand remains energized as long as that code is beingreceived. The relay, in turn, controls the appropri-ate wayside signal. When another code is received,another relay is energized as long as that code isbeing received. When a train enters the track cir-

cuit, the code-following relay is deenergized, andthis fact is used to indicate thee presence of a train.Typical interruption rates for these circuits are 75,120, and 180 times per minute,

In a.c. track circuits, either power-frequency oraudiofrequency, the a.c. signal is turned on and offat a selected rate. Since the switching rates for thecoded signals are so much slower (I-3 per second)than the frequencies of the a.c. signals applied tothe track circuit (50-150 per second), many cycles ofthe a.c. signal occur during the time that the codesignal is switched on. The coded track signal can bereceived by wayside equipment at the far end of thetrack circuit and used to control wayside signals orit can be received on board a train and used to con-trol the speed of the train. The presence of a trainstops the operation of the code-following relay andindicates occupancy of the track circuit. The codedtrack signals are received on board a train by a pairof coils mounted near the front of the leading carjust a few inches above each of the two runningrails and in front of the first set of wheels and axle,The magnetic field from the electric current carriedin the rails produces a signal in these coils (some-times called antennas), and this signal is processedor decoded to determine the switching rate andhence the speed command. The decoded speedcommand is used in an automatic system to controlthe speed of the train. In a semiautomatic system,the decoded speed command is displayed to thetrain operator who then regulates train speedmanually.

Binary Coded Track Circuits

This technique is sometimes used with audio-frequency a.c. track circuits. It could also be usedwith power-frequency a.c. circuits, but customarilyit is not. Instead of turning the track circuit signal onand off (rate modulation), the frequency of thetrack signal is changed from one to the other of twodiscrete frequencies. This technique is particularlyadaptable to digital systems in which one frequencycorresponds to the transmission of a “l” and theother frequency corresponds to the transmission ofa “0.” The track circuit receiver responds to both ofthe signaling frequencies that are used. When atrain enters the track circuit, the amplitude of thesignals at the track circuit receiver is reduced belowsome threshold and this information is used as anindication of the presence of the train. On board thetrain two antennas or coils are mounted near the

197

front of the lead car close to the running rails and in mine the speed command. In an automatic system,front of the first set of wheels and axle. As with the the decoded speed command is used to control thecoded track circuits, the magnetic field from the train speed, In a semiautomatic system, the decodedelectric current carried in the rails produces a signal speed code information is displayed to the trainin these coils, and this signal is decoded to deter- operator who controls train speed manually,

198

Appendix C

DESIGN CHARACTERISTICS

OF SELECTED RAIL RAPID TRANSIT SYSTEMS

This appendix is a tabulation of the ATC designcharacteristics and engineering features of fiveoperating rail rapid transit systems:

Bay Area Rapid Transit (BART)

Chicago Transit Authority (CTA)

Massachusetts Bay Transportation Authority(MBTA)

New York City Transit

Port Authority Transit

Authority (NYCTA)

Corporation (PATCO)

Listed vertically at the left of the tables are thegeneric functions which must be accomplished toprovide train protection, train operation, trainsupervision, and system communications. 101 A r -rayed beside these functions are descriptions of theequipment and techniques employed in the fivetransit systems. The major distinction is betweenmanual and automated techniques, with supple-mentary material to indicate specific engineeringand operational features.

None of the rail rapid transit systems describedhere is completely manual or completely automatic.

Iolsee appendix A for a definition and description of thesefunctions.

All represent various combinations of manual andautomatic train control—the particular mixturebeing determined by local needs and conditions, thehistory of engineering development in each locale,and (for the newer systems, at least) the fundamen-tal design philosophy. Generally speaking, NYCTAand CTA are the least automated of the five transitsystems, although both have a considerable amountof automation of train protection functions. TheRed Line of MBTA represents a higher level ofautomation, incorporating some automatic trainoperation features in addition to basic automatictrain protection. The other MBTA l ines areequivalent to NYCTA or CTA in the extent ofautomation, PATCO is still more automated, withvirtually all train protection and operation func-tions assigned to machine components. On theother hand, PATCO has almost completely manualmeans of train supervision. BART is the most high-ly automated of the five systems. Train protectionand operation are fully automatic, but monitored byan onboard operator. Train supervision is alsolargely automated, with extensive use made ofcentral computers to accomplish functions that areperformed by dispatchers and towermen in othertransit systems. The order of listing in the table in-dicates progressive increase in the general level oftrain control system automation.

199

—- ——————

TRAIN CONTROL SYSTEM CHARACTERISTICS

TRAIN PROTECTION FUNCTIONS NYCTA CTA

Train Detection

Monitoring of track occupancy

Train Separation

Collision prevention, primarily byblocks to ensure safe separation

and speed limits to ensure safestopping distance

Movement Commands

Speed and stopping commands totrains

Overspeed Protection

Comparison of command andactual speed to ensure that civilspeed limits are not violated

Speed Determination

Sensing and display of actualtrain speed

Interlocking

Prevention of conflicting trainmovement through switches andalong routes

Train and Track Surveillance

Monitoring the right-of-way forobstructions, persons on track,broken rails, etc.

Monitoring condition of train

systems.

Conventional voltage level trackcircuits of two types:

single rail, power frequency,hardwireddouble rail, power frequency,hardwired

Fixed blocks (length: 40-1200 ft.)Relay logicMinimum design headway: 11/2 min.

Conventional voltage level trackcircuit of three types:

single rail, power frequency,hardwireddouble rail, power frequency,hardwireddouble rail, audio frequency,hardwired

Fixed blocks (length: 300-2000 ft.)Relay logicMinimum design headway: 11/2 min.

Wayside signals Mixture of wayside and cab signalsThree-aspect block signal system Three-aspect block signal system(Proposed cab signal system will Five cab signal speed commandshave 70,50, 35, 25, 15, and 3 (70, 35,25, 15,0 mph)mph speed commands and a cabsignal cutout. Absence of apositive command is interpretedas O mph.)

Wayside signals with timers and Mixture of cab signals withtrip stops automatic overspeed protection

and wayside signals with timers andtrip stops

Estimated by motorman, no speed- Tachometer, with speedometer inometer in cab except on new R 44 caband R-46 cars.

Mixture of electro-mechanical and Mixture of mechanical, electro -all-relay mechanical, electro-pneumatic,

and all-relay

Visual track surveillance (primarily Visual track surveillance (primarilyby motorman with some assistance by motorman with some assistancefrom conductor) from conductor); cab signal for

broken rail protection

Status of some carborne systems is Status of some carborne systems ismonitored automatically and monitored automatically and dis-displayed by annunciator placards in played by annunciator placards inthe cab. the cab.

200

MBTA PATCO BART

Conventional voltage level trackcircuits of three types:

single rail, power frequency,h a r d w i r e d double rail, power frequency,hardwireddouble rail, audio frequency,hardwired

Fixed blocks (length: 425-2100 ft.)Relay logicMinimum design headway: 11/2 min.

Mixture of wayside and cab signalsThree-aspect block signal systemEight cab signal speed commands(70, 65,50,40,25, 10,0 mph andyard speed)

Conventional voltage level trackcircuits of two types:

single rail, power frequency,hardwired (yard only)double rail, power frequency,hardwired (revenue tracks)

Fixed blocks (length: 295-3400 ft.)Relay logicMinimum design headway: 2 min.

Cab signalsFive speed commands (75, 40,30,20, 0 mph) to ATO system whichcontrols speed

Mixture of cab signals with automatic Cab signals with automaticoverspeed protection and wayside overspeed protectionsignals with timers and trip stops

Low voltage level track circuits:double rail, audio frequency,multiplex one rail; powerfrequency track circuits in yards

Fixed blocks (length 75-1100 ft.)Solid-state logicMinimum design headway: 11/2 min.

(With sequential Occupancy Releasesystem, headways are restricted to

2 min.)

Cab signalsEight speed commands (80, 70,50,36, 27, 18,6, 0 mph) to ATO systemwhich controls speed

Cab signals with automaticoverspeed protection

Estimated by motorman, no speed- Tachometer, with speedometer in Tachometer, with speedometer inometer in cab except for Silverbird cab cabcars on Red Line

Mixture of electro-mechanical and All-relayall-relay

All -relay

Visual surveillance (primarily by Visual surveillance by train operator Visual surveillance by train operatormotorman with some assistance cab signal for broken rail protection cab signal for broken rail protectionfrom train guards); cab signal forbroken rail protection (Red Lineonly)

Status of some carborne systems is Status of some carborne systems is Status of some carborne systems is

monitored automatically and dis- monitored automatically and dis- monitored automatically and dis-played by annunciator placards in played by annunciator placards in played by annunciator lights in the

the cab. the cab. cab.

201

TRAIN CONTROL SYSTEM CHARACTERISTICS–Continued

TRAIN SUPERVISION FUNCTIONS NYCTA CTA

Schedule Design and Implementation

Planning of service in light ofanticipated demand, availableequipment, and environmentalconditions (includes orders toexecute the plan)

Train Identification

Manual Manual

Determination of the route anddestination of a train

Mixture of manual and automatic

(Automatic only on R-44 and R-46cars, where passive unit on trainresonates when excited by waysideequipment )

Mixture of manual (by trainoperator or towerman) andautomatic (passive unit ontrain resonates when excitedby wayside equipment or opticalscanning of identity panel ontrain by wayside equipment)

Train Dispatching

Control of train departures fromterminals (or waypoints) inaccordance with schedule

Route Assignment and Control

Mixture of manual and automatic(electro-mechanical clock)

Mixture of manual and automatic(electro-mechanical clock)

Selection and assignment of routesto be followed by trains, includingperiodic update reports by trainsas to identity, location, anddestination

Mixture of manual control by localtowerman and automatic controlbased on train identity information ortrack circuit occupancy

Mixture of manual methods (bycentral control remotely or bytowerman locally) and automaticcontrol based on train identityinformation or track circuitoccupancy

Performance Monitoring

Following the progress of trainsagainst the schedule

Visual observation (model boardsin towers and central control)Also manual check-off at towers

Visual observation (model boardsin towers and pen graph recordersat central control)

Performance Modification

Adjustment of movement com-mands or revision of schedulein response to traffic conditions

Verbal instructions to adjust runningspeed or station stops; remotelycontrolled starting signals to delaydeparture from terminals and controlpoints

Verbal instructions to adjust runningspeed or station stops; remotelycontrolled starting signals to delaydeparture from terminals and controlpoints

Alarms and Malfunctions Recording

Alerting to malfunctions andproblems and recording of time,location, and nature

Yard Train Control

Manually activated electrical alarm-ing and manual recording based onverbal reports

Manual alarming and recording basedon verbal reports

Train assembly, routing, andmovement with in yards and toand from revenue tracks

Manual train operation and a mix-ture of manual and automatedswitching

Manual

202

MBTA PATCO BART

Manual Manual Schedule prepared manually and fedinto central computer

Mixture of manual and automatic Automatic (passive unit on train Automatic (active unit on train

(optical scanning of identity panel resonates when excited by wayside transmits identity to wayside

on train by wayside equipment) equipment) transceiver for relay to terminals,stations, and central control)

Mixture of manual and automatic Automatic (electro-mechanical clock) Automatic (computer controlled)

(electro-mechanical clock)

Mixture of manual methods (by

central control remotely or bytowerman locally) and automaticcontrol based on train identityinformation or track circuitoccupancy.

Visual observation (model boardsin towers and central control)

Mixture of manual control by Automatic (trainborne destinationcentral control remotely and information transmitted to waysideautomatic control based on train equipment which automatically setsidentity information or track route); manual control (by centralcircuit occupancy or local controllers) available as an

alternative or back-up mode

Visual observation (model board at Visual observation (model boards atcentral control) central control and towers) with

computer-aided display and alerting

Verbal instructions to adjust running Verbal instructions to adjust running Automatic, station dwell time andspeed or station stops; remotely speed or station stops; remotely train performance mode (speed and/orcontrolled starting signals to delay controlled starting signals to delay acceleration) controlled by centraldeparture from terminals and control departure from terminals and control computer (can be selected by computerpoints points automatically or by manual input)

Manual alarming and recording based Manual alarming and recording based Automatic alarming and recordingon verbal reports on verbal reports for some events; manual inputs to

computer record also possible

Manual Manual Manual (special hostling controlpanel)

203

.———

TRAIN CONTROL SYSTEM CHARACTERISTICS–Continued

TRAIN OPERATION FUNCTIONS NYCTA CTA

Velocity Regulation

Control of actual speed inrelation to command (civil)

speed

Station Stopping

Stopping train in alignmentwith station platform

Door Control

Opening and closing of doorsat stations

Train Starting

Departure from station

Manual

Manual

Manual

Manual

Manual, by conductor Manual, by conductor (oroperator on single-car trains)

Manual, by operation of propulsion Manual, by operation of propulsioncontrol control

COMMUNICATION SYSTEMS NYCTA CTA

Train – Central Radio

Train – Station No direct link, relayed throughcentral control

Train – Wayside Radio

Central - Station Telephone; also public addresssystem on platform at some stationsand automatic train departuresigns at some terminals

Radio and dial telephoneCentral – Wayside

Station – Wayside

Station – Station

Wayside - Wayside

Outside Emergency Assistance

*

Dial telephone

Dial telephone

Radio and dial telephone

Walkie-talkie radio net for police,central control and key dispatchers,other assistance summoned throughcentral control

Train phone

No direct link, relayed throughcentral control

No direct link, relayed throughcentral control

Telephone; also public addresssystem on platform at some stationsand automatic train departuresigns at some terminals

Dial telephone; also public addressto certain key towers and terminalsupervisors

Dial telephone

Dial telephone

Dial telephone

Dial or direct-line telephone

204

MBTA PATCO BART

Automatic on Red Line, with manualoperation (at full speed) available asalternative mode.Manual on other lines

Manual

Manual by train guard (conductor)

Manual, by operation of propulsioncontrol

Automatic, with manual operation(at, full speed) available as alternativemode.

AutomaticStop command triggered whentrain passes fixed wayside point;braking effort to stop in requireddistance reckoned from wheelrevolution

Manual by train guard (motorman)

Manual, by depressing start button

Automatic, with manual operation(at reduced speed) available as aback-up mode or if track conditionsdictate

AutomaticContinuous stop command generatedby wayside equipment; brakingeffort to stop in required distancereckoned from wayside measuringpoints.

Automatic, with manual override

Automatic

MBTA PATCO BART

Radio

No direct link, relayed throughcentral control

Radio

Telephone and public addresssystem on station platforms; somestartees equipped with walkie-talkieradios

Radio and dial telephone

Dial telephone

Dial telephone

Dial telephone

Police and fire each on separateradio network; utilities contactedby telephone

Train phone

No direct link, relayed throughcentral control

No direct link, relayed throughcentral control

Telephone, public address, closed-circuit TV, and call-for-aid phonesat automatic fare collection gates

Dial telephone, radio in trucksand work trains, walkie-talkiefor trackside workers

Dial telephone

Dial telephone

Radio and dial telephone

PATCO police on system radionetwork; outside police on separatenetwork or contacted by telephone;fire and utilities contacted bytelephone

Radio

No direct link, relayed throughcentral control

Radio

Telephone and public address system;automatic signs on station platformsindicating train arrival and destination

Radio and dial telephone

Dial telephone

Dial telephone

Radio and dial telephone

BART police on system radio net-work; fire and utilities contacted byphone

205

Appendix D

GLOSSARY OF TRAIN CONTROL TERMS

The language of rail rapid transit and train con-trol technology contains many specialized termsthat may be unfamiliar to the general reader. Thisglossary has been prepared as an aid to understand-ing the terminology used in the report. It is also con-templated that the glossary may be useful as areference for additional reading on the subject ofATC and transit system engineering. For thisreason, the list of terms defined here has been ex-panded to include some background items notneeded for the immediate purpose of reading thisreport.

The principal source of the definitions presentedhere is the Lexicon of Rail Rapid Transit Safety-Related Terminology, prepared by the Safety Tech-nology Applied to Rapid Transit (START) Commit-tee of the American Public Transit Association,January 1975. The START Lexicon, in turn, drawsextensively on earlier work by the Association ofAmerican Railroads and the U.S. Department ofTransportation. In addition to START, othersources consulted include General Order No. 127 ofthe Public Utilities Commission of the State ofCalifornia, August 1967, and several technicalspecifications prepared by WMATA. In all cases,however, the responsibility for interpretation andfor the accuracy and completeness of the defini-tions offered here rests with the authors of thisreport.

ACKNOWLEDGING DEVICE—a manual deviceused by the train operator to forestall automaticbrake application on a train equipped withautomatic train stop or to silence the sounding ofa cab indicator on a train equipped with cab sig-naling. (See Audible Cab Indicator.)

ASPECT—the visual indication presented to an ap-proaching train by a wayside signal; also, the dis-play presented by a cab signal to an operator inthe cab. The aspect is said to be “clear” (proceedat civil speed) or varying degrees of “restrictive.”

False Clear Aspect—the aspect of a signal thatconveys an indication less restrictive thanintended.

False Restrictive Aspect—the aspect of a sig-nal that conveys an indication more restric-tive than intended.

ATTENDANT-a transit employee on board atrain in service whose principal duties are tooversee safety, provide security, and assist inemergency situations (as distinct from a trainoperator, motorman, who is responsible for run-ning the train).

AUDIBLE CAB INDICATOR-an alerting device,on a train equipped with cab signals, designed tosound when the cab signal changes and to con-t inue sounding unt i l acknowledged. (SeeAcknowledging Device.)

AUDIO-FREQUENCY TRACK CIRCUIT—a track

circuit energized by an electrical current alter-nating in the audio-frequency range(15,000-20,000 Hz); also called “high frequency”or “overlay” track circuit.

AUTOMATIC BLOCK SIGNAL SYSTEM—aseries of consecutive blocks governed by blocksignals, cab signals, or both, actuated by occupan-cy of the track or by certain conditions affectingthe use of a block; such as an open switch or a carstanding on a turnout and blocking the maintrack. (See also Block and Manual Block SignalSystem.)

A U T O M A T I C C A R I D E N T I F I C A T I O N — asystem that automatically provides positiverecognition and transmission of individual carnumbers as they pass a fixed wayside point.

AUTOMATIC TRAIN CONTROL—the method(and, by extension, the specific system) forautomatically controlling train movement, en-forcing train safety, and directing train opera-tions. ATC includes four major functions:

Automatic Train Protection (ATP)--assuringsafe train movement by a combination oftrain detection, separation of trains runningon the same track or over interlockedroutes, overspeed prevention, and route in-terlocking.

207

Automatic Train Operation (ATO)--controll-ing speed, programed station stopping, dooroperation, performance level modification,and other functions traditionally assignedto the train operator and conductor.

Au toma t i c T ra in Supe rv i s i on (ATS)—monitoring of system status and directingtraffic movement to maintain the scheduleor minimize the effect of delays.

Communication (CS)—interchanging infor-mation (voice, data, or video) betweensystem elements separated by distance.

AVAILABILITY—the portion of time that asystem is operating or ready for operation;mathematically, the probability that a system orsystem element will be operational when re-quired, expressed as the ratio of mean time be-tween failure to the sum of mean time betweenfai lure plus mean t ime to restore. [A =M T B F/ (M T B F + M T T R)] (See also Mean TimeBetween Failure and Mean Time to Restore.)

BASE PERIOD—the nonrush hour period of week-day transit system service. (See also PeakPeriod,)

BERTH—the space assigned for a train of specifiedlength when stopped at a station platform or in aterminal zone. (See Terminal Zone.)

BERTHING—the positioning of a train in itsassigned berth.

BLENDING—the automatic and simultaneous ap-plication of dynamic and friction braking, wherethe effort of each is continuously proportioned toachieve the required total braking effect.

BLOCK—a length of track of defined limits, the useof which is governed by block signals, cab sig-nals, or both.

Absolute Block-a block into which no train isallowed to enter while it is occupied byanother train.

Permissive Block--a block into which a trainis allowed to enter even though occupied byanother train.

BLOCK SIGNAL-See Signal.

BRAKE ASSURANCE—the function provided bya subsystem within the automatic train operationsystem that will cause the emergency brakes of avehicle to be applied when the actual braking

rate of the vehicle is less than the braking rate re-quested by the automatic train control system.

BRAKING—the process of retarding or stoppingtrain movement by any of various devices:

Dynamic Braking—a system of electricalbraking in which the traction motors areused as generators and convert the kineticenergy of the vehicle into electrical energy,which is consumed in resistors and, in sodoing, exert a retarding force on the vehi-cle.

Friction Braking—braking supplied by amechanical shoe or pad pressing against thewheels or other rotating surface; also called“mechanical braking. ”

Regenerative Braking-a form of electricalbraking in which the current generated bythe traction motor is returned to the trac-tion power supply for use in propellingother trains. (In ordinary dynamic brakingthe generated power is dissipated in resis-tors.)

There are two methods of controllingbrake application:

Closed-Loop Braking-continuous modula-tion (by means of feedback) under thedirection of the automatic train operationsystem or the human operator. (See Closed-Loop Principle,)

O p e n - L o o p B r a k i n g — b r a k i n g w i t h o u tmodulation through feedback from theATO system.

BRAKING EMERGENCY—irrevocable unmodu-lated (open-loop) braking to a stop usually at ahigher rate than that obtained with a full servicebrake application.

BRAKING, FULL SERVICE-a nonemergencybrake application that obtains the maximumbrake rate consistent with the design of the pri-mary brake system. Full service braking can bereleased and reapplied.

BRAKING, SERVICE-braking produced by theprimary train braking system,

CAB SIGNAL SYSTEM—a signal system wherebyblock condition and the prevailing civil speedcommands are transmitted and displayed directlywithin the train cab. The cab signal system may

208

be operated in conjunction with a system of fixedwayside signals or separately. (See also Signal.)

CATENARY—the wire or wires above the track(including the messenger, supports, and insula-tion) that carry electric energy for the propulsionof trains. (See also Contact Rail.)

CENTRAL CONTROL—the place from whichtrain supervision and direction is accomplishedfor the entire transit system; the train commandcenter.

CIRCUIT, TRACK-an arrangement of electricalequipment, including the rails of the track, thatforms a continuous electrical path used for thepurpose of detecting the presence of trains on therails; the track circuit may also be used to com-municate commands or other information be-tween the wayside and the train.

Check-In/Check-Out—a track circuit systemthat detects the entrance of the front end ofa train into a block and the departure of therear end of a train from a block for the pur-pose of determining block occupancy.

Coded Track Circuit—a track circuit in whichthe feed energy is varied or interruptedperiodically for the purpose of transmittingcommands or instructions to the train oroperating train detection apparatus.

Fail-Safe Circuit-a circuit designed to princi-ples which will cause the actuated device toassume its most restrictive position (or astate generally known to be safe) when anyelement of the circuit or system fails.

Vital Circuit--an electrical circuit that affectsthe safety of train operation.

CIVIL SPEED-See Speed Limit.

CLOSED-CIRCUIT PRINCIPLE—the principle ofcircuit design employing a circuit that is nor-mally energized and, on being deenergized or in-terrupted, causes the controlled function toassume its most restrictive condition.

CLOSED-LOOP” PRINCIPLE—the principle ofcontrol system design in which the response of asystem (feedback) is continuously comparedwith the controlling signal to generate an errorsignal.

CLOSING IN—running a following train toward aleading train that is either stopped or running

slower than the following train. (See also Closingup.)

CLOSING UP—running a following train to a posi-tion that will allow it to couple with a stoppedleading train.

COAST—the moving condition of a car or trainwhere the propulsion is inactive and, usually, acertain minimum braking is applied. (See alsoFreewheeling.)

CONDUCTOR-an attendant whose main func-tion is to operate train doors.

CONSIST (noun)—the number, type, and specificidentity of cars that compose a train.

CONTACT RAIL-a rail, mounted on insulatorsalongside the running rails, that provides electricenergy for the propulsion of trains. (Also knownas “Third Rail.”)

CROSSOVER—two turnouts, arranged to form acontinous passage between two parallel tracks.

DEADMAN CONTROL-a safety device that re-quires continuous pressure or activity to remainactivated; used to detect the inattention or dis-ability of a train operator.

DEPARTURE TEST-an operational test made ina yard or on a transfer track before permitting theunit to enter revenue service.

DISPATCH—to start a train into revenue servicefrom a terminal zone, transfer track, or desig-nated intermediate point.

DISPATCHER-a person at central control whosefunction is to dispatch trains, monitor trainoperation, and to intervene in the event ofschedule disruption or when any change in serv-ice or routing is required. (Also called “LineSupervisor” or “Central Supervisor.”)

DOWNSTREAM—for a given direction of travel,locations that will be reached after passing agiven point (equivalent to the AAR term “in ad-vance of”).

DWELL (or DWELL TIME)—the elapsed timefrom the instant a train stops moving in a stationuntil the instant it resumes moving,

ENTRANCE—EXIT ROUTE CONTROL—asystem of interlocking control that automaticallyalines switches and clears signals to form a trainroute in response to manual inputs designating

209

the entrance and exit points of the desired route.(Also called “N-X.”)

FACING MOVEMENT—the movement of a trainover points of a switch which face in the direc-tion in which the train is moving. (See also Trail-ing Movement.)

FAIL-SAFE--a characteristic of a system whichensures that a fault or malfunction of any ele-ment affecting safety will cause the system torevert to a state that is known to be safe; alter-natively, a system characteristic which ensuresthat any fault or malfunction will not result in anunsafe condition.

FALSE OCCUPANCY--an indication of track oc-cupancy when no train is present.

FREEWHEELING-a mode of operation in whichthe train is allowed to roll freely without tractiveor braking effort being applied. (See also Coast.)

FREQUENCY SHIFT KEYED (FSK)--a techniqueused with high-frequency a.c. track circuits, inwhich the frequency of the track signal is variedbetween two or more discrete states to conveyinformation (used as an alternative to ratemodulation where the track circuit is turned onand off as an information code).

FROG-a track structure, used at the intersectionof two running rails, to provide support forwheels and passageway for their flanges, thuspermitting wheels on either rail to cross theother. A frog may either be fixed or have mova-ble points like a switch.

GATE—the limit of an interlocked route where en-try to that route is governed by a signalingdevice. 102

Fixed Gate--the limit of an interlocked routebeyond which automatic operation of trainsis never permitted.

HEADWAY—the time separation between twotrains traveling in the same direction on the sametrack, measured from the instant the head end ofthe leading train passes a given reference pointuntil the head end of the train immediatelyfollowing passes the same reference point.

HOSTLER-an employee assigned to operate carsor trains manually within the yard or mainte-nance area.

Hz (HERTZ)—the unit of frequency equal to 1cycle per second.

IMPEDANCE BOND-a device of low resistanceand relatively high reactance, used to provide acontinuous path for the return of propulsion cur-rent around insulated joints and to confine alter-nating current signaling energy within a trackcircuit.

I N D U C T I V E L Y C O U P L E D I M P E D A N C EBOND--an impedance bond in which transmit-ter energy and receivers are inductively coupledinto a track circuit,

INSULATED JOINT-a joint placed between abut-ting rail ends to insulate them from each otherelectrically.

INTERLOCKING--an arrangement of signals andcontrol apparatus so interconnected that functionsmust succeed each other in a predetermined se-quence, thus permitting train movements alongroutes only if safe conditions exist.

Automatic Interlocking-an interlocking con-trolled by logic circuits so that movementssucceed each other in proper sequencewithout need for manual activation or con-trol.

Manual Interlocking—an interlocking oper-a t ed manua l ly f rom an i n t e r l ock ingm a c h i n e , s o i n t e r c o n n e c t e d ( e i t h e rmechanically or electrically) that move-ments succeed each other in proper sequen-cy.

Relay Interlocking-an interlocking in whichlocking is accomplished electrically by in-terconnection of relay circuits.

INTERLOCKING LIMITS—the length of track be-tween the most remote opposing home signals ofan interlocking.

INTERLOCKING MACHINE-an assemblage ofmanually operated levers or like devices for con-trolling the switches, signals, and other apparatusof an interlocking. (Also cal led “SwitchMachine,”)

INTERLOCKING ROUTE-a route between twoopposing interlocking signals.

JERK—the rate of change of acceleration (thesecond derivative of velocity), expressed in unitsof miles per hour per second per second(mphpsps, mph/see/see, or mph/sec2).

JUNCTION-a location where train routes con-verge or diverge.lo2Tht,sf, terms ;] rp pet; II 1 ia r to the BA RT s~st[’m.,.

210

KEY-BY—the act of lowering a trip stop in order topass a signal displaying a stop indication; socalled because of the use at one time of a key bythe train operator to actuate the mechanism forlowering the trip stop. Key-by today operatesautomatically without a key,

LOCKING-establishing an electrical or mechani-cal condition for a switch, interlocked route,speed limit, or automatic function such that itsstate cannot be altered except by a prescribedand inviolate sequence of actions.

Approach Locking-electric locking effectivewhile a train is approaching within aspecified distance a signal displaying anaspect to proceed and which prevents, untilafter the expiration of a predeterminedtime interval after such signal has beencaused to display its most restrictive aspect,the movement of any inter locked orelectrically locked switch, movable pointfrog or derail in the route governed by thesignal and which prevents an aspect to pro-ceed from being displayed for any conflict-ing route.

Electric Locking—an electrical circuit arrange-ment by means of which levers of an in-terlocking machine, switches, or other sig-nal apparatus is secured against operationunder prescribed conditions.

Indication Locking-electric locking whichprevents actions that would result in an un-safe condition for a train movement if a sig-nal, switch, or other operative unit fails tomake a movement corresponding to that ofits control.

Occupancy Detector Locking--electric lockingwhich prevents the movement of a trackswitch while the track circuit or circuitssurrounding that switch are occupied by atrain.

Route Locking—electric locking, effectivewhen a train passes a signal displaying anaspect for it to proceed, that prevents themovement of any switch in the routegoverned by the signal and prevents theclearing of a signal for any conflictingroute.

Time Locking-electric locking that preventsthe operation of any switch in the route (or

for any conflicting route) until expiration ofa predetermined time interval after a signalis restored to its most restrictive indication.

Traffic Locking--electric locking which pre-vents the actuation of devices for changingthe direction of traffic on a section of trackwhile that section is occupied or while asignal displays an aspect for a movement toproceed into that section,

Sectional Release Locking—a route locking soarranged that, as a train clears a section ofthe route, the locking affecting that sectionis released. (Also called “Trailing ReleaseLocking.”)

MAINTAINABILITY—the property of a systemthat allows it to be repaired and restored tooperating condition after a component malfunc-tion or failure; maintainability is often expressedas mean time to restore (or repair).

MANUAL BLOCK SIGNAL SYSTEM-a blocksignal system operated manually, usually basedon information transmitted by telephone ortelegraph.

MARRIED PAIR—two semipermanently coupledcars that share certain essential components andare usually operated as a unit.

MASTER CONTROLLER-a carborne device thatgenerates control signals to the propulsion andbraking systems,

MEAN TIME BETWEEN FAILURES (MTBF)—the average time that a system or component willoperate without fai1ure or ma1function;

m a t h e m a t i c a l y , M T B F = ( o p e r a t i n gtime) /(number of failures). MTBF is the measureof reliability.

MEAN TIME TO RESTORE (MTTR)—theaverage time required to restore a system or com-ponent to operation after a failure; this time ismeasured from the time troubleshooting andrepair work is begun until the system or compo-nen t i s aga in ope rab le ; ma thema t i ca l ly ,MTTR = (cumulative corrective maintenancet ime) / (number of fai lures) . MTTR is themeasure of maintainability,

MODEL BOARD-a reproduction of the trackassemblage (not necessarily to scale) equippedwith lights and other indicators, used for the pur-

21172-683 C) - 76 - 15

pose of train supervision and traffic control (Alsocalled “Train Board”).

MOTORMAN-See Operator.

MTBF-See Mean Time Between Failures.

MTTR-See Mean Time to Restore.

NORMAL DIRECTION—the prescribed directionof train traffic as specified by the rules; usually,the direction in which all regularly scheduledrevenue service operations are conducted.

N–X-See Entrance-Exit Route Control.

OPERATOR—the transit employee on board thetrain having direct and immediate control overthe movement of the train. (Also called “Motor-man.”)

OPPOSING TRAIN—a train moving in the direc-tion opposite to another train on the same track.

OVERSPEED CONTROL—that onboard portionof the carborne ATC system that enforces speedlimits in a fail-safe manner.

PABX—a des igna t i on u sed i n t he na t i ona ltelephone system to denote a privately ownedtelephone system that operates by the use of dial-ing, such as that used in some transit systems forcommunication between stations or waysidelocations and central control.

PEAK PERIOD—the period during a weekdaywhen system demand is highest ; usual ly7:30-9:30 a.m. and 4:30-6:30 p.m. (Also called“Rush Hour.”) (See also Base Period)

POINT-See Switch Point.

PROPERTY—literally, the right-of-way, track,structures, stations, and facilities owned or oper-ated by a transit agency; but used generally as asynonym for the operating agency itself. (Seealso Territory.)

RAIL RAPID TRANSIT—a mode of transporta-tion operating in a city or metropolitan area andhigh-speed speed passenger cars run singly or intrains on fixed guideways in separate rights-of-way from which all other vehicular and foottraffic is excluded. Tracks may be located in un-derground tunnels, on elevated structures, inopen cut, or at surface level. There are very few,if any, grade crossings; and rail traffic has theright-of-way at such intersections. Cars aredriven electrically with power drawn from an

212

overhead electric line by means of pantograph orfrom an electrified third rail. Rail rapid transitmay use steel wheels on steel rails or pneumatictires on wooden, steel, or concrete guideway.

RELAY-a device operated by variation in the con-dition of one electric circuit and used to effectthe operation of other devices in the same oranother circuit; commonly, an electromagneticdevice to achieve this function.

Track Relay--a relay receiving all or part ofits operating energy through conductorshaving the track rails as an essential part.

V i t a l Re lay—a re l ay , mee t ing ce r t a instringent specifications, designed so thatthe probability of its failing to return to theprescribed state after being deenergized isso low as to be considered, for all practicalpurposes, nonexistent.

RELIABILITY—the probability that a system orcomponent thereof will perform its specifiedfunction without failure and within prescribedlimits; reliability is often expressed as a meanfailure rate (MTBF).

REVENUE SERVICE—transportation of fare-paying passengers on main line routes.

REVERSE DIRECTION—train movement op-posite to the normal direction. (See NormalDirection.)

REVERSE RUNNING-operation of a train in thereverse direction.

ROUTE-a succession of contiguous blocks be-tween two controlled gates or interlocked sig-nals.

Conflicting Routes—two or more routes (op-posing, converging, or intersecting) overw h i c h m o v e m e n t s c a n n o t b e m a d esimultaneously without possibility of colli-sion.

Normal Route--a prescribed route, a route inthe normal direction of train travel.

Reverse Route--a route opposite to the normalroute.

ROUTE REQUEST—registration at an interlockingof a desired interlocked route.

RUNTHROUGH—intentionally passing a stationplatform without making a scheduled stop.

.

SEMAPHORE-a wayside signal device by whichindications are given by the position of a mova-ble arm in daylight hours and by the color of alight in darkness.

SHUNT-a conductor joining two points in anelectrical circuit so as to form a parallel or alter-nate path through which a portion of the currentmay pass.

SHUNTING SENSITIVITY-the maximum impe-dance that, when placed at the most adverseshunting location, will cause the track circuit toindicate the presence of a train.

SIDING—a track auxiliary to the main track, usedfor meeting, passing, or storing trains.

SIGNAL-a means of communicating direction orwarning.

Block Signal-a fixed signal at the entrance ofa block governing trains entering and usingthat block,

Cab Signal--a signal in the train operator’scab that governs the movement of that trainby conveying the automatic block aspectsand the prevailing speed command.

Clear Signal--a signal displaying the aspectindicating to proceed.

Home Signal-a fixed signal at the entrance ofa route or block governing trains enteringand using that route or block.

Opposing Signals-wayside signals governingtrain movements in opposite directions overthe same stretch of track.

Time Signal-a signal that controls trainspeed by requiring that a certain timeelapse between entering and leaving ablock.

Wayside Signal-a signal of fixed locationalong the track right-of-way.

SIGNAL ASPECT-See Aspect.

SLIDE (WHEEL)—the condition, during braking ordeceleration, where the surface speed of thewheel is less than train speed,

SLIP (WHEEL)—the condition, during accelera-tion, where the surface speed of the wheel isgreater than train speed. (Also called “Spin.”)

SLIP-SLIDE SYSTEM-an onboard system forautomatically detecting and correcting slip andslide by making compensating adjustments ofpropulsion and braking to maintain optimumtraction (wheel-rail adhesion),

SPEED

Civil Speed (Limit)—the maximum speedallowed in a specified section of track asdetermined by physical limitations of thetrack structure, train design, and passengercomfort.

Safety Speed (Limit)—the maximum speed atwhich a train can safely negotiate a givensection of track under the conditions pre-vailing at the time of passage. (Safety speedmay be less than or equal to civil speed,)

Schedule Speed-the speed at which a trainmust operate to comply with the timetable;mathematically, the distance from terminalto terminal divided by the time scheduledfor the trip (including station stops),

SPEED PROFILE-a plot of speed against distancetraveled.

SPEED REGULATOR-an onboard subsystem,usually part of the automatic train operation(ATO) system, that controls acceleration andbraking to cause the train to reach and maintain adesired speed within a given tolerance,

SPIN-See Slip.

STOP

Emergency Stop-stopping of a train by an ap-plication of the emergency brake, which—after initiation-cannot be released untilthe train has stopped.

Full Service Stop-a train stop achieved by abrake application, other than emergency,that develops the maximum brake rate.

Penalty Stop--irrevocable open-loop brakinginitiated by an onboard automatic systemor by a wayside trip stop as a result of ablock violation or uncorrected overspeed.

Programed Stop-a train stop produced byclosed-loop braking such that the train isstopped at a designated point according to apredetermined speed-distance profile.

213

Stop Signal--a signal indication requiring atrain to stop and stay stopped and permit-ting no exceptions such as running atreduced speed, movement within restrict-ing limits, or similar alternatives.

Train Protection Stop-a train stop initiatedby the automatic train protection (ATP)system.

SWITCH—a device that moves rails (switchpoints) laterally to permit a train to transfer fromone track to another. (See also Frog,)

Facing Point Switch-a track switch withpoints facing toward approaching traffic.

Trailing Point Switch-a track switch withpoints facing away from approachingtraffic.

SWITCH POINT-a movable tapered track rail,with the point designed to fit against the stockrail.

TERMINAL ZONE-a length of track, withinwhich the prescribed running direction can bereversed while it is occupied by a train,

TERRITORY—that portion of a route or route net-work characterized by a particular mode ofoperation or type of equipment, e.g., cab signalterritory, multiple track territory,

THIRD RAIL-See Contact Rail.

TRACK

Double Track—two parallel tracks, usuallywith each reserved for running in onedirection only.

Main Track-a track extending through yardsand between stations, upon which trainsare operated in revenue service or the useof which is governed by signals.

Reversible Track--a section of track on whichthe prescribed direction of running can bereversed if it is unoccupied and the oppos-ing home signals are at stop.

Single Track--a main track on which trainsare operated in both directions.

Transfer Track-a track in a yard area wheretransfer between main track and yardmodes of operation takes place.

TRACK CIRCUIT—(See Circuit, Track,)

TRAFFIC REGULATION--a train supervisoryfunction making use of changes in dwell time,performance level, acceleration rates, or othertrain performance characteristics to maintain in-tended traffic patterns and system stability,

TRAIN-a consist of one or more cars combinedinto an operating unit, (See also Consist.)

TRAIN BOARD—(See Model Board.)

TRAIN DETECTION EQUIPMENT—the trackcircuits and associated apparatus used to detectthe presence of trains in blocks,

TRAIN IDENTIFICATION-a method of desig-nating trains by means of such information astrain number, destination, or length; may be ac-complished automatically for functions such asrouting or dispatching.

TRAIN ORDERS—instructions used to govern themovement of trains manually, usually writtenand hand-delivered,

TRANSFER ZONE-a zone where changeoverfrom manual to automatic operation, or vice ver-sa, may be made. (See also Transfer Track, underTrack.)

TRIP STOP-a mechanical arm, located on thewayside, that can initiate a penalty brake ap-plication on a train that passes it by engaging abrake-triggering device (trip cock) on the train.Trip stops may be fixed, i.e., permanently posi-tioned in the tripping position; or they may beraised and lowered in response to signal indica-tions,

TURNBACK POINT-a point along the track, notat a terminal, where a train may reverse directionif allowed by the train control system, (See alsoTerminal Zone,)

TURNOUT-an arrangement of switch points andfrog with closure rails that permits trains to bediverted from one track to another,

UPSTREAM—track locations that, for a givenreference point and direction of travel, lie behindthe train and have been passed by it.

YARD-a network of tracks for making up trainsand storing cars.

214

Appendix E

CHRONOLOGY OF TRAIN CONTROLDEVELOPMENT

The history of train control technology in railrapid transit is interwoven with railroad engineer-ing. Most of the train control techniques applied inrail rapid transit have their origin in railroading,from which they are either borrowed directly oradapted to the special circumstances of the urbanset t ing. For this reason, many t rain controlengineers consider ATC in rail rapid transit simplyan extension of the field of railroad signaling.However, there are some distinct differences, bothin the technology and i ts appl icat ion. Thesimilarities and differences are evident in thechronology of train control development presentedhere.

The development of signaling and train controltechnology may be separated into two periods, with1920 as the dividing point. Before 1920 the majorareas of technological advance were interlockingcontrol and block signaling (manual and automatic).After 1920, the demand for moving heavier trafficat higher speeds and with increased safety led tomajor developments such as centralized traffic con-trol, continuous cab signaling, coded track circuits,and automatic train control. Generally, innovativesignaling and train control technology for rail rapidtransit was derived from railroads and laggedbehind railroad application by about 10 years. Therewere some notable exceptions; the development ofautomatic junction operation and automatic traindispatching was pioneered in rail rapid transit. Veryrecently, since roughly 1960, there has been someexperimentation with techniques and equipmentsolely for rail rapid transit and small people-moversystems.

The major source of this material is AmericanRailway Signal ing Principles and Pract ices,Chapter l—History and Development of RailwaySignal ing, published by the Associat ion ofAmerican Railroads, Signal Section, 1954. Supple-mentary information, particularly on rail rapid tran-sit technology in recent years, was assembled fromv a r i o u s s o u r c e s , i n c l u d i n g m a n u f a c t u r e r ’ sbrochures, local transit agency reports, and techni-cal journals.

1832

1843

1851

1853

1853

1860

1863

1866

The first fixed signal system in Americawas installed on the New Castle &Frenchtown RR, The signals were ball-shaped objects mounted on masts at 3-mile intervals. The signals were raisedand lowered by a signalman to indicatepermissible speed—low meaning stopand stay and high meaning proceed atfull speed, The latter indication gave riseto the expression “highballing.”

The first mechanical interlocking wasinstalled at Bricklayer’s Arms Junction inEngland. It was a simple machine oper-ated by a signalman who worked theswitches with his hands and the signalswith his feet.

Morse code electric telegraph was firstused in train operation for sending trainorders on the New York & Erie RR.

The Philadelphia & Reading RR installedsignal towers for giving information toapproaching trains on the occupancy ofthe track in advance.

Open-circuit manual block signaling wasfirst used in England.

Gate signals were initiated in America.A stop indication was displayed by plac-ing a red banner or disc on top of the gateduring the day. A red light was displayedat night.

Closed-circuit (fail-safe) manual blocksignal ing, using the space intervalmethod of operation, was first employedin America on the United New JerseyCanal & RR Co. between Kensington, Pa.(Philadelphia), and Trenton, N.J.

The first automatic electric block systemwas installed on the New Haven Systemat Meriden, Corm. Hall enclosed disc sig-nals, open circuit, were operated by trackinstruments.

215

1868

1870

1871

1872

1873

1876

1876

1880

1881

1885

216

The Pennsylvania RR used a type oftrain order signal which was under thecontrol of the train dispatcher who couldset it in the stop-danger position at anyremote station by means of a selectivedevice operated over the regular Morsetelegraph circuit.

The f i r s t i n t e r lock ing mach ine i nAmerica was installed at Top-of-the-Hill, a junction at Trenton, N. J., on theCamden and Amboy Division of thePennsylvania RR.

A system of automatic block signals,comparable with presently used equip-ment, was installed on the New York &Harlem RR and the Eastern RR.

The first installation of closed d.c. trackcircuit, invented by Dr. William Robin-son, was made at Kinzua, Pa., on thePhiladelphia & Erie RR.

The Robinson closed-circuit track blockfor switch protection was first put intouse on the Philadelphia & Erie RR.

The first power interlocking of the Burnpneumatic type was put in use on thePennsylvania RR at Mantua “Y,” WestPhiladelphia, Pa.

The Boston & Lowell and the Boston &Providence RRs introduced the Robin-son e l ec t romechan ica l s i gna l fo rautomatic blocking, controlled by directcurrent track circuits.

The first automatic train stop was placedin trial service on the Middle Division ofthe Pennsylvania RR. A glass tube in thetrain air line located on the locomotivenear the rails was designed to be brokenby a “track trip” set in operating positionwhen the signals were in the stop posi-tion.

The first interlocking of the hydraulictype was installed by the Union Switch &Signal Co. at Wellington, Ohio, for acrossing of the Wheeling & Lake Erie Ry.with the Cleveland, Cincinnati, Chicago& St. Louis Ry.

The “Dutch Clock” device for establish-ing time intervals (headways) between

1885

1889

1893

1900

1900

1901

trains was in use on the New York, NewHaven & Hartford RR and the New YorkCentral & Hudson River RR. When oper-ated automatically by a treadle device onthe rail, the passing train released apointer which started to move around adial divided into three segments eachrepresenting 5 minutes. The pointermovement was controlled by an escape-ment so that it moved across the dial in aperiod of 15 minutes. Headway for thetrain ahead was thus indicated up to 15minutes.

The first electric detector locking for in-terlocked track switches was installed bythe Pennsylvania RR at the Pittsburgh,Pa., terminal by using depression trips toground the indication circuit.

The first electric interlocking employingdynamic indication, invented by John D.Taylor, was installed at East Norwood,Ohio, at the crossing of the Baltimore &Ohio Southwestern RR and the Cincin-nati and Northern RR.

The first low-voltage, direct-current,motor-operated automatic semaphoreblock signals were installed on theCentral RR of New Jersey in Black Dan’sCut, east of Phillipsburg, N.J. They weretwo-position lower-quandrant signalswith the motor and driving chain outsidethe mast.

The first three-block indication was in-stalled on the Pennsylvania RR betweenAltoona and Cresson, Pa. The signalswere two-posi t ion, lower-quadrant ,home and distant automatic semaphores,

In Acton Town, England, an illuminatedtrack diagram was first used in connec-tion with resignaling on the District Ry.due to electrification, It dispensed withseparate track indicators and broughttogether all track occupancy informationon the plan of tracks and signals, therebyfacilitating the work of the signalmanhandling traffic.

The Taylor Signal Co. put in service thefirst electric interlocking embodying the“dynamic indication” principle, at Eau

1901

1901

1903

1906

1907

1909

1911

1912

Claire, Wis., on the Chicago, St. Paul,Minneapolis & Omaha Ry.

The Boston Elevated installed specialpolarized d.c. track relays. This was thefirst attempt to operate track circuits on arailroad where propulsion power wassupplied by electricity and the rails wereused as the medium for current return.

The Boston Elevated made the first per-manent installation of an automatic trains top sy s t em, w h i c h c o n s i s t e d o fmechanical wayside trips engaging brakecontrol apparatus on the moving car.

The North Shore RR of California madethe first installation of a.c. track circuitsfor automatic block signals.

The first signal system with a.c. trackcircuits on a road using a.c. propulsionpower was installed on the New York,New Haven & Hartford RR, The trackcircuits were the two-rail type, 60 Hz,with impedance bonds. Propulsion cur-rent was 25 Hz.

The first automatic interlocking for theprotection of a railroad crossing was in-stalled at Chester, Va., at a crossing ofthe Tidewater & Western Ry. with theVirginia Railway, Power & Light Co.

The Erie RR installed automatic signal-ing for train operation by signal indica-tion on a two-track division, 139.7 milesin length, which directed trains to: (1)stop and hold main track, (2) take siding,(3) proceed on main track regardless ofsuperior trains.

The absolute permissive block system(APB), developed by the General Rail-way Signal Co., was first installed on theToronto, Hamilton & Buffalo RR be-tween Kinnear and Vinemount, Ontario,Canada, using direct-current semaphoresignals.

Train movements on the Chesapeake &Ohio Ry. were directed for the first timeby signal indication without writtentrain orders.

1912

1914

1915

1919

1920

1923

1925

1926

Cab signals were first used on an electricrailway, the Indianapolis & CincinnatiTraction Co. ●

The cam controller for control of powerapplication to d.c. propulsion motors wasfirst used in the Chicago Rapid Transitco.

The American Railway Associat ionadopted rules which permitted trainoperation on single track by controlledmanual block signal indications, super-seding timetable and train orders.

The Buffalo, Rochester & Pittsburgh Ry.made the first trial installation of theGeneral Railway Signal Co. intermittentinductive train stop system. This systemused magnetic induction to transfer sig-nals from wayside controls to trainequipment.

The first installation of automatic speedcontrol in the US. was that of the ReganSafety Device Co. intermittent electricalcontact ramp-type train control systemon the Chicago, Rock Island & Pacific RRbetween Blue Island and Joliet, Ill.

The Pennsylvania RR placed in service,experimentally, the first installationanywhere of the continuous inductive cabsignal and train control system coveting43.5 miles of single track and 3.4 miles oftwo-track, between Lewistown and Sun-bury, Pa. It was the first instance wherevacuum tubes were used for purposesother than in communication circuits.This installation also was the first timethat cab signals were used in lieu ofwayside signals for operating trains bysignal indication.

The first permanent installation of cabsignals without wayside automatic blocksignals was made on the Atchison,Topeka & San ta Fe Ry . , be tweenChillicothe, Ill., and Ft. Madison, Iowa.The equipment was a Union Switch &Signal Co. three-speed continuous in-ductive-type train control device.

The Illinois Central RR was the first toequ ip an ope ra t i ng d iv i s i on w i thautomatic train stop and two-indication

217

●

1927

1930

1931

1932

1933

1934

1937

continuous cab signals without waysideautomatic block signals,

The first General Railway Signal Co.centralized traffic control system was in-stalled on the New York Central RR be-tween Stanley and Berwick, Ohio. Thef i r s t dua l - con t ro l e l ec t r i c swi t chmachines, which provided for eitherhand or electric operation, were in-troduced on this installation,

The first use of the all-relay interlockingprinciple, as a substitute for indicationparts and magnets at the levers of a largeinterlocking machine equipped withmechanical locking, was at ClevelandUnion Terminal, Ohio,

The New York Central RR installed asystem of four~block indication signalson a line equipped with automatic blocksignals in heavy suburban traffic terri-tory.

The Philadelphia subways installed amodified type of the three-wire circuitcode scheme of centralized traffic con-trol.

The Pennsylvania RR was granted per-mission by the ICC to convert all itslocomotives equipped with the codedcontinuous train stop system to thecoded continuous cab signal system withwhistle and acknowledger. This wasdone with the understanding that thePennsylvania RR would voluntarily ex-tend cab signal territory to include mostof its main line trackage.

The first installation of coded track cir-cuits on steam-operated territory wasmade be tween Lewis tone and Mt .Union, Pa., on 20 miles of four-trackmain line on the Pennsylvania RR, Theaverage length of track circuit was 5,201feet. Energy was coded storage batteryfor three and four-indication waysidesignals, with coded 100 Hz a.c. superim-posed for continuous cab signals,

The first installation of a relay-type in-terlocking with push-button automaticselection of routes and positioning ofswitches and signals, General Railway

1939

1939

1940

1940

1940

1940

Signal Co. Type “N-X” (entrance-exit),was made at Girard Junction, Ohio, onthe New York Central RR.

A four-indication, four-speed coded,continuous train control system was in-stalled on suburban cars of the KeySystem, Southern Pacific and Sacramen-to Northern Railroads operating over theSan Francisco-Oakland Bay Bridge,California, The system was designed tohandle 10-car mult iple-uni t t ra insoperating on a l-minute headway, Theinstallation included an N–X interlock-ing system with a train describer andautomatic operation of a single switch.

The first application of coded detectortrack circuits in interlocking was madeby the Norfolk & Western Ry.

The first installation of coded track cir-cuits for continuous cab signaling with-out wayside automatic signals in steamterritory, developed by the Union Switch& Signal Co., was made between Conpitand Kiskiminetas Junctions, Pa., on thePennsylvania RR,

The Pennsy lvan ia RR in s t a l l ed acentralized traffic control system be-tween Harmony and Effingham, Ill.,using the Union Switch & Signal Co,two-wire, 35-station time code type forthe first time on a multiple-connectedline circuit in which the line wires werecontinuous throughout the territory, andwhich provided for the coordination ofthe code circuit and communication cir-cuits over the same line wires, This wasthe first installation of a centralizedtraffic control system to employ a two-wire code line circuit in which all thefield locations were connected in multi-ple across the line wires,

The first installation of reversible codedtrack circuits in single-track territorywith centralized traffic control wasmade between Machias and Hubbard,N.Y., on the Pennsylvania RR.

The first installation of absolute per-missive block (APB) signaling with threeand four indications with coded track

218

—

1943

1944

1946

1948

1949

1951

1951

circuits was made onWestern Ry., betweenEvergreen, Va.

The first installation of

the Norfolk &Petersburg and

coded track cir-cuits using polar reverse codes withthree-indication signaling for either-direction operation was made on the St.Louis Southwestern Ry.

The f i r s t i n s t a l l a t i on o f no rma l lydeenergized coded track circuits forcentralized traffic control on single trackwas placed in service between Laredoand Polo, Me., on the Chicago, Mil-waukee, St. Paul & Pacific RR.

The Pennsylvania RR demonstrated thefeasibility of centralized traffic controloperation over commercial communica-tion circuits, including beamed radio.The test was made over approximately1,130 miles of Western Union carriertelegraph circuit including about 90miles of beamed radio. This was the firsttime beamed radio was used for this pur-pose.

The first use of automatic train dispatch-ing in rail rapid transit was by thePhiladelphia Rapid Transit Co. (nowSEPTA). The device employed a perfor-ated opaque tape driven by a clockmechanism. A beam of light scanningthe tape triggered a photoelectric cellthat automatically activated startinglights at terminals.

The Chicago Transit Authority initiatedexperiments in the use of radar for traindetection and separation assurance.

The Pennsylvania RR installed a three-speed continuous inductive train controlsystem in which the limits were 20 milesper hour with no code, 30 miles per hourwith 75 code, 45 miles per hour with 120code, and no speed limit with 180 code.

CTA began the use of automatic traind i spa t ch ing w i th r emo te ove r r i decapability fromsystem, whichclock, pen graphment, and lineoperation.

.

central locations. Theemploys a mechanicalrecorders of train move-supervision, is still in

1951

1952

1953

1953

1955

1959

1961

1962

1964

A portable radio, called “Dick Tracy,”was first used by yard switchmen on theSouthern Ry. in connection with cou-pling cars in the classification yard andtransferring them to the departure yard.

The Erie RR placed in service at Water-boro, N, Y., in connection with theestablishment of a remotely controlledinterlocking, a system of automatic trainidentification, This system automaticallyidentifies the direction and the numberof a train as it clears a manual block on abranch line.

The first installation of cab signalingusing transistors in place of vacuumtubes was placed in service on the NewYork, New Haven & Hartford RR by theGeneral Railway Signal Co.

The first installation using transistors in-stead of vacuum tubes in safety-type(vital circuit) carrier equipment wasmade on the Pennsylvania RR. Theequipment was developed by the UnionS w i t c h & S i g n a l D i v i s i o n o fWestinghouse Air Brake Co.

A crewless remote-controlled passengertrain was demonstrated on the NewYork, New Haven & Hartford RR.

The inductive train phone was first usedin raiI rapid transit by the Chicago Tran-sit Authority.

A completely automatic subway trainwas placed in service on the shuttle runbe tween T imes Squa re and GrandCentral Station in New York. A motor-man was on board for emergencies, buthe was not involved in normal operationof the train and often spent his timereading the newspaper,

A crewless freight t rain operat ingsystem was tested on the Canadian Na-tional RR.

Automatic train operation (ATO) equip-ment, intended for use in the BARTsystem, was operationally tested atThorndale on the Chicago Transi tAuthority North-South route.

219

1966

1966

1967

1969

1971

Four automatic train control systems forBART were demonstrated at the Diablotest track--one using the moving blockconcept, two using coded track circuits,and the other using a “trackwire” com-munications link and wayside controlequipment.

Fully automated vehicle operation andinnovative methods of train control weredemonstrated for the Transit Express-way (Skybus) system at South Park, Pa.,by the Port Authority of AlleghenyCounty (Pittsburgh).

Audio-frequency track circuits in a railrapid t ransi t appl icat ion were f i rs tplaced in regular service by the ChicagoTransit Authority.

Revenue service was begun on thePATCO Lindenwold Line. After amanually initiated start, train operationis completely automatic until the doorsare opened at the next station.

An automatic people-mover system wasplaced in operation at the Tampa Air-port. This system incorporates some ofthe ATC elements originally demon-strated at South Park,

1972

1972

1973

1974

1974

Fourwere

automatic people-mover systemsdemonstrated at TRANSPO ’72,

Washington, D. C., under the auspices ofthe U.S. Department of Transportation.

Revenue service was initiated on theFremont-MacArthur por t ion of theBART system, Train operation, includ-ing start, berthing, and door operation, isentirely automatic but under the super-vision of an onboard operator,

The Satellite Transit System, featuringautomatic crewless vehicle operation,was placed in service for passengers atthe Seattle-Tacoma (Sea-Tac) Airport.

The AIRTRANS system at Dallas/Ft.Worth Airpor t opened for service .Operating on 17 interconnected routes,AIRTRANS has automatic crewlesstrains to carry passengers, baggage,freight, and refuse within the airportcomplex.

Demonstration of the Morgantown (W,Va,) PRT system was conducted, Smallvehicles, operating on a fixed guideway,circulate under automatic control andwithout onboard operators,

220

Appendix F

PERSONS AND ORGANIZATIONS VISITED

During the course of this study, visits were made to each operating transit agency and to other organiza-tions with an interest in the design and operation of rail rapid transit systems. The following is a list of thepersons interviewed and their organizational affiliation. Individual recognition of their contributions to thestudy is not possible, except in a few cases where their observations are cited directly in the report. TheOTA staff and the Urban Mass Transit Advisory Panel are grateful for the cooperation of these people andfor the generous gift of their time and interest. Persons interviewed by the OTA staff are listed on the left.Those interviewed by the technical consultants, Battelle Columbus Laboratories, are listed on the right.

OPERATING TRANSIT AGENCIES

Boston—Massachusetts Bay Transportation Authority (MBTA)

Jason B. Baker Larry BulongieSupt., Equipment Engineering Signals and Communications

Francis X. Capelle James A. BurnsDeputy Supt., Signals and Communications General Supt., Equipment Maintenance

Joseph P. Dyer Raymond M. CaddiganGeneral Supt., Power and Signals Supt. of Rail Lines

William Malone Francis X. CapelleSafety Department Deputy Supt., Signals and Communications

David Marcham Frank CrowleyTransportation Department Deputy Supt., Maintenance

Col. Warren J. HigginsDirector of Operations

Larry MaddalenaElectrical Supervisor

Chicago-Chicago Transit Authority (CTA)

Joseph Bensen Kendrick BissetLibrarian S u p e r v i s o r , S i g n a l D e s i g n .

James Blaa Paul CleaverManager of Transportation Supt., Signals and Communications

Thomas Boyle Arthur SandbergDirector of Safety Manager, Engineering

Albert Brandt James StewartSignal Maintainer, Lake-Harlem Shops Manager, Equipment Engineering

Daphne Christensen Ralph TracyScience Advisor to the Chairman of the Board Director of Transportation Services

221

Paul C. CleaverSupt., Signals and Communications

Adel El DibManagement Systems

Judy GennesonLibrarian

Edward HenryPublic Safety

Harold HirschManager of Operations Planning

Paul KoleGeneral Manager, Finance

George KramblesGeneral Manager, Operations

Herbert LowensteinDistrict Supervisor (North)

Katherine MoriartySupervisor, Collection Agents

Evan C. OlmsteadManager of Maintenance

A. R. SandbergManager of Engineering

Theodore SzewcSupervisor, Signals Maintenance

Ralph TracyDirector of Transportation Services

Joanne VlecidesDevelopment Planning

Thomas WolgemuthDirector, Plant Maintenance

Chicago—Illinois Department of Transportation

B. G. CunninghamFormer Assistant Secretary of

Transportation (Illinois DOT)

Cleveland—Cleveland Transit System (CTS)

Elmer CorlettDirector of Personnel

Bernard HamperEmployment Supervisor

Leonard KraynikChief Research Analyst

222

Michael McKennaSupt., Power and Plant

Philip RockwoodManager of Maintenance

Myron SilsbySupt. of Rapid Transit

Bernard WilchekSupervisor of Management Development,

Training and Safety

Dear WilliamsonManager of Operations

Donald YuratovacResearch Specialist, Research and Planning

New York—New York City Transit Authority (NYCTA)

Charles Kalkhof Louis AlessiAssistant General Superintendent, Assistant Supervisor,Maintenance of Way Car Maintenance

William O’Neill Frank T. BerryAsst. Supervisor of Signals Executive Officer,

and Maintenance

Joseph Calderone

Irwin Cohen

Frank De Maria

Operations

Supt. of Rapid Transit

Seymour DornfeldDivision Engineer, Signals andCommunications

Donald GillChief Industrial Engineer

Charles KalkhofAssistant General Superintendent,Maintenance of Way

Harold J. McLaughlinAssistant General Supt., RapidTransit Transportation

Dennis NewmanDeputy Subdivision Engineer,Maintenance of Way

Jack RoggSupt. of Engineering and ProductionControl

223

Daniel T. ScannellSenior Executive Officer, OperationsManagement

Bart SheehanRailroad Signal Specialist,Maintenance of Way

Ludwig StanitschSenior Transit Management Analyst

Thomas SullivanSupt., Rapid Transit

Philadelphia, Camden—Port Authority Transit Corporation (PATCO)

David L. Andrus, Jr. James F. ElderSupervisor of Traffic and Planning Supt. of Operations

Howard C. Conings Joe FioriSafety and Insurance Supervisor General Foreman,

Robert B. Johnston Robert B. JohnstonGeneral Manager General Manager

Robert S. Korach Robert S. Korach

Signals and Communications

Assistant General Manager and Supt. ofOperations

John A. Lane J.Administrative Assistant to the General

Manager

Assistant General Manager andSupt. of Operations

William VigrassSupt. of Equipment

John J. McBride D. R. WolfeCaptain of PATCO Police Supt. of Way and Power

Herbert McCrearyStatistician

J. William VigrassSupt. of Equipment

San

Clay YostEquipment Engineer

PATCO ConsultantsGibbs & HillJack R. Shepard

Chief Systems Engineer (Ret.)

F r a n c i s c o — B a y A r e a R a p i d T r a n s i t ( B A R T )

Malcolm BarrettGeneral Counsel

Ward D. Belding, Jr.Senior Economic Analyst

Ray CarrollDirector of Maintenance

Krishna HariTrain Control Systems Engineer

George AnasWilliam Briner

Supt. of Line Operations

Ray CarrollDirector of Maintenance

Marvin DenowitzManager of Quality Control

Krishna HariTrain Control Systems Engineer

224

Fred HarmonTransportation Division

William F. HeinDirector of Planning

Mary-Lou MulhernLibrarian

Eugene P. NunesSafety Supervisor

Philip O. OrmsbeeAssistant to General Manager

William M, ReedyFormer Member, BART Board

William J. RhineDirector of Engineers

Ronald E. RypinskiSupervisor of Maintenance Planning

George M. SillimanFormer Member, BART Board

BART Consultants

Hewlett-PackardDavid CochranLeonard Cutler

Lawrence Berkeley LaboratoryDon M. EvansD. Theodore Scalise

Roy W. Harris, TRWManager, BART Safety Availability Project

Clarence LovellMember, Special Panel, Public Utilities

and Corporations Committee, CaliforniaSenate

Bernard M. OliverMember, Special Panel, Public Utilities

and Corporations Committee, CaliforniaSenate

William WattenburgIndependent Consultant

San Francisco—State of California

William F. HeinDirector of Planning

Mary-Lou MulhernLibrarian

William J. RhineDirector of Engineering

R. SomertonTraining Supervisor

BART Consultants

Parsons-Brinckerhoff-Tudor-BechtelWalter Quintin of the Bechtel Corp.TRW

Hal BuchananRoy Harris, Project ManagerEverett Welker

Alfred E. AlquistChairman, Public Utilities andCorporations Committee,California Senate

225

— .

James CooneyOffice of the Legislative Analyst

Wayne KeithleyOffice of the Legislative Analyst

James R. MillsMember, Public Utilities andCorporations Committee,California Senate

A. Alan PostLegislative Analyst

S a n F r a n c i s c o — C a l i f o r n i a P u b l i c U t i l i t i e s C o m m i s s i o n

James K. GibsonDirector, Transportation Division

Leo L. LeeSenior Electrical Engineer

William L, OliverTransportation Division

Herman W. PrivetteSenior Transportation Supervisor

Theodore E. RogersChief, Railroad Operations andSafety Branch

In addition to contacts with the preceding people and organizations who are involved in operating transitagencies, OTA technical consultants from Battelle Columbus Laboratories interviewed the following per-sons.

A G E N C I E S I N P L A N N I N G O R C O N S T R U C T I O N A C T I V I T Y

Atlanta—Metropolitan Atlanta Rapid Transit Authority (MARTA)

Al Locke, System Safety EngineerJohn J. Tucker, Equipment Design ManagerErnest Young, Equipment Design Engineer

ConsultantsParsons-Brinckerhoff-Tudor-BechtelDale H. Fencken, Supervising Engineer, ATC

Baltimore—Mass Transit Administration (Maryland)

Walter J. Addison, AdministratorCarl Buhlman, Manager, Equipment and Equipment SystemsFrank W, Hearne, Director, Rapid Transit Development DivisionBernard Walker, Senior Electrical Engineer

ConsultantsDaniel, Mann, Johnson, and Mendenhall/Kaiser EngineersP. Morris Burgess, Chief Systems EngineerDavid Hammond, Project ManagerLewis Sanders, Senior Systems Engineer

226

Buffalo—Niagara Frontier Transportation Authority (NFTA)

Kenneth Knight, Manager of Metro ConstructionEugene Lepp, Manager of Systems

Denver—Regional Transit District (RTD)

Dr. J. Edward Anderson, AdvisorGeorge Billman, Manager, Technology AssessmentAubrey J. Butts, System DesignCarlos DeMoraes, Assistant Executive Director, DevelopmentR. J. Farrell, Command and Control SpecialistRalph Jackson, Director of PlanningPaul Newcomb, Director, System DesignJohn Simpson, Executive DirectorA. J. Weaver, Subsystem RequirementsW. Wild

ConsultantsSystems Management Contractor (TRW Systems et al.)E. Bagerstos, Reliability/SafetyR. K. Boyd, Assistant Program Manager, ControlsA. F. Ems, Reliability/SafetyRobert Johnson/SimulationHerman Wells, Controls

Miami—Metropolitan Dade County Transit Authority

E. Randolph Preston, Asst. Coordinator for Transportation Development

ConsultantsKaiser EngineersJames AllisRalph MasonEugene Stann, Project Manager

Minneapolis-St. Paul—Twin Cities Area Metropolitan Transit Commission

Camille Andre, Executive DirectorJohn Jamieson, Director of Transit DevelopmentDouglas Kelm, Chairman, Metropolitan Transit CommissionWilliam Marshall, Systems EngineerRobert Pearson, Project Manager, Fixed-Guideway Systems

ConsultantsDe Leuw CatherFrank Smith, Small Vehicle Study

OthersDr. J. Edward Anderson, University of MinnesotaJ. Kiedrowski, Staff Member, Minnesota Senate Metropolitan Urban

Affairs Committee

Pittsburgh—Port Authority of Allegheny County (PAAC)

72-683 0 - 76

John Mauro, Executive DirectorJames Maloney, Acting Manager, Early Action ProgramRobert Sedlock, Systems Technology

16

227

ConsultantsKaiser EngineersThomas R. Gibson, Systems EngineeringZoltan Stacho, Project Manager

Washington, D.C .—Washington Metropolitan Area Transit Authority (WMATA)

C. David Allen, Systems Maintenance EngineerHoward W. LyonDavid Q. Gaul, Assistant Director, Office of Equipment DesignWarren Quenstedt, Deputy General ManagerWilliam RandolphRalph Sheldon, Train Control Engineer

ConsultantsGibbs and HillJoseph Smith, Senior Transportation EngineerJack Shepard, Chief Systems Engineer (Retired)

AUTOMATED AIRPORT SYSTEMS

Dallas-Fort Worth Airport—Dallas-Fort Worth Regional Airport Board (AIRTRANS)

Dennis Elliott, Manager of EngineeringDalton Leftwitch, Operations Supervisor, AIRTRANSDavid Ochsner, Manager, AIRTRANSDavid Slaboda, Maintenance Supervisor, AIRTRANS

SupplierLTVRichard Condrey, Manager, Ground TransportationAustin Corbin, Project ManagerJohn Loutrel, Operations Research

Seattle-Tacoma International Airport—Port of Seattle (Sea-Tac)

Arthur Krause, Airport Maintenance SuperintendentEugene Sagar, Operations Manager, Port of SeattleDon Shay, Director of Aviation, Port of Seattle

ConsultantsThe Richardson AssociatesHarry Linden, Assistant Administrative Director

SupplierWestinghouse Electric CompanyJoseph Borkowski, Maintenance Manager

SUPPLIERS

General Railway Signal Corp. (GRS), Rochester, N.Y.

Dr. John Freehafer, Manager of Advanced EngineeringArthur Gebhardt, Vice President, MarketingMark Sluis, Vice President, Engineering

228

Westinghouse Air Brake Company (WABCO), Union Switch and Signal Division,Swissvale, Pa.

Homer Hathaway, Senior Systems Consultant

Westinghouse Electric Company (WELCO) Transportation Division, Pittsburgh, Pa.

Philip Gillespie, Manager of Market DevelopmentJames H. King, Reliability EngineerDr. Robert Perry, Manager, Train Controls and Vehicle System Engineering

I N S T I T U T I O N S

Amalgamated Transit Union, Washington, D.C.

Walter Bierwagen, Member. General Executive Board and Director ofPublic Affairs

Earle Putnam, General Counsel

American Public Transit Association (APTA), Washington, D.C.

Robert Coultas, Deputy Executive DirectorJack Hargett, Legislative LiaisonB. R. Stokes, Executive Director

F E D E R A L

Transportation Systems Center (TSC),

Robert Casey, Systems Analyst

AGENCIES

Cambridge, Mass.

Harry Hill, Man-Systems TechnologyRobert Pawlak, Project Manager, ATC ProgramChan Watt, Systems Safety/Reliability/Maintainability

Urban Mass Transportation Administration (UMTA), Washington, D.C.

Steven Barsony, Acting Director, Morgantown Division, Office of R&DEdward J. Boyle, Manager, Safety Division, Office of Transit ManagementRay Brunson, Systems Development Branch, Office of R&DVincent DeMarco, Systems Development Branch, Office of R&DDuncan MacKinnon, Chief, Advanced Development Branch, Office of R&DPaul Spencer, Staff Engineer, Rail Branch, Office of R&D

Federal Railroad Administration (FRA), Washington, D.C.

C. M. Bishop, Chief, Signals Branch, Standards and Procedures DivisionRolf Mowatt-Laarsen, Chief, Standards and Procedures Division, Bureau of SafetyWilliam Paxton, Chief, Maintenance of Way Branch, Standards and Procedures Division

National Transportation Safety Board (NTSB), Washington, D.C.

J. Emerson HarrisRobert JewellThomas Styles, Chief, Railroad Safety Division

229

Appendix G

BIOGRAPHIES OF URBAN MASS TRANSITADVISORY PANEL

George Krambles, ChairmanGeneral Operations ManagerChicago Transit Authority (CTA)Chicago, Ill.

Mr. Krambles is responsible for the transportation, maintenance, and operations planning departmentswhich together include more than 11,000 of CTA’s 12,500 employees. Prior to assuming the position ofGeneral Operations Manager, he served as operating manager in charge of the transportation and shops andequipment departments from 1972 to 1973. He was superintendent of research and planning related to serv-ice and marketing from 1965 to 1972, and superintendent of operations for the transportation departmentfrom 1961 to 1965. In addition, he was named manager in 1964 for the 2-year Skokie Swift mass transporta-tion demonstration grant project.

Mr. Krambles is a graduate of the University of Illinois and a registered professional engineer in Illinois.After a brief period with the Indiana Railroad, he joined the Chicago Rapid Transit Company, serving in themechanical and electrical departments. His work during this period included maintenance and constructiondesign as well as power system operation.

Walter J. BierwagenVice President and Director of Public AffairsAmalgamated Transit UnionAFL–CIO .

Mr. Bierwagen has spent most of his professional life in a leadership role in organized labor. From 1951 to1964, Mr. Bierwagen was president and business agent of the Washington local of the Amalgamated TransitUnion representing Washington transit employees. In this role he was the primary officer who directed col-lective bargaining and legislative programs of interest to his union. Subsequently he became vice presidentof the international union, engaging in legislative representation before Congress on behalf of the membersof the Amalgamated Transit Union, and conducting collective bargaining negotiations on behalf of manyATU local unions in the Eastern section of the United States. In addition he has been involved in State ac-tivities for the labor movement by serving as vice president of the Maryland State Federation, AFL–CIO, aswell as a principal officer of the Washington Central Labor Council. His interests have gone beyond thelabor movement, including active leadership in Group Health Association, and the development of transithealth, welfare, and pension funds on the local level. As mentioned above, Mr. Bierwagen’s principal ac-tivity has been in the legislative arena. In that role, he played an important part in the development of theUrban Mass Transportation Act of 1964 and its amendments, especially the requirements to protectemployees, and in the development of similar legislation at the State level.

231

—

Robert A. BurcoDeputy DirectorOregon Department of TransportationSalem, Oreg.

Mr. Burco has been active in the field of urban transportation policy research since 1967, performing studiesfor local and State governments and the Federal Government. Most recently he has headed his own consult-ing firm specializing in transportation and environment policy. Previously he was employed in a researchcapacity at the Stanford Research Institute and at Bell Telephone Laboratories. Mr. Burco received a B.S. andM.S. from Stanford University and a second master’s degree from the University of California at Berkeley.

He has been involved professionally in the activities of the Transportation Research Board and has lecturedwidely on the subject of transportation. His clients have included a number of public interest groups, theCalifornia State Legislature, Office of the Mayor of Los Angeles, and the California Department ofTransportation. He is familiar with transportation policy in Europe, Canada, and Japan, having worked oninternational assignments with the Organization for Economic Cooperation and Development in Paris. InSeptember 1975, he joined the administration of Governor Straub as Deputy Director of the Oregon Depart-ment of Transportation.

Jeanne J. FoxAssociate Director, ResearchJoint Center for Political Studies

Mrs. Fox has conducted research on transportation as a public policy issue since 1971. Prior to her presentposition, she was a consultant at Mark Battle Associates, and before that she was employed at the UnitedStates Information Agency.

She is a graduate of the University of Minnesota.

Mrs. Fox was the principal author of Urban Transportation: Minority Mobility in the 70’s (DC-RDG-12),prepared for UMTA, Civil Rights Division. She was one of two principal investigators and co authors forUMTA research which resulted in the following three reports:

Transportation for the Elderly and the Handicapped (DOT-UT-533), Marketing Techniques and the MassTransit System (DOT-UT-533),(DOT-UT-533).

She also wrote the “Public Transit

Lawrence A. Goldmuntz

Marketing Techniques and the Mass Transit System—A Handbook

in the Spotlight, ” Focus, 1974.

President, Economics & Science Planning, Inc.Washington, D.C.

Dr. Goldmuntz is a consultant and the present Visiting Professor of Engineering and Public Policy at Car-negie-Mellon University in Pittsburgh, Pa. Prior to his present positions, he worked in the Office of theAssistant Secretary of Transportation for Research and Technology and chaired the Metroliner SteeringCommittee, charged by the Secretary of Transportation to supervise the completion of the Northeast cor-ridor Washington-New York high speed rail project. He also served as Executive Secretary of the Air Traffic’Control Advisory Committee at DOT, as Assistant Director for Civilian Technology in the Office of Scienceand Technology of the Executive Office of the President, and as chairman of a committee for the Office ofScience and Technology which reviewed the Cumulative Regulatory Effects on the Cost of AutomotiveTransportation (RECAT). Previously in private industry, he served as President of TRG Inc., a research anddevelopment organization involved in air traffic control and other electronic systems.

Dr. Goldmuntz is a graduate of Yale University where he received his B.E.E. (1947) and M.E.E. (1948)degrees and h@ Ph.D. (1950) in Applied Science.

232

Dorn C. McGrath, Jr.Chairman, Department of Urban and Regional PlanningThe George Washington UniversityWashington, D.C.

Professor McGrath has served as an advisor or consultant to a variety of agencies concerned with transpor-tation and urban growth policy, including the House Committee on Public Works, the House Committee onBanking and Currency, the U.S. Commission on Population Growth and the American Future, the U.S.Aviation Advisory Commission, the Port of Oakland, and the North Central Texas Council of Governments.He is a member of the Environmental Studies Board of the National “Academy of Sciences and Chairman ofthe Environmental Advisory Board for the Chief of Army Engineers.

Professor McGrath is a graduate of Dartmouth College where he received his B.A. degree. He received hisM.A. in City Planning from Harvard University.

He is a principal author of the National Academy of Sciences report, Jamaica Bay and Kennedy Airport,published in 1970, and has written numerous articles on transportation and related community developmentplanning.

Bernard OliverVice-President of Research and Member of the Board of DirectorsHewlett-Packard CompanyPalo Alto, Calif.

Dr. Oliver previously worked on the development of automatic tracking radar, television transmission, in-formation theory and efficient coding systems, as a member of the technical staff of the Bell TelephoneLaboratory (New York) from 1940 to 1952. He joined Hewlett-Packard in 1952 as director of research andwas appointed Vice-President of Research and Development in 1957. Presently, he is also a lecturer inelectrical engineering at Stanford University and a member of the Science and Technology Advisory Com-mittee to the California State Assembly. He recently served on a State senate panel investigating the safetyaspects of the BART system.

Dr. Oliver is a 1935 graduate of Stanford University where he received his A.B. and M.S. degrees in electricalengineering. In 1940 he received his Ph.D. degree in electrical engineering from Cal Tech.

He is the author of numerous technical articles and holds over 50 U.S. patents in the field of electronics. Heis a Fellow of the IRE and has served as vice president and later president of the IEEE.

Simon ReichSupervising Engineer-SignalsGibbs & Hill, Inc.New York, New York

Mr. Reich took his present position in the Signals department, after serving in the Transportation andSystems departments of Gibbs & Hill. He has an extensive background in automatic train control (ATC). Hehas worked on the design of signaling and control systems and the technical coordination of the WMATArail rapid transit system, including the ATC and communication functions. Prior to association with Gibbs &Hill, he was involved in the design of train control demonstration systems for the BART Concord test track,the CTA Lake Street Line, the CTS Airport Extension, and the MBTA South Shore Project-all while work-ing for the General. Railway Signal Company. Other activities of Mr. Reich at GRS include the engineeringfor automatic train operation systems and train control research and development for rapid transit lines andrailroads. These included projects on automatic train operation for the NYCTA 42d Street Shuttle, and a cabsignaling system for the Netherlands State Railway, This work involved design of train-borne equipmentand instrumentation to measure train performance.

Mr. Reich is a 1959 graduate of the Polytechnic Institute of Brooklyn where he also received a B.S. inPhysics.

233

Frederick P. SalvucciSecretary, Executive Office of Transportation and ConstructionCommonwealth of MassachusettsBoston, Mass.

Secretary Salvucci has been occupied, both professionally and personally, with the problems of urban plan-ning and transportation in the Boston area since his formal training at the Massachusetts Institute of Tech-nology. Prior to his appointment as Secretary, Mr. Salvucci held transportation and management jobs withthe Boston municipal government and worked as a transportation planner for the Boston RedevelopmentAuthority,

Mr. Salvucci helped to organize and found Urban Planning Aid, an advocacy planning group established toprovide technical and planning assistance to low-income and other community groups. He has also been ac-tive in the Greater Boston Committee on the Transportation Crisis, a public transportation advocacy group,the Massachusetts Air Pollution and Noise Abatement Committee, an organization concerned with promot-ing a shift from air to rail travel, and the East Boston Neighborhood Council,

Mr. Salvucci’s formal training at M.I.T. concluded in 1962 with an M.S. in civil engineering. He was namedFulbright Scholar in the 1964-65 academic year.

Thomas Chapman Sutherland, Jr.Assistant Dean, School of Architecture and Urban PlanningPrinceton UniversityPrinceton, N.J.

Dean Sutherland, prior to his present position, was Assistant Director, Office of Research and Project Ad-ministration at Princeton and before that, Assistant to the Chairman of Princeton’s Department ofAstrophysical Sciences, He has served as former Chairman of the South New Jersey Group of the SierraClub, a Trustee of the Stony Brook-Millstone Watersheds Association, Vice Chairman of the Princeton Con-servation Commission, and recently a member of the New Jersey Solid Waste Advisory Council. DeanSutherland is presently Chairman of the University Environmental Advisory Committee and a member ofthe Gateway Citizens Committee.

He is a graduate of the United States Naval Academy. While in the Navy, he served in the submarine serviceand, as a staff member of the Navy’s Office of Special Projects, worked on the development of the Polarismissile.

Dean Sutherland has authored numerous articles on conservation, ornithology, astronomy, and railroads.He is co author of the book, The Way to Go: The Coming Revival of U.S. Rail Passenger Service (Simon &Schuster), published in 1974. He is presently on the Board of Directors of the National Association ofRailroad Passengers (NARP).

Stewart F, TaylorVice President & Director Mass TransportationSanders & Thomas, Inc.Pottstown, Pa.

Mr. Taylor has been a consultant on transportation projects for public agencies at the Federal, State, andlocal level and for private corporations, Before becoming a consultant, he served with the former Penn-sylvania Railroad in various staff and management positions, He was also Chairman of the 1975 NationalConference on Light Rail Transit, jointly sponsored by the Urban Mass Transportation Administration, theNational Research Council, the American Public Transit Association, and the University of Pennsylvania.

Mr. Taylor is a graduate of Yale University and Harvard Law School. He has authored numerous articles andpapers on transportation, The most recent is entitled, “Urban Transportation—Another Alternative,”published by the Heritage Foundation of Washington, D.C. His work has appeared, on several occasions, inthe United States Congressional Record.

234

Appendix H

REFERENCES

American Transit Association, “Vandalism andPassenger Security, ” September 1973.

Association of American Railroads, Signal Section,American Railway Signaling Principles and Prac-tices, Chapter 1: History and Development of Rail-way Signaling, Chicago, Ill., 1954.

Air Transport World, “Facts and Figures,” Vol. 12,

No. 1, January 1975.

Battelle Columbus Laboratories, Safety in UrbanMass Transportation, Report No. UMTA RI-06-0005-75-1, prepared for Urban Mass TransportationAdministration, Washington, D, C., April 1975,

Department of Transportation, TransportationSystems Center, Safety and Automatic Train Con-trol for Rail Rapid Transit Systems, Report No.DOT-TSC-OST-74-4, prepared for the Office of theSecretary, Washington, D, C., July 1974,

Grose, V. L., “Constraints on Application ofSystems Methodology to Socio-Economic Needs,”presented to First Western Space Congress, SantaMaria, Calif., October 1970.

Highway Research Board, Task Force on UrbanMass Transportation Safety Standards, Safety in Ur-ban Mass Transportation: the State of the Art, forthe National Research Council, Washington, D. C.,September 1973.

Institute for Rapid Transit, Moving People Safely:Safety Guidelines for Urban Rapid Transit Systems,Washington, D. C., January 1974 (2d ed.).

King, J. H., “Making Transit Reliability a Reality,”Westinghouse Transway News, October 1975.

MITRE Corp., Research Requirements of the RapidRail Industry, Report No. UMTA-TRD-90-71, pre-pared for Urban Mass Transportation Administra-tion, Washington, D. C., June 1971. (PB 204 438)

National Transportation Safety Board, SpecialStudy of Rail Rapid Transit Safety, Report No.NTSB-RSS-71-1, Washington, D. C., June 1971.

National Transportation Safety Board, SafetyMethodology in Rail Rapid Transi t SystemDevelopment t, Report No. NTSB-RSS-73-1,Washington, D. C., August 1973.

Reistrup, P., President of the National RailroadPassenger Corporation (Amtrak), Statement beforethe Subcommittee on Transportation of the Com-mittee on Appropriations, United States Senate,April 7, 1975.

Transit Development Corporation, Safety Prioritiesin Rail Rapid Transit, Report No. UMTA-DC-06-0091-75-1, prepared for Urban Mass TransportationAdministration, Washington, D.C., March 1975 (2Vols.),

United States Congress, Urban Mass TransportationAct of 1964, P.L. 88-365, 78 Stat. 302. (49 U.S.C.1601).

United States Congress, Department of Transporta-tion Act (October 15, 1966), P.L. 89-670, 80 Stat. 931(49 U.S.C. 1653).

United States Congress, Federal Railroad Safety Actof 1970, P.L. 91-458, 84 Stat. 791 (45 U.S.C. 421).

United States Congress, National Mass Transporta-tion Assistance Act of 1974, P.L. 93-503, Stat. 1565(49 U.S.C. 1601).

235

Appendix I

CONGRESSIONAL LETTERS OF REQUEST

U N I T E D S T A T E S S E N A T E,

C O M M I T T E E O N A P P R O P R I A T I O N S ,

Washington, D. C., February 25, 1974.Hon. EDWARD M. KENNEDY,Chairman, Technology Assessment Board,House Annex, Washington, D.C.

D EAR MR. CHAIRMAN : On behalf of Senator Robert C. Byrd, Chairmanof the Transportation Subcommittee, and Senator Clifford P. Case, theSubcommittee’s Ranking Minority Member, I am transmitting the at-tached technology assessment request to you.

With kindest personal regards, I amSincerely,

J O H N L. MC C L E L L A N, C h a i r m a n .

Enclosure.

U N I T E D S T A T E S S E N A T E,

Washington, D. C., February 6, 1974.Hon. JOHN L. MC C L E L L A N,

Chairman, Senate Appropriations Committee, New Senate Office Build-ing, Washington, D.C.

D EAR MR. CH A I R M A N : We would like to enlist your support for aprompt and thorough study of automation in federally supported urbanrail transit projects.

This matter of increasing concern to our Subcommittee arises becauseseveral large cities, including Baltimore and Atlanta, are planning auto-mated train systems and are or will be seeking substantial federal fundingwithin the next two years.

At the same time, serious questions have arisen as to whether and towhat degree Automated Train Control (ATC) should be used in rail tran-sit.

The recent experience with San Francisco’s new rail system, known asBART, has helped focus attention on this problem.

Original plans for BART called for a fully automated system requiringno on-board train operator. This has not worked out because of a series ofmalfunctions in the ATC system. Costly patch-up work, with substantialfederal help, is underway, but complete automation of BART now ap-pears out of the question.

In light of the BART experience we should be alert to see to it that thesame expensive mistakes are not made in other federally supported urbanrail transit projects involving Automated Train Control.

237

At present, there is no means of assuring that the mistakes made in theBART project will not be repeated.

A draft study just completed by the Department of Transportation’sTransportation Systems Center states that train control “typicallyreceives little priority and emphasis” even though—as the study empha-sizes—this choice of system greatly affects revenue, safety, including, weadd, the serious matter of crime prevention, and operation and mainte-nance costs. The DOT study did not purport to deal with cost and cost sav-ings in detail, but it did state that there seemed to be an “intuitiveconclusion that an automated system should be more economical than aman-operated system in achieving or surpassing a given level of service orsafety. ”

The Congress and this Committee should not accept an “intuitive”judgment on matters of such cost and complexity.

There are at least two questions that require particular study: (1) towhat extent should urban rail transit systems be automated? and (2) howshould these projects be planned and executed?

The appropriate body to carry out such an independent, in-depth studyfor this Committee is Congress’ Office of Technology Assessment. Underthe provisions of the “Technology Assessment Act of 1972” (P.L. 92–482,Sec. 3(d), (l)), we ask that you transmit to the Chairman of the Tech-nology Assessment Board our request for a study that would:

1. Assess the state of automated train control technology and itsapplication to existing and planned rail transit systems.—What ma-jor research is underway and what is its objective? What train con-trol systems are being considered for transit projects now in theplanning stage? What are the characteristics of these systems andhow are they similar to or different than those of BART and otherhighly automated systems in use?

2. Assess the testing methods by which the workability of auto-mated train projects is determined .—To what extent are prototypesbuilt and tested? What has been the lesson of BART and other re-cent projects concerning the necessity for system testing duringdevelopment? What provisions have been made for the testing oftrain control systems now being planned?

3. Assess the process by which new rail transit systems or exten-sions of existing systems are planned and executed; evaluate theadequacy and professionalism of cost, safety, including crime pre-vention, and other analyses used.—What criteria are used, par-ticularly in determining degree of automation? To what extent areeconomic tradeoffs (i.e., cost of partially manual vs. fully automatedsystem) explicitly considered? How and to what extent is publicoversight maintained throughout the project? What federal require-ments, if any, apply to these federally assisted projects?

Your assistance in transmitting this request will be appreciated,Sincerely,

R O B E R T C. B Y R D,

Chairman, Transportation Appropriations Subcommittee.C L I F F O R D P. CA S E,

Ranking Minority Member,Transportation Appropriations Subcommittee.

238U. S. GOVERNMENT PRINTING OFFICE : 1976 0-72-683