DOT/FAA/RD-92/26DOT-VNTSC-FAA-92-8Research and Development ServiceWashington, DC 20591

Crew Resource Management:An Introductory Handbook

307-:

wx" lfi;v100

m X\7

RSPA/VNTSC

«tejMAG

-\

493 :•

COCKPIT HF PROGRAM

James E. DriskellFlorida Maxima Corporation

Richard J. AdamsAdvanced Aviation Concepts, Inc.

Final Report

August 1992

This document is available to the publicthrough the National Technical Information ServiceSpringfield, Virginia 22161

©U.S. Department of TransportationFederal Aviation Administration

NOTICE

This document isdisseminated underthe sponsorshipof the Department of Transportation in the interest

of information exchange. The United States Governmentassumes no liability for its contents or use thereof.

NOTICE

The United States Government does not endorseproducts or manufacturers. Trade or manufacturers'

names appear herein solely because they are consideredessential to the object of this report.

REPORT DOCUMENTATION PAGE Form ApprovedOMB No. 0704-0188

Public reporting burden for this collection of information is estimated to average 1 hour per response, including thetime for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, andcompleting and reviewing the collection of information. Send comments regarding this burden estimate or any otheraspect of this collection of information, including suggestions for reducing ~ " ' " "Services, Directorate for information Operations and Reports, 1215 Jefferson77?ri7-«0?, and to the Off ire of Management and Budget r Paperwork Reducf-- '

reducing this burden, to Washington Headquartersefferson Davis Highway, Suite 1204, Arlington, VAdnrtion Prompt r0704-nlfifi\ Washington, Be 70503.

1. AGENCY USE ONLY (Leave blank) 2. REPORT DATEAugust 1992

3. REPORT TYPE AND DATES COVEREDFinal Report

4. TITLE AND SUBTITLE

Crew Resource Management: An Introductory Handbook

6. AUTHOR(S)James E. Driskell, Richard J. Adams

7. PERFORMING ORGANIZATION NAME(S) AND AODRESS(ES)

Florida Maxima Corporation147 East Lyman AvenueWinter Park. FL 32789

Advanced Aviation Concepts, Inc.10356 Sandy Run RoadJupiter, FL 33478

9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES)U.S. Department of TransportationFederal Aviation AdministrationResearch and Development ServiceWashington, DC 20591

11. SUPPLEMENTARY NOTES

12a. DISTRIBUTION/AVAILABILITY STATEMENT

This document is available to the public through the NationalTechnical Information Service, Springfield, VA 22161

5. FUNDING NUMBERS

A1070/FA1EZ

8. PERFORMING ORGANIZATIONREPORT NUMBER

DOT-VNTSC-FAA-92-8

10. SPONSORING/MONITORINGAGENCY REPORT NUMBER

DOT/FAA/RD-92/26

12b. DISTRIBUTION CODE

13. ABSTRACT (Maximum 200 words)

Recent research findings suggest that crew resource management (CRM) training canresult in significant improvements in flightcrew performance. The objectives of thishandbook are to foster an understanding of the background and philosophy of CRM and toprovide an overview of the development, implementation and evaluation of CRM training.Currently, CRM programs have been implemented successfully at a number of airlines,large and small, civil and military. The variety of CRM training programs suggestthat there are a number of ways to achieve effective CRM.

14. SUBJECT TERMS

Crew Resource Management (CRM), Air Carrier Training, Flight Crew Performance, Air CarrierAccident Prevention, Team Coordination

15. NUMBER OF PAGES62

16. PRICE CODE

17. SECURITY CLASSIFICATIONOF REPORTUnclassified

NSN 7546-61-286-5500

18. SECURITY CLASSIFICATIONOF THIS PAGE

Unclassified

19. SECURITY CLASSIFICATION

OF ABSTRACT

Unclassified

20. LIMITATION OF ABSTRACT

Standard Form 298 (Rev. 2-89)Prescribed by ANSI Std. 239-18298-102

PREFACE

Recent research findings suggest that crew resource management (CRM) training canresult in significant improvements in flightcrew performance. The objectives ofthis handbook are to foster an understanding of the background and philosophy ofCRM and to provide an overview of the development, implementation and evaluationof CRM training. Currently, CRM programs have been implemented successfully ata number of airlines, large and small, civil and military. The variety of CRMtraining programs suggest that there are a number of ways to achieve effectiveCRM.

iii

METRIC/ENGLISH CONVERSION FACTORS

ENGLISH TO METRIC METRIC TO ENGLISH

LENGTH (APPROXIMATE) LENGTH (APPROXIMATE)1 inch (in.) o 2.5 centimeters (cm) 1 millimeter (mm) = 0.04 inch (in)1 foot (ft) a 30 centimeters (cm) 1 centimeter (cm) = 0.4 inch (in)1yard (yd) = 0.9 meter (m) 1 meter (m) = 3.3 feet (ft)1 mile (mi) = 1.6 kilometers (km) 1meter(m) = 1.1 yards (yd)

1 kilometer (km) = 0.6 mile (mi)

AREA (APPROXIMATE) AREA (APPROXIMATE)1 square inch (sq in. in2) = 6.5 square centimeters (cm2) 1 square centimeter (cm2) = 0.16 square inch (sq in, in2)1 square foot (sq ft, ft2) = 0.09 square meter (m2) 1 square meter (m2) = 1.2 square yards (sq yd, yd2)1 square yard (sq yd, yd2) = 0.8 square meter (m2) 1 square kilometer (kn2) = 0.4 square mile (sq mi, mi2)1 square mile (sqmi, mi2) = 2.6 square kilometers (km2) 1 hectare (he) = 10,000 square meters (m2) = 2.5 acres1 acre = 0.4 hectares (he) = 4,000 square meters (m2)

MASS - WEIGHT (APPROXIMATE) MASS - WEIGHT (APPROXIMATE)1ounce (oz) = 28 grams (gr) 1 gram (gr) = 0.036 ounce (oz)1 pound (lb) ° .43 kilogram (kg) 1 kilogram (kg) = 2.2 pounds (lb)

1 short ton o 2,000 pounds (lb) = 0.9 tonne (t) 1 tonne (t) = 1,000 kilograms (kg) ail short tons

VOLUME (APPROXIMATE) VOLUME (APPROXIMATE)1 teaspoon (tsp) = 5 milliliters (ml) 1 milliliter (ml) = 0.03 fluid ounce (fI oz)

1 tablespoon (tbsp) = 15 milliliters (ml) lliter (I) = 2.1 pints (pt)1 fluid ounce (fl oz) = 30 milliliters (ml) 1 liter (1) = 1.06 quarts (qt)

lcup(c) = 0.24 liter (1) 1 liter (1) = 0.26 gallon (gal)1 pint(pt) ° 0.47 liter (I) 1 cubic meter (m*) = 36 cubic feet (cu ft, ft*)1 quart (qt) = 0.96 liter (1) 1 cubic meter (m3) a 1.3 cubic yards (cu yd. yd))

1 gallon (gal) = 3.8 liters (I)1 cubic foot (cu ft, W) = 0.03 cubic meter (m*)1 cubic yard (cu yd, yd3) = 0.76 cubic meter (m')

TEMPERATURE (EXACT) TEMPERATURE (EXACT)((x-32){5/9>]°F = y°C [{9/5)y + 32]°C = x °F

QUICK INCH-CENTIMETER LENGTH CONVERSION

— | i f f f 5 6 7 8 9 11 1 1 1 1

)

CENTIMETERS 01234 5 6 7 89 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

QUICK FAHRENHEIT-CELSIUS TEMPERATURE CONVERSION

-40° -22° -4° 14° 32' 50° 68°

^ 1 1 1 1 1 1 186° 104° 122° 140° 158° 176° 194° 212'

I 1 1 1 1 1 1 !1 1 1 1 1 1 1C -40° -30° -20° -10° 0° 10° 20°

1 1 1 1 1 1 1 130° 40° 50' 60° 70° 80° 90° 100°

For more exact and or other conversion factors, see NBS Miscellaneous Publication 286, Units of Weights andMeasures. Price S2.50.SD Catalog No. CI 3 10286.

IV

TABLE OF CONTENTS

Section Paae

1. INTRODUCTION 1

Objectives 1CRM Background and Philosophy 2Principles of Crew Resource Management 8CRM Training 9Summary 10Overview of the Handbook 11

2. CRM SKILLS I3

Communication Processes and Decision-Making .... 16Communication 16Assertiveness 20Decision-Making 23

Team Building and Maintenance 24Leadership 24

Workload Management and Situational Awareness ... 28Workload Management, 28Stress Management 31

3. IMPLEMENTING CRM TRAINING 33

The Systems Approach to Training 33Step 1: Developing CRM Training 33Step 2: Implementing CRM Training 45Step 3: Evaluating CRM Training 48A Final Note on CRM Training 51

4. SUMMARY 53

GLOSSARY 55REFERENCE MATERIALS 59

VI

Chapter 1: Introduction

Aviation Safety! This concept is readilyembraced by everyone in the aviationcommunity from flight crews to support staffto management. This was not always the casewith Crew Resource Management (CRM). Theconcept of crew resource management has beenboth blessed and cursed by those in aviation.It has been cursed because the emphasis oncrew resource management is relatively new,and people often have a healthy, skepticalreaction to new ways of doing things.

But CRM is also blessed by many because ofwhat it can accomplish. Recent researchfindings suggest that crew resourcemanagement training can result in significantimprovements in flightcrew performance. Notsurprisingly, a growing number of people inthe aviation community, from airlinemanagement to flight crews themselves, haveembraced crew resource management as aneffective approach to reducing flight errorsand increasing aviation safety. Currently,CRM programs have been implementedsuccessfully at a number of airlines, largeand small, civilian and military.

Objectives of This HandbookThe objectives of this handbook are to fosteran understanding of the background andphilosophy of Crew Resource Management, andto provide an overview of the development,implementation, and evaluation of CRMtraining. This handbook is written for Part135 and Part 121 carrier operators andmanagement, and is designed to serve as asupplement to Advisory Circular 120-51 asrevised, Crew Resource Management.

CRM Background and PhilosophyIt is useful to distinguish between thephilosophy of crew resource management andthe implementation of crew resourcemanagement (CRM training). There is generalagreement within the aviation communityregarding the principles underlying CRM.Most agree on key CRM concepts and the needto focus on crew skills and performance.However, there is less consensus regardinghow to implement CRM training. In fact,various training programs have appeared whichmeet the specific needs of individual users.The variety of CRM training programs suggeststhat there are a number of ways to achieveeffective crew resource management.

What follows is a brief history of crewresource management, a discussion ofprinciples, and finally, an overview of CRMtraining.

We've been flying for over 90 years. Why CRMnow? The concept of crew resourcemanagement is not new. Anyone who thinksthat the Wright brothers did not makeeffective use of the resources at theirdisposal in 1903 at Kitty Hawk is certainlymistaken. Similarly, military and civilianpilot training programs have touched on CRMtopics for years. NASA's John Lauber recallsthe saying that if an idea is new, itprobably isn't good, and if it is good, itprobably isn't new. So, while the conceptsunderlying CRM are not new, what is new isthe heightened emphasis on crew resourcemanagement as one key to increased aviationsafety.

From the 1950s to the 1990s we have witnesseda steady decline in aviation accidents (seeFigure l). This decline in aviationaccidents has been attributed to betterequipment, better training, and betteroperating procedures. However, this happybig picture of system safety masks sometroubling data. As Figure 2 illustrates, asaccidents related to equipment weaknesseshave decreased, accidents attributed to humanweaknesses have increased. A comparison of

(0UJa:

<(0 Q.UJ UJ

5 °

d

rrUJa.(0HZUJooo<

<

ZZ<

(0z><o

zUJooo<U.Ozo

oQ.Ooro.UJ>

UJor

Figure1. Total accident rate forcommercial aircraft, worldwide,1959-1990.(Excludes sabotage, mffitary action, turbulence and evacuation Injuries.)(BoeingCommercial Airplane Group, 1991)

Figure 1 and Figure 2 suggests two points.First, Figure 1 indicates that after a sharpdrop in the 1960s, accident rates haveleveled off from 1970 through 1990. Second,the trends in causes of accidents illustratedin Figure 2 show that human error hasremained a major contributing factor inaviation accidents during these latter years.

HUMAN CAUSES

MACHINE CAUSES

TIME

Figure 2. Changes inaccident causal factors overtime. (InternationalCivil Aviation Organization, 1984)

Industry estimates of causal factors in aircarrier accidents are shown in Figure 3. Bya conservative estimate, well over 60% ofaircraft accidents have been attributed tocrew-related actions. In brief, it seemsthat the "human factors" contribution toaviation accidents may be a difficult problemto solve.

Percent of Total Accidents with Known Causes10 20 30 40 50 60 70

1959 - 1990

Last 10 years (1981 - 1990)

Figure3. Primary causal factors for commercial aircraftaccidents, worldwide, 1959-1990.(Boeing Commerdal Airplane Group, 1991)

Concern with the factors underlying theseaccidents led NASA researchers in the 1970sto conduct a series of interviews with linepilots to investigate their perceptions ofaviation mishaps. Charles Billings, GeorgeCooper, and John Lauber found that one mishapcomponent consistently mentioned by pilotswas inadequate training. Even moreinteresting, these researchers found that itwas not technical training that these pilotsfelt they lacked, but training in leadership,communication, and crew management. In otherwords, traditional training had done anexcellent job of imparting stick and rudderskills, but these pilots felt that they

needed more training in orew ooordination. Asubsequent analysis of jet transportaccidents between 1968 and 1976 revealed morethan 60 that involved problems with crewcoordination and decision making (Cooper,White, & Lauber, 1979).

These preliminary results, coupled with adogged determination to pursue answers toproblems that line pilots had identified,encouraged NASA researchers to conductfurther research and analysis. In oneclassic simulator study, B-747 flight crewswere observed in a highly realistic simulatedline trip from New York's Kennedy Airport toLondon (see inset: New York to London MinusOne Engine). During this tightly-scriptedscenario, an oil pressure problem forced thecrew to shut down an engine. The crew had todecide where to land the plane. Thisdecision was further complicated by ahydraulic system failure, bad weather, poorair traffic control, and a cabin crew memberwho demanded attention at the worst possiblemoments. Researchers found that there was awide variation in the performance of crewsduring this simulation. Most problems arose

New York to London Minus One Engine: The RuffellSmith Simulator Study.Because the scenario involveda high gross takeoff weight,followed by an engine shutdown with a subsequentdiversion, the crew needed to dump fuel to reduce theaircraft's weight to maximize landingweight. As in actualline operations, this was a very busy period. Inone case,after the captain decided to dump fuel, the captain and thefirst officer together decided that 570,000 pounds was thecorrect target landing weight. They reached the decisionwithout consulting the flight engineer or any aircraftdocumentation. The flight engineer then calculateda dumptime of 4 minutes 30 seconds, which the captainacceptedwithout comment even though itwas approximatelyone-thirdthe actual time required. Without prompting, the flightengineer recalculatedthe dump time to the nearlycorrectfigure of 12 minutes.Instead of dumping for12 minutes, however, the flightengineer stopped afteronly3 minutes, perhapsbecause hereverted to his original, erroneous estimate or because hemisread the gross weightindicator. Unsatisfied, he againstarted to recalculate, but the failure of the No. 3 hydraulicsystem interrupted him.During the next eight minutes, the flight engineerwassubjected to a highworkload, but then noticed thatthe grossweightwas toohighanddecidedto refigure the fuel. Duringthattime, he was interrupted further anddidnothing moreabout the fuel until the captain, noticing the grossweight

indicator at 647,000 pounds, decided to make anover-gross-weight landing. A minuteand a half later, theflight engineerrechecked the fuelas partof the landingchecklist and became concerned about the gross weight.He spent a minute and a half rechecHng calculationsandannounced that the aircraft's gross weight computer must bein error. Two minutes later, the simulator lands at 172 knotswith only25 degrees of flap: a 1,000 foot-per-minute descentabout 77,000 pounds over the correct weight, on a short, wetrunway.

Duringthe 32 minutes between the decision to dump fueland the landing, the flight engineer was Interrupted 15 timeswhile performingspecific tasks tailoring the amount of fuel tobe dumped Inrelationto the conditions and length of thelanding runway. Nine of the interruptionscame directlyorindirectlyfrom the captain, four from the cabin crew member,and two fromequipment problems. The flight engineer wasnever able to complete and verify his fuel calculations anddump times without interruption, either by a routinepartofstandardoperatingprocedure,or by a request fromthecaptain or the cabin crew member. He thus becameoverloadedand his workbecame fragmented. The captainfailedto recognize the situation and so did nothing to resolveit

(Lauber, 1964)

not from a lack of technical knowledge orskills, but from poor resource management.Crews whose performance included a high rateof errors did a poor job of communicating,setting priorities, and sharing workload.Crews making few errors did a better job ofmanaging available resources.

In a subsequent analysis of the cockpit voicerecordings from this study, Foushee and Manos(1981) found that those crews whocommunicated more and who acknowledged theexchange of information made fewer errors.

Is CRM Training Necessary?

Factors related to faulty crew performance account for well overhalfof aircarrier accidents.These include:

Eastern Airlines, Lockheed L-1011, Miami, Florida, December 29,1972.United Airlines, DC-8, Portland, Oregon, December 28,1978.Allegheny Airlines, Inc., BAC 1-11, Rochester, NY, July 9,1978.Air Florida. Boeing B-737. Washington, DC, January 13.1982.Air Illinois, Hawker Siddley 748-2A, Pinckneyville, Illinois, October 11.1983.Galaxy Airlines. Lockheed Electra-L-188C, Reno, Nevada, January 21,1985.Air Ontario, FokkerF-28, Dryden, Ontario, March 10,1989

This early work by Lauber, Cooper, Foushee,and many others culminated in the firstNASA/Industry Workshop on Resource Managementon the Flight Deck in 1979. This eventconverged the efforts begun by the militaryand by commercial carriers in this area.Subsequently, in the early 1980s, programswere developed and implemented by some aircarriers, including United Airlines, KLM, PanAm, Trans Australia Airlines, and others.Other events, such as the 1986 NASA/MACWorkshop on cockpit Resource ManagementTraining, and the biennial aviationpsychology symposia organized by Dick Jensenat Ohio State University, provided

opportunities to review progress in CRMprogram development.

The FAA officially recognized the value ofCRM types of training during this period byallowing a LOFT training period to be used asan approved period of training which could besubstituted for certain pilot's recurrentproficiency checks. More recently, SFAR 58,The Advanced Qualification Program (AQP),passed into law in 1990, has given greatlyexpanded latitude to air carriers with regardto training. One of the conditions of theAQP training option is that CRM training beincluded. It is projected that CRM may oneday be required in all formal aircrewcertification requirements.

The CRM concept has continued to evolve overthe last decade, guided by extensivefederal/university/industry research and bylessons learned from the implementation ofCRM programs at a growing number of airlines.FAA Advisory Circular 120-51 as revisedprovides a contemporary statement of CRMconcepts. This document underscores severalrecent developments in CRM:

• CRM has come to embody the entire flightoperations team, including the cabin crew,air traffic controllers, maintenance, andother groups that interact with thecockpit crew. A shift in terminologyreflects this emphasis: Cockpit ResourceManagement is now more appropriatelytermed Crew Resource Management.

• A second recent initiative is theintegration of CRM skills with traditionaltechnical flying skills. Whereas CRMprograms stress the acquisition ofcrew-related skills, it is thought thatthese skills should ultimately beintegrated with technical skills in thenormal training and evaluation process.In other words, both technical skills andCRM skills interact to determineperformance on the flightdeck.Accordingly, these skills should betrained and evaluated together as part ofthe total training program.

CRM programs have been in place for a timesufficient to allow a body of researchevaluating CRM training effectiveness toaccumulate. The research results indicateclear evidence of positive changes inaircrew performance following theintroduction of CRM.

This brief look at the background of CRM hasnecessitated the omission of many importantcontributions by many people. However, it isnoteworthy that CRM program development hasbeen driven by inputs from line pilots, notdreamed up in some ivory tower. Accordingly,CRM has become widely accepted within theaviation community.

Principles of Crew Resource Management

CRM is the effectiveutilization of all availableresources-hardware,software, andpersonnel-to achievesafe, efficient flightoperations.

CRM is defined as the effective utilizationof all available resources—equipment andpeople—to achieve safe, efficient flightoperations. Resources include autopilots andother avionics systems; operating manuals;and people, including crew members, airtraffic controllers, and others in the flightsystem. Therefore, the concept of effectiveCRM combines individual technical proficiencywith the broader goal of crew coordination,thus integrating all available resources toachieve safe flight.

The following principles are fundamental tothe CRM concept:

• Effective performance depends on bothtechnical proficiency and interpersonalskills.

• A primary focus of CRM is effective teamcoordination. The team encompasses theflight crew (cockpit and cabin),dispatchers, air traffic controllers,maintenance and others.

• CRM focuses on crew members' attitudes andbehaviors.

• Effective CRM involves the entire flightcrew. CRM is not simply a responsibilityof the captain, nor should CRM training beviewed as captain's training. Allcrewmembers are responsible for effective

CRM Training

management of the resources available tothem.

The acquisition of effective CRM skillsrequires the .active participation of allcrewmembers. Effective resourcemanagement skills are not gained bypassively listening to classroom lectures,but by active participation and practice,including the use of simulations such asLine-Oriented Flight Training (LOFT).

CRM training should be blended into thetotal training curriculum, includinginitial, transition, upgrade, andrecurrent training.

CRM training programs come in many forms.Limited CRM training programs are nowavailable off-the-shelf from various sources.Specific organizations develop CRM programsto meet their own particular needs andcorporate culture. Therefore, someonereviewing current CRM training programs islikely to find a variety of programs andprogram acronyms. These include:

• Flight Operations Resource Management(FORM) ,

• Flight Deck Management (FDM),

• Aircrew Resource Management (ARM),

• Aircrew Coordination Training (ACT),

• Flight Team Management (FTM).

This diversity reflects the difference insize, type of aircraft, mission, trainingfacilities, equipment, and financialresources of operators. Accordingly, nosingle training program is likely to meet therequirements of all operators.

All CRM training programs are built on theprinciples outlined above. FAA AdvisoryCircular 120-51 as revised may serve to builda consensus on program content by suggestingbasic CRM skills to be included in anyprogram of instruction. These skills aregrouped into three clusters:

Summary

The overall goal of CRMis the blending oftechnical skills andhuman skills to supportsafe and efficientoperation of aircraft.

1. communications and Decision Skills. Thiscluster of skills includes behaviors relatedto communications and decisionmaking,including:

• assertiveness

• communications

• decision making

• conflict resolution

2. Team Building and Maintenance Skills.This cluster focuses on human interaction andteam management skills including:

• leadership

• team management

3. Workload Management and SituationalAwareness. This cluster reflects skillsrelated to managing stress and workload,including:

• mission planning

• stress management

• workload distribution

These skills will be examined more closely inthe following chapter.

Crew resource management represents anapproach to improving aviation safety thatwas born of real life experiences of airlinepilots. They realized that technical skillalone was not enough to manage safely acomplex flight system. CRM emphasizes theeffective utilization of all resourcesavailable to the flight crew, includingequipment and people. In addition torespecting the importance of traditionalstick and rudder skills, CRM focuses on thoseother skills required for effective crewperformance. The overall goal of CRM is theblending of technical skills and human skillsso as to support safe and efficient operationof aircraft.

10

Research and experience have both shown thatthe best CRM training is like other effectivetraining - it will include three learningelements: awareness, practice, andreinforcement. CRM training should notfollow any single outline, however. It ismost effective when it is developed to meeteach user's unique set of needs.

Overview of the HandbookChapter 1 has introduced CRM. The followingthree chapters provide suggestions andexamples on what CRM is and how CRM trainingprograms can be provided.

Chapter 2 presents an explanation of thebasic CRM skills. This explanation citescases illustrating effective and ineffectiveutilization of these skills.

Chapter 3 provides guidance on developing,implementing and evaluating CRM training.

Chapter 4 provides a brief summary.

Following the text, a glossary is presentedas a key to the terms used in CRM training.Finally, a bibliography is included toprovide supplementary reference material.

11

12

Chapter 2: CRM Skills

The "Right Stuff" formodem-day flight operations includes bothindividual technical proficiency and crew resourcemanagement skills.

There are many skills required to fly anairplane safely. Some of these are referredto as technical or "stick and rudder" skills.Major categories of technical skills include:

Motor Skills: the physical control ofaircraft systems, aircraft attitude, andnavigation.

Procedural Skills: the execution ofstandard, abnormal, and emergency operatingprocedures.

Information Skills (Knowledge): the use ofinformation required to conduct safe airoperations in areas such as federalregulations, weather, and aircraft systems.

These skills constitute the technicalproficiency of crewmembers. As noted inChapter 1, these skills formed the primarybasis for the selection and training ofaviators for most of this century.

These skills are necessary for modernaircraft operations, but by themselves arenot sufficient to ensure safe flight. Inother words, these individual technicalskills must be paired with other crew-relatedskills to achieve safe flight operations.For example, it is not enough that acrewmember possesses the appropriatetechnical knowledge; each crewmember mustalso have the skills necessary to receive andto transmit information efficiently in thecrew setting—communication skills. Thecrewmember who tends to ignore input fromothers can be a hazard during normal flight,and can be disastrous in emergencyconditions. Therefore, technical skills mustbe integrated with other crew-related skills,defined in Chapter 1 as CRM skills, to ensuresafe flight.

CRM skills, those skills related to effectivecrew resource management, may be grouped intothe following categories:

13

Communication Processes and Decision Making:skills related to effective communicationsand crew decisions.

Team Building and Maintenance: skillsrelated to leadership/followership andmaintaining a supportive team environment.

Workload Management and situationalAwareness: skills related to operationalawareness, planning, and managing stress andworkload.

Figure 4 provides an overview of the skillsthat determine flight performance. Note thatFigure 4 indicates that both technical skillsand CRM skills are necessary for effectiveflight performance. This view is consistentwith the recent initiative to integratetechnical and CRM skills in flight operationsand training.

Furthermore, each cluster of skills presentedin Figure 4 is broken down into basic orprimary-level skills. For example, specificskills that compose the CommunicationsProcesses and Decision Making cluster includecommunication skills, assertiveness skills,and decision making skills. The three majorCRM skill clusters provide one convenient wayto classify CRM skills. However, it is theprimary-level skills that form the basis forCRM training.

CRM skills within each of the three skillclusters are described in the followingsections. Each skill will be describedbriefly. Synopses of NTSB accident reportswill illustrate how skills can effect crewperformance. The purpose of this chapter isto provide an overview of selected CRMskills, and to demonstrate the importance ofthese skills to flight safety.

14

H

Briefings/

Debriefings

Inquiry/Assertion

Crew

SelfCritique

ConflictResolution

Communication

Decisions

COMMUNICATION

PROCESSESAND

DECISION

MAKING

Is

Leadership

Concernfor

Operation

InterpersonalClim

ate

GroupClim

ate

Automation

Management

Preparation

Planning

Vigilance

Workload

Distribution

Distraction

Avoidance

TEAMBUILDING

AND

MAINTENANCE

WORKLOAD

MANAGEMENT

ANDSITUATIONAL

AWARENESS

CREWRESOURCE

MANAGEMENT

SKILLS

g'ssjsg:

VisualAcq.and

Interp.ofHazards

System

Control

AircraftControl

MOTOR

ACTIVITIES

SKILLS

INDIVIDUALANDCREW

BEHAVIORS

^

Emergency

Abnormal

FlightManeuvers

Communications

SystemOperation

PROCEDURAL

ACTIVITIES

SKILLS

TECHNICAL/

FLIGHTCONTROL

SKILLS

Figure

4.

Skills

underlying

effective

flight

operations

CompanyPolicies

Laws,Rules,Advisories

Security/Haz.M

aterials

AirTrafficControl

Weather

AircraftOperations

INFORMATION

BASE

KNOWLEDGE

Communication Processes and Decision Making.The first cluster of CRM skills includesthose related to effective communication anddecision making. Three primary-level CRMskills within this cluster are described inthe following: Communication, Assertiveness,and Decision Making.

Communication

One of the most significant variablesrelevant to crew performance is theinformation flow within the cockpit andbetween the cockpit and other sources. Theeffective transfer of information is acomplex process, and requires thatinformation be conveyed when needed,transferred clearly, attended to by thereceiver, understood and acknowledged by thereceiver, and clarified if needed. There arenumerous opportunities for breakdown in thisprocess.

The Aviation Safety Reporting System (ASRS)was created in 1976 by NASA and the FederalAviation Administration to provide a database for anonymous reports of aviationincidents. From the earliest months, itbecame obvious that common deficiencies inthe exchange of flight information werefrequently being noted in the reports to theASRS.

Billings & Reynard(1981) analyzed a largegroup of the incidentsreported to the ASRS,finding that over 70% ofthe reports containedevidence of error in thetransfer of information.One of the most commoncommunication problems(37% of the reportedincidents) was failureto initiate theinformation transfer

i ,- ir- :- tncorrpiete or Comm not NMracafcMf Eguprrcnecomm oirNM

messagerrndtonfern

rriAjnderslooo >«ua

Figure 5. Communications errors reported in theASRS (Billings &Reynard. 1961).

process. In most of these cases, the neededinformation almost always existed, but it wasnot made available to those who needed it.Another common problem (37% of the incidents)was inaccurate, incomplete, ambiguous, orgarbled messages. Other problems included

16

the failure to transmit the message at theappropriate time (13%). In 11% of the cases,the message was either not received or wasmisunderstood. Only 3% of the informationtransfer problems were attributed toequipment failure.

Foushee and Manos (1981) also reviewed theASRS data base to examine incidents involvingcommunications problems. They observed thefollowing communication problems:

35% of the reports cited problems dealing with poor understanding anddivision of responsibilities. Often, the lackof appropriateacknowledgments and cross-checking was a factor.16%were due to interference with pertinent cockpit communications byextraneous conversations between cockpit crewmembers or betweencockpit crewmembers and cabin crew.15% of the incidents were due to information which one or morecrewmembers believed they had transferred, but due to interference orinadequacy of the message, was not transferred successfully.12% reported a total lack of communication between crewmembers.Within this category, there were numerous examples of crewmembers notcommunicating regarding errors even when they had access to the correctinformation.

10% of the communication problems cited were due to overconfidence orcomplacency. Often, crewmembers assumed that everyone elseunderstood what was happening, when in fact, they did not.

This and related research suggests that:

• Overall, there is a tendency for crews whocommunicate more often to perform betterthan crews who communicate less.

• When more information regarding flightstatus is transferred, there are fewererrors related to system operation.

• Crews who frequently acknowledge commands,inquiries and observations tend to makefewer errors.

In general, effective communication issupported by the following behaviors:

• Convey information clearly, concisely, andin a timely manner.

• Use standard terminology.

17

Advocate concerns and suggestions clearlyand assertively.

Acknowledge communications.

Provide information as required.

Repeat information.

* Ask for clarification when needed.

• Resolve conflicts constructively.

The following accident summary illustratesthe importance of these behaviors.

ACCIDENT SUMMARY - OnJuly19.1989. at about 3pmlocal time, a DC-10 operated byUnited Airlines as flight 232,experienced a catastrophic failure of the No. 2 tailmountedengineduring cruise flight. Shortly afterthe enginefailure, thecrewnotedthat the hydraulic fluid pressureandquantity hadfallen to zero inall threeredundant hydraulic systems. Theengine failure precipitated damage that severed the threehydraulic systems, leaving the flight control systemsinoperative. Approximately oneminute afterthe enginefailure,the flightdata recorder indicated no further movement of theflight control surfaces.

The onlymeans of control for the flight crewwas from theoperating wingmounted engines. The application ofasymmetricpowerto these engines changed the roll attitude,hencethe heading. Increasing anddecreasing power hadalimited effect on the pitch attitude. The airplane tendedtooscillate aboutthe center of gravity inthe pitchaxis, itwas notpossible to control the pitch oscillations with anydegree ofprecision. Moreover, becauseairspeed Isprimarily determinedby pitchtrimconfiguration and power, therewas no directcontrol of airspeed. The crew found thatdespite theirbestefforts, theairplane would not maintain a stabilized Rightcondition. The airplane subsequently crashedduring anattemptedlanding at SiouxGatewayAirport, Iowa. Therewere265passengers and 11 crewmembers onboard. One flightattendant and 110 passengerswere fatally injured.EVENT HISTORY - About 1 hour and7 minutes after takeoff.theflight crew heard a loud bang oranexplosion, followed byashuddering of the airframe. The following sequenceofeventsis inchronological order and is presentedto summarizethetypeandvariety of communications required.1. The flight crew determined that theNo. 2aft(tail mounted)engine hadfailed. The captain called for the engine shutdownchecklist. While shutting downthe engine, the secondofficer(flight engineer) observed that the systemshydraulic pressureand quantity gauges indicated zero.2. The first officer advised that he could not control theairplane as itentered a right descending turn. Thecaptain tookcontrol oftheairplane and confirmed that itdid notrespond toflight control inputs.

3. Thecaptain reduced thrust tothe No. 1engine andtheairplane beganto roll to a wingslevel attitude.4. A flight attendantadvised the captainthat a UALDC-10training checkairman wasseated inthepassengercompartment andhadvolunteered hisassistance. Thecaptainimmediately invited the airman into the cockpit.5. At the requestof the captain, the check airman re-enteredthe passenger cabin andperformed avisual inspection of theairplane's wings. He returned and reported thatthe inboard

ailerons wereslightly up, notdamaged,and thatthe spoilerswere lockeddown. Therewas nomovementof the primaryflightcontrol surfaces.6. The captaindirectedthe check airmanto operatethethrottles to freehimselfand the first officerto attempttomaintain command of the flight controls. The check airmanadvisedthat the No. 1 and No.3 engine thrust levers couldnotbe used symmetrically, so he used twohands to manipulatethe throttles. Even so, he said that the airplane had acontinuoustendency to turnright and Itwas difficult to maintaina stable pitch attitude.7. The captainreported to the approachcontroller that theflight had no elevatorcontrol, they mighthave to make a forcedlanding andaskedthecontroller for the ILS frequency, headingto the runwayand lengthof the runway. Hethen instructedthesecondofficer to startdumping fuel usingthe quickdump.8. The captain asked the senior flight attendant Ifeveryone inthe cabinwas ready. She reported Inthe affirmative and thatshe observeddamage on one wing. The captainsent thesecond officerback to inspect the empennage visually.9. The secondofficer returned andreported damage to theright andlefthorizontal stabilizers. The captain replied "that'swhatIthought." The captain thendirected the flightcrew tolocktheir shoulder harnessesandto puteverything away.10. Several seconds later, the controller alerted thecrewmembers to a 3,400 foot tower obstruction located 5 milestotheir right andaskedhow steepa right turn theycould make.The captain responded thattheyweretrying to makea 30degree bank. A crewmember commented that "I cant handlethatsteepof bank." The first officer stated, "we're gonna haveto try it straight ahead Al..."11. The captain reported the runway Insightand thankedthecontroller for his help. The controller statedthatthe runway theflight had linedup withwas closed, but he added "that'll worksir. we're getting the equipment offthe runway." The captainasked its length andthe controller reported 6.600 feet. Twelveseconds later the controller statedthattherewas an open fieldat the end of the runwayand that the winds wouldnot be aproblem.

12. During the final 20 seconds beforetouchdown, theairspeed averaged 215knots, sinkrate was1.620 feet perminute andsmoothoscillations in pitch and roll continued. Thecaptain recalled getting a high sinkrate alarm from the groundproximity warning system and that at 100 feet above theground, the nose of the airplane beganto pitchdownward.First contact wasmadebythe right wing tip followed bytheright main landing gear. Theairplane skidded tothe right ofthe runway, ignited,cartwheeled and came to rest in aninverted position. (NTSB, 1990a)

18

The quality and efficiency of the crewcommunications in UAL flight 232 is onefactor that minimized the loss of life inthis catastrophe. Specific communicationsbehaviors such as clarity (event #2),conciseness (event #5), timeliness (event #7)and acknowledgement (event #10) arerepresented in the event history.

A more intensive analysis of crewcommunication during this accident wasperformed by Predmore (1991), who broke downthe Cockpit Voice Recorder (CVR) transcriptinto the following communication categories:

• Command-Advocacy (CMD-ADVOC)

• Inquiry

• Incomplete-Interrupted (INCOMPL)

• Reply-Acknowledge

• Observation

The following chart represents the cockpitand radio communications from flight 232during this emergency.

26 28 30 32 34 36 38 40 42 44 46 48 50 62 54 56 58 60CVR TIME (In mlmiUi)

0CMD/ADVOC H OBSERVATION • INQUIRY• INCOMPL ESrEPLY/ACKNOWL

Figure 6. Communications of the UAL flight 232crew. (Predmore, 1991)

Figure 6 reveals that the crew maintained aremarkably high level of communicationoverall. A second important factor is theconsistent level of acknowledgement evidentduring this event. It is important to notethat the crew of United 232 had receivedprior CRM training.

19

Assertiveness

Accident reports reveal a number of instancesin which crewmembers failed to speak up evenwhen they had critical flight informationthat might have averted a disaster. In mostcases, this hesitancy involved a copilot orflight engineer who failed to question acaptain's actions or to express an opinionforcefully to the captain. These types ofincidents lead to the conclusion thatcrewmembers are often unwilling to state anopinion or to take a course of action, evenwhen the operation of the airplane is clearlyoutside acceptable parameters.

NASA's H. Clayton Foushee reported anincident from the ASRS data base whichillustrates this phenomenon. This reportdescribed a situation in which air trafficcontrol had instructed the aircraft to leveloff at 21,000 feet. As the aircraft reachedits assigned altitude, the copilot noticedthat the captain was allowing the airplane tocontinue climbing. The copilot alerted thecaptain, but not forcefully enough for thecaptain to hear. The copilot tried again andpointed to the altimeter, at which point thecaptain stopped the climb and began descentback to the assigned altitude. The copilotsummed up the reasons for his actions asfollows:

The captain saidhe hadmisreadhis altimeter and thought he was 1000feet lower than hewas. Ibelieve themain factor involvedherewasmyreluctance to correct the captain. This captain is very "approachable''and Ihad no reason to holdback. ItIsjust a bad habitthat Ithink a lot ofcopilots haveofdouble-checking everything we say before we sayanything to the captain. (Foushee, 1982, p. 1063)

Assertiveness involves the ability to requestinformation from others, make decisions, andcarry out a course of action in a consistentand forceful manner. Assertive behaviorincludes:

• Inquiry: inquiring about actions taken byothers and asking for clarification whenrequired.

• Advocacy: the willingness to state what isbelieved to be a correct position and toadvocate a course of action consistentlyand forcefully.

20

• Assertion: stating and maintaining acourse of action until convinced otherwiseby further information.

Crewmembers are often hesitant to speak upfor several reasons:

• Sometimes crewmembers fail to questionothers' behavior because they are hesitantto point out incompetent behavior and toembarrass a captain or crewmember. Komich(1985) cites one copilot's hesitancy tocorrect a captain: "He's slow to catch hisairspeed and if I speak up every time he'sten knots low, it'll sound like aninstructional ride, so since he usuallycatches it at ten, I'll speak up atfifteen." Others have expressed fear ofcausing animosity and possibly creating areputation as that of someone who is"difficult to work with."

• Crewmembers sometimes hesitate to speak upbecause they perceive the captain as toointimidating. This kind of captain seeshimself as omnipotent, with the othercrewmembers there only to serve hiswishes, not to make any contributions tothe decision-making process. Foushee(1982) cites one extreme case:

Air traffic control had issued a speedrestriction. The copilot acknowledged andwaited for the captain to slow down.Assuming that the captain hadn't heard themessage, he repeated, "Approach said slow to180." The captain's reply was, "I'll do whatI want." Air traffic control inquired as towhy the aircraft had not been slowed, advisedthe crew that they had nearly collided withanother aircraft, and issued a new clearance,which the captain also disregarded. Followinga further advisory from the copilot, thecaptain responded by telling the copilot to"just look out the damn window."

The status structure of the cockpit maycontribute to non-assertiveness amongcrewmembers. The captain has authority inthe cockpit and the responsibility forflight operations. People are naturallyhesitant to question those who have higher

21

status. Therefore, people tend to bedeferential to those in command. Second,crewmembers may simply assume that, sincethe captain is in charge, that the captain"knows what he is doing."

However, a lack of assertive behavior on thepart of crewmembers may have disastrousconsequences, as the following incidentillustrates.

ACCIDENT SUMMARY - About 1815 Pacific StandardTime on December 28,1978, United AirlinesFlight173, aDC-8-61 aircraftcrashed intoa wooded area during anapproachto the Portland International Airport. The aircrafthad delayed southeast of the airportforone hourwhile theflightcrewcoped with a landinggear malfunctionand preparedthe passengers foran emergency landing.The NTSB determined the probable cause of the accident wasthe failure of the captainto monitorproperty the aircraft's fuelstate and to properly respond to the crewmembers' advisoriesregarding fuel. Contributingto the accident was the failure ofcrewmembers either to fully comprehend the crtticality of thefuel state or to assertively communicate their concerns to thecaptain.EVENT HISTORY - The first problem faced by the captainwas the unsafe landing gear indication duringthe initialapproachto Portland International Airport. This indicationfolloweda loud thump, an abnormal vibration, and an abnormalaircraft yaw as the landinggear was lowered.1. At 1712:20, Portland Approachrequested,"United oneseven three, contact the tower,one one eight pointseven."The flight responded. "Negative, we'll staywithyou. We'll stayat five. We'll maintain abouta hundredand seventy knots. Wegot a gear problem. We'll let you know."2. At 1746:52, the first officerasked the flight engineer,"Howmuch fuel we got...r The flight engineerresponded,"Fivethousand." The first officeracknowledged the response.3. At 1748:00. the first officerasked the captain,"... what'sthefuel show now...?" The captainreplied, "Five." At 1749, aftera partially unintelligible comment by the flight engineerconcerningfuel pump lights, the captainstated, "That's aboutright, the feed pumpsare starting to blink." At this point,according to air traffic control data, the aircraftwas about 13nauticalmiles south of the airport.4. About1750:20, the captain askedthe flight engineer to"Giveus a currentcardonweight. Figure aboutanotherfifteenminutes." The first officerresponded,"Fifteen minutes?" Towhichthe captain replied, "Yeah,giveus threeor fourthousand poundson topof zero fuel weight." The flightengineer then said,"Notenough. Fifteen minutesis gonnareallyrun us low on fuel here."5. From 1752:17 to about 1753:30. the flight engineer talked toPortland and discussed the aircraft's fuelstate, the numberofpersonsonboard the aircraft andthe emergency landingpreparations at the airport.6.At 1756:53, the first officer asked,"Howmuch fuel yougotnow?" The flight engineer responded that4,000poundsremained, 1,000 pounds in each tank.7. From 1757:30 until 1800:50, the captain and the first officerengaged inconversation which included discussions ofgiving

the flight attendants ample time to prepare forthe emergency,cockpit procedures in the event of an evacuationafter landing,and the procedures the captainwould be using duringtheapproach and landing.8. AT 1802:44. the flightengineer advised, "We got aboutthree on the fuel and that's It." The aircraft was about 5nauticalmiles south of the airporton a southwest heading.9. At 1802:44. PortlandApproach asked Flight173 fora statusreport. The first officer replied,"Yeah, we have indicationourgear is abnormal, nil be our intention, in about five minutes, tolandon two eight left. We would likethe equipment standingby. Our indications are the gear is down and locked. We'vegot our people prepared foran evacuation In the event thatshould become necessary."10. At 1803:14, Portland Approach asked that Flight 173advise them when the approach would begin. The captainresponded, "They've about finished in the cabin. I'dguessabout another three, four, five minutes." At this time, theaircraft was about 8 nautical miles south of the airport.11. At 1806:19.the first flight attendantentered the cockpitThe captainasked, "Howyoudoing?" She responded,"Wed, Ithinkwe're ready." At this time, the aircraft was about 17nautical mites south of the airport. Almost simultaneouslythefirstofficer said, "I think you just tost number four followedimmediately by adviceto the flight engineer,"Betterget somecrossfeeds open there."

12.At 1806:46, the first officer toldthe captain, "We'regoing tolose an engine" The captainreplied, "Why?" at 1806, the firstofficer again stated,"We'relosing anengine." Again thecaptainasked, "Why?" The first officerresponded, "Fuel."13.At 1807:12. the captain calledPortland Approachandrequested," Would likeclearance foran approach Intotwoeight left, now." The aircraftwas about 19 nauticalmitessouth-southwest of the airport and turningleft. This was thefirst requestfor an approach clearance from Flight 173.14. At 1813:21, the flight engineer stated, "We've lost twoengines, guys." At 1813:25.he stated, "We Justlost two

one and two."

15. At 1813:38, the captain said,They're allgoing. We canlmakeTroutdale." The firstofficer said, "We can't makeanything."16. At 1813:46, the captain toldthe first officer, "Okay.Declare a mayday." At 1813:50,the first officercalledPortlandInternational Airport Towerand declared."Portland Tower,United oneseventythree heavy, Mayday. We're- theenginesare flaming out. We're going down. We're notgoing to be ableto make the airport" This was the last radio transmission fromFlight 173.(NTSB, 1979)

22

In this accident, the flight engineer wasincreasingly concerned about the criticalfuel situation, making several observationsto the captain that were not acknowledged. Inits report, the NTSB stated:

Admittedly, the stature of a captain and his management style may exertsubtle pressure on his crew to conform to his way of thinking. Itmayhinder interaction and adequate monitoring and force anothercrewmember to yield his right to express an opinion. (NTSB, 1979)

Decision MakingDecision making is a topic that may at firstglance seem to be an individual matter.After all, the captain.is the final authorityand responsible for flight decisions.However, aircrew decision making is a groupprocess, and clearly illustrates thecollective nature of crew resourcemanagement. There are a number of hardware,software, and human resources available inthe cockpit, including other crewmembers,ATC, dispatch, and various sources ofinformation. The decision maker who does notrely on input from other crewmembers and fromother flight team members outside the cockpitis more likely to make poor decisions.

Although decisions are certainly founded onaeronautical knowledge, flying skills, andexperience, it is often difficult to describehow decisions are made in actual flyingsituations. It has generally been assumedthat learning to make good decisions could beattained only through experience. However,research has shown that aircrew decisionmaking skills can be shaped through training.

The decision making process may be brokendown into the following five steps:

1. Recognizing or identifying the problem.Does a problem exist that requires action?

2. Gathering information to assess thesituation. This step requires determiningwhat information is needed, who has theneeded information, and whether theinformation is verified by othercrewmembers and resources.

3. Identifying and evaluating alternativesolutions. This step includes evaluating

23

the advantages as well as the risksassociated with each alternativeidentified, and selecting the optimumalternative.

4. Implementing the decision. This stepincludes executing the decision andproviding feedback on actions taken tocrewmembers.

5. Reviewing consequences of the decision.This step involves evaluating theconsequences of the decision and revisingthe decision if consequences are not asanticipated.

Some decisions, especially those that must bemade under extreme time pressure, must beseat-of-the-pants decisions. In these cases,there is very little time to gather allavailable information or to evaluatealternative solutions. These situations callfor intuitive decision making, which is basedon gut reactions, or more specifically, isbased on past experience and training.However, these emergency situations arerelatively rare. Host situations allowsufficient time to make a more deliberate oranalytical decision. This decision moreclosely follows the steps outlined above.Analytical decision making uses the resourcesavailable to the decision maker and resultsin more informed decisions.

Team Building and MaintenanceTeam Building and Maintenance skills includethose skills related to fostering effectiveteam performance.

LeadershipThe term "leadership" implies that this skillis relevant only to the captain. There aretwo reasons why this is not true. First, aflightcrew is a team with a clearlydesignated leader: the captain. The captainas designated leader retains the authorityand responsibility for flight operations.However, there are times when othercrewmembers must play functional leadershiproles. A functional leader may carry outleadership duties for a specialized task on atemporary basis, such as a takeoff or

24

landing. In this case, the crewmember mustdirect task activities and serve as afunctional leader to carry out that task.

Second, leadership would more properly becalled leadership/followership. Leadershipis a reciprocal process, and there arebehaviors that both a leader and a followermust apply to ensure effective performance.For example, one leader behavior might be toprovide direction for carrying out a task;correspondingly, one follower behavior mightbe to provide feedback on performance of thetask. In other words, leader behaviors areless effective without complementary followerbehaviors.

Leadership is not just "captain's" material.All crewmembers must perform leadershipduties in some situations. Furthermore,leadership is not a one-way process, butrequires both leader actions and effectivecrewmember responses.

Understanding the leadership role requires anunderstanding of what it is that leaders do.Effective leaders perform four primaryfunctions:

1. Regulating Information Flow. The leadermust regulate, manage, and direct the flow ofinformation, ideas, and suggestions withinthe cockpit crew and between the cockpit crewand outside sources. This function includesthe following behaviors:

• Communicating flight information

• Asking for opinions, suggestions

• Giving opinions, suggestions

• Clarifying communication

• Providing feedback

• Regulating participation

2. Directing and Coordinating CrewActivities. The leader must function as crewmanager to provide orientation, coordinationand direction for group performance. Thisfunction includes:

• Directing and coordinating crew activities

• Monitoring and assessing crew performance

• Providing planning and orientation

25

• Setting priorities

• Delegating tasks

3. Motivating crewmembers. The leader mustmaintain a positive climate to encourage goodcrewmember relations and to invite fullparticipation in crew activities. Thisfunction includes:

• Creating proper climate

• Maintaining an "open" cockpit atmosphere

• Resolving/preventing conflict

• Maintaining positive relations

• Providing non-punitive critique andfeedback

4. Decision-making. The leader isultimately responsible for decisions. Thisfunction includes:

• Assuming responsibility for decisionmaking

• Gathering and evaluating information

• Formulating decisions

• Implementing decisions

• Providing feedback on actions

The following excerpt from an NTSB accidentreport illustrates the errors that can occurwhen certain leadership and followershipbehaviors are applied poorly or not at all.

ACCIDENT SUMMARY - On October 11.1983. Air IllinoisFlight 710, a HawkerSiddley 748-2A was being operatedbetween Chicago, Illinoisand the Southern IllinoisAirport,Carbondale, Illinois,with an intermediate stop at Springfield,Illinois. At 2020 central daylight time (CDT), Flight710departed Springfield with seven passengers and threecrewmembers on board. About 1.5 minutes later, Flight 710called Springfield departure control and reported that Ithadexperienced a slight electrical problembut that itwascontinuing to Its destination about 40 minutes away.The flight toward Carbondalewas conducted in instalmentmeteorologicalconditions. The cloud bases in the areawere at2,000 feet MSL with tops at 10,000 feet. Visibility below thecloud bases was 1 mile in rain, and there were scatteredthunderstorms in the area.

The CockpitVoice Recorder transcriptshowed that shortlyafter takeoff, Flight 710s left generatorsuffered a completemechanical failure and that in responding to the failure of theleft generator, the firstofficermistakenly isolatedthe right

generator from the airplane's d.c. electrical system. Allsubsequent attempts to restorethe rightgeneratorto theairplane'sd.c. distribution systemwere unsuccessful, and theairplane proceededtoward Carbondale relying solelyon itsbatteries for d.c. electrical power.EVENT HISTORY - The flightwas about 45 minutes behindschedule when it arrived at Capitol Airport, Springfield, Illinois.The flightcrew remained on boardwhile the airplanewasfueled. At 2011:44, when Flight710 requested its IFRclearance, Italso requested 5,000 feet forits enroute attitude,as opposed to the IFR flight planof 9,000 feet stored in theARTCC computer. At 2019:40, Springfieldtower clearedFlight710 for takeoff.1. At 2021:14, Flight 710 contacted departure controlandinformed the controllerthat Itwas climbing through 1,500 feet.The departure controlleradvised the flightthat he had it Inradar contact, cleared It to climb to and maintain 5,000 feet,and clearedit to proceeddirectto Carbondale after it receivedthe Carbondale VOR.

26

2. At 2021:34, Flight710 informed the departure controller thatit had experienced a "slight electrical problem..." The controllerasked the flight if itwas going to return to Springfield, and theflight reported that it did not intend to do so.

3. At 2022:10, the flight told departure control that "We'd liketo stay as lowas we can," and then requested and was clearedto maintain 3,000 feet. The controllerasked the flight Ifhecould provide any assistance, and the flight responded, "We'redoing okay, thanks."4. At 2023:54, the first officer told the captain that "the left(generator) is totally dead, the right(generator) is puttingoutvoltage but Icant get a toad on It." About 30 seconds later, hereported "zero voltage and amps on the left side, the right isputtingout 27.5 volts but Icant get it to come on the line." At2025:42, he told the captain that the battery powerwas goingdown "pretty fast."5. At 2027:24, the captain called Kansas City center andstated that he had an "unusual request." He asked clearanceto descend to 2,000 feet "even ifwe have to go VFR." He alsoasked the controller "to keep an eye on us Ifyou can." Thecontroller told the flight that he could not clear it to descend.The captain thanked the controller and continued to maintain3,000 feet.

6. At 2028:45, the captain said, "Beacons off... and Nav lightsareoff." At 2031:04, the firstofficer reminded the captain thatCarbondale had a 2,000 foot celling and that the visibility was 2mileswithlightrain and fog. There was no replyoracknowledgement from the captain.7. At 2033:07, the flightattendant came forward and thecaptain asked her if she could work with what she "had backthere." The flight attendantreported that the onlylightsoperating in the cabinwere the reading lights, the lightsby thelavatory, the baggage light and the entrance lights. The captain

instructedherto briefthe passengers that he had turned off theexcess lightsbecause the airplane had experienced"a bitof anelectrical problem..."but that they were going to continue toCarbondale.

8. At 2038:41 (17 minutes after the initial failure), the firstofficer told the captain, "Well, when we... started losing the leftone I reached up and hit the right(isolate button) tryingtoisolatethe rightside because Iassumed the problemwas theright side but they (the generators) both still went off."9. At 2044:59, in response to the captain's request, the firstofficer reported that the battery voltage was 20 volts. At2049:23, Kansas City center requested Flight710 to changeradio frequencies. The flight acknowledged the request, whichwas the last radio communication from Flight710.10. At 2051:37. the first officer told the captain, "I donl know ifwe have enough juice to get out of this." At 2052:12, thecaptain asked the first officer to "watch my altitude, I'mgoing togo down to twenty-four hundred (feet)." He then asked the firstofficer if he had a flashlightand to have it ready. At 2053:18,the first officer reported, "We're losing everything, ...down toabout thirteen volts," and at 2053:28, he told the captain theairplane was at 2,400 feet11. At 2054:00, the captain asked the first officer if he had anyinstruments. The first officer asked him to repeat, and at2054:16, the captainasked "Doyou have any instruments, doyou have a horizon (attitude director indicator)?" Flight 710crashed near Centralis, Illinois VORTAC located about 40nautical mites north of the Southern Illinois Airport. Threecrewmembers and seven passengers were killed in the crash.(NTSB, 1984)

This accident involved an HS-748-2A aircraftwhich experienced a generator failure atnight. Proper procedures were not followed,causing disconnection of the secondgenerator. A series of poorly managedactions followed, including an attempt tomake the destination on battery power alone.Several behaviors related toleadership/followership are identifiable:

• Poor monitoring and assessment of crewactivities (Event 8)

• Little feedback or acknowledgement ofactions (Event 6, 8)

• Utilization of critical informationunknown (Event 6)

Furthermore, a crucial error occurred whenthe co-pilot reminded the captain of IFRweather at the destination, got no responsefrom the captain, and did not press the issueuntil it was too late.

27

This accident shows that it is sometimesdifficult to apply vital team behaviorsrequired in the cockpit such as leadership,planning, problem solving, delegating,motivating, and setting priorities. Thisdifficulty reinforces the importance of CRMtraining.

Workload Management and Situational AwarenessThese skills reflect the extent to whichcrewmembers maintain awareness, prepare forcontingencies, and manage workload andstress.

Workload ManagementWorkload management includes preparation,vigilance and avoidance of distractions andcomplacency. Pilots interviewed aboutworkload management offered the followingtips:

1

Preparation - "Commit SOP' (Standard Operation Procedures), limitations andemergency procedures to memory, to free up mental capacity to deal withunforeseen events."

Planning - "Before each flight, I typically spend about one hour at homereviewing the route and airport information."

Vigilance - "Be especially vigilant when everything is going well." and "Neverassume anything, but verify and cross-check all critical information."

Complacency - "Avoid complacency. The minute you think something wont hurtyou, it will."

Distractions - "Maintain a terrain awareness and a general knowledge of thetopography over which you are flying."

(from Kelly, 1991)

Workload varies according to the phase of anoperation, from the routine of preflightplanning and enroute cruise to the highworkload of a low visibility instrumentapproach. Either workload level can bedangerous.

28

4.2%

c

%

12.9%

Takeoff

1%

Accidents often occur when workload demandsexceed crew capabilities. Figure 7illustrates the phases of flight in whichmost accidents occur. Note that takeoff,approach, and landing phases account for mostaircraft accidents. These phases are alsoperiods of high crew workload. If anydistraction or irregularity occurs duringthese phases, an accident is much more likelyto occur than at other times.

8.5%

InitialClimb

Percent of Accident*

6.4%

Climb

5.7%

Cruise

13% | 60%

Exposure - Percent of Flight Time

6.8%

Descent

7.0%

InitialApproach

HoldingPattern

21.3% 27.2%

I

Fnure 7. Percent oftotal commercial aircraft accidents byphase of flight,worldwide, 1959-1990. (Boeing Commercial AirplaneGroup, 1991)

Paradoxically, low workload can also be ahazard to safety. Crews may be less alertduring long cruise segments. These lowworkload periods are times when complacency,forgetfulness and drowsiness are most common.Examination of the errors associated with lowand high workloads reveals that performancefollows a YERKES DODSON arousal curve likethe one depicted in Figure 8.

29

Higlj

8c(0

(OPTIMAL AROUSAL)

0.

(UNDERSTIMULATED) (OVERSTIMULATED)

LowArousal

High

Figure 8. Very low or very high arousal can degrade performance.

The following accident emphasizes therelationship between arousal and performance.Crucial elements identifiable in this excerptinclude the importance of unbroken attentionto instrument scan and the insidious role ofdistractions.

ACCIDENT SUMMARY - On December 29,1972, anEastern Airlines Lockheed L-1011 crashed in the Evergladesabout 18 miles west-northwest ofMiami International Airport(MIA). There were nomajor mechanical problems, severeweather phenomena orcrew incapacitation. The flight divertedfrom its approach because the nose landing gear positionIndicating systemoftheaircraft did notindicate that thenosegearwas locked inthe downposition. The aircraft climbed to2.000 feet MSL and followed aclearance to proceed west fromthe airport at that altitude. During that time the crew attemptedtocorrect themalfunction andtodetermine whether ornotthenose landing gear was extended. Unfortunately, during thatperiod, workload management in terms offlying, navigating andcommunicating was totally ignoreddue to fixation on therelativelyminor failure.

The flight wasconducted inclear weather conditions withunrestricted visibility. However, the accidentoccurred indarkness with nomoon. The flight wasuneventful until theapproach toMIA. The landing gear handle was placed in the"down" position during the preparation for landing, and thegreen light, which indicates to the crew that the landing gear isfully extended and locked, failed to illuminate. The captainrecycled thelanding gear, but thegreen light still failed toilluminate.

TheNational Transportation Safety Board determined that theprobable cause ofthis accident was failure oftheflight crew tomonitor the flight instruments during the final four minutes offlight, and todetect an unexpected descent soon enough toprevent impact with the ground. Preoccupation with amalfunction ofthe nose gear position indicating systemdistracted the crew'sattention from the instruments andallowed the descent to go unnoticed.EVENT HISTORY - Eastern Airlines Flight 401 was ascheduled passenger flight from theJohn F. Kennedy

International Airport InJamaica, NewYorkto the MiamiInternational Airport inMiami. Florida.1.At2334:05, EAL 401 called theMIA tower and stated, "Ah.tower, this is Eastern, ah, four zeroone, it lookslike we'regonna have tocircle, wedon't have a light onour nose gearyet"

2. At 2334:14, the tower advised, "Eastern four oh one heavy,roger, pull up, climb straight ahead totwo thousand, goback toapproach control, onetwenty eight six."3.At2335:09, EAL 401 contacted MIA approach control andreported, "All right, ah,approach control, Eastern four zeroone, we're right over theairport here and climbing totwothousand feet, in fact, we'veJust reachedtwothousand feetand we've got togetagreen light onournosegear."4.At2336:04. thecaptain instructed the first officer, who wasfrying theaircraft, toengage theautopilot. The first officeracknowledged the instruction. Subsequently, the captain tookover the frying responsibilities. The first officer successfullyremoved thenose gear light tens assembly, but itjammedwhenhe attempted to replace it.5. At 2337:06, the captain instructed the second officertoenter the forward electronics bay, below the flight deck, tocheck visually thealignment ofthenosegear Indices.6.At2337:48. approach control requested the flight toturn lefttoa heading of 270degreesmagnetic. EAL 401acknowledged the request and turned tothe new heading.Meanwhile, the flightcrew continued their attempts to free thenose gearposition light lens from its retainer, without success.At 2338:34, the captain againdirected the second officertodescendInto the forward electronics bayandcheckthealignment of the nose gear indices.

30

7. At 2340:38, a half-second C-chord, which indicated adeviation of +/- 250 feet from the selectedaltitude, soundedinthe cockpit. No crewmember commented on the C-chord. Nopitch change to correct the loss ofaltitude was recorded. Ashort timelater, thesecond officer raised hishead into thecockpit and stated, "I cant see it, ITs pitch dark and Ithrow thelittle light. Iget. ah. nothing." The flightcrew and an EALmaintenance specialist whowasoccupying theforwardobserver seat thendiscussed the operation of the nosewheelwell light. Afterward, the specialist went Into theelectronics baytoassist the secondofficer.8.At2341:40, MIA approach control asked, "Eastern, ah, fourohonehow are things coming" along outtherer Thisquerywasmade a few seconds after thecontroller noted analtitudereading of900 feet In the EAL alphanumeric block on his radardisplay. The controller later testified that rr«rnentary deviationsIn altitude information on the radar displaywerenot

uncommon; and that more that onescanonthedisplay wouldberequired toverify adeviation requiring controller action.9.At2342:05, the first officer said. "We did something toaltitude." The captain's replywas "WhatT The first officerasked, "We're still attwo thousand, rightr The captainimmediately exclaimed, "Hey. whafs happening hereT10. At2342:10, the first ofsixradio altimeter warning "beep"sounds began; they ceased Immediately before the sound ofinitial ground impact. The aircraft crashed while In aleft bankof28degrees. The aircraft was destroyed. Therewere163passengers and acrewof13 aboard the aircraft Ninety-fourpassengers and five crewmembers received fatal injuries. Another occupants received injuries which varied from minor tocritical.

(NTSB. 1973)

The distraction shown in this accident reportwas an operational one. Ironically, thedistraction itself was not serious. Thecrew's mismanagement of the distractioncaused it to be fatal.

Distractions can also come from outside thecockpit in the form of traffic, weather,unexpected rerouting, etc. The keys todealing with distractions in order to avoidcatastrophic consequences lie in the crew'sability to focus on aircraft control andhazard avoidance.

The crew should be prepared to avoiddistractions. The captain can fly theaircraft and delegate tasks that mightinterfere. Or vice-versa. While thedistraction in the L-1011 accident demandedimmediate attention, it was the captain'sresponsibility to set priorities and todelegate responsibilities or make workassignments. Tragically, the entire crewbecame absorbed in the distraction at theexpense of aircraft control.

Stress ManagementThere are two types of stress that candegrade flight performance. One type hasbeen called background stress. Backgroundstressors are chronic stress factors that arein the background of our everyday activities.They include job stress, stress to maintainschedules, fatigue, family stress, and the

31

stresses imposed by organizationalrequirements. Sometimes no single source ofstress may seem very serious, but thecombined or cumulative effects of stressorscan lead to degraded performance. In otherinstances, major life stressors such asdivorce or death of a loved one can have illeffects on one's performance. Counseling oncoping skills has proven effective incontrolling background stressors.

A second type of stress is acute stress.Acute stress is the overload that occurs in ahigh intensity event, such as an unfamiliarflight irregularity. This type of stressoccurs all at once, and the results can becatastrophic. Acute stress results inseveral negative consequences:

• "Tunnel vision", or the restriction ofattention to only part of a task

• Rigidity of response, or maintaining asingle course of action even thoughconditions have changed

• A tendency to scan alternatives lesseffectively during decision making, and

• "Ballistic" decision making; makingdecisions without thinking through theconsequences of a decision.

The intense time pressure, unfamiliarity, andoverload inherent in acute stress conditionsresult in a narrowed view. In turn,awareness is reduced regarding the hazards ofthe task, and the resources available to meetthose hazards.

32

Chapter 3: Implementing CRM Training

Early versions of CRM training were largelyattempts to adapt existing training materialsthat had been developed for other purposes,such as management training. But it becameclear that a more complete method was needed.

The Systems Approach to TrainingThe systems approach to training has beenused successfully in a variety of settings,including the military, airlines, and otherindustries. While the systems approach totraining exists in many forms, it generallyprovides guidelines for training in threesteps: DEVELOPING the training, IMPLEMENTINGthe training, and EVALUATING the training.It is not a rigid rule book but rather ageneral process for building trainingprograms.

What follows is a simplified scheme for CRMtraining, using the systems approach.

STEP1: Developing CRM Training

Training program development activities maybe grouped in three areas: (1) NeedsAssessment; (2) Setting PerformanceObjectives; and (3) Preparing a TrainingPlan.

1. Needs Assessment

The first step in the development of asuccessful CRM training program is to assessthe organization's training needs. There aremany ways to assess training needs, includingsurveys, reviews of incidents, and studies byspecial advisory committees.

The scope of the needs assessment depends onthe specific organization's size and

33

Management support isessential to effectiveCRM training.

complexity. A comprehensive needs assessmentcan help an organization identify trainingneeds most vital to its operations and toarrange other needs by priority.

An organization may not be entirely receptiveto the new training that it needs. Oneuseful tool for assessing an organization'sattitudes toward CRM training is the "CockpitManagement Attitudes Questionnaire",developed by Helmreich (1984). (SeeAppendix) This questionnaire can provideuseful readings on embedded attitudes towardtraining. Further, this type of measurementencourages people to get involved in thedevelopment of the training program.Training needs exist on at least threelevels: the individual, the group, and theorganization. We have mentioned theindividual in focusing on the group. But wehave said little about the organization, isthe organizational culture supportive of CRMtraining? is management on board? A CRMprogram has to be supported from the top ofthe organization to the bottom. Thereforesome training program may be needed for 'management. Management training may consistof distributing materials on crew-relatedincidents; posting information on CRMsuccesses such as the United Airlines flight232 story; or running a seminar formanagement describing the nature and value ofa CRM program, in any case, it is essentialto build management support early in proaramdevelopment.

To assist in the assessment of needs, a setof practical questions can be asked. Avariety of answers are appropriate. Thespecific answers depend on the uniquecharacteristics of the user organization.However, there have been some valuablegeneral lessons learned from the CRM trainingprograms that have been implemented atvarious airlines. Some of these lessonsfollow.

WHO?

Who should develop the training? Variouspeople can and do develop CRM trainingprograms. Programs can be bought off theshelf or developed from scratch. They can be

34

developed by outside consultants or byin-house experts. Some sort of <^™J»Vamong these options often provides the bestapproach.Two hints have proven helpful in CRM programdevelopment. First, there are excellentmodels of CRM training available, includingprograms developed by several airlines and bythe military aviation community. And manypublications are available covering lessonslearned over the last decade on building asuccessful CRM training program.Second, most organizations have found that ateam approach is effective in developing aCRM training program. Many organizations donot have training specialists on staff, inthis case, the design and development teamshould include both outside trainingconsultants and in-house representatives.In-house representatives should comprise awide array of managers, flightcrew, checkairmen, and instructors. This team approachavoids the pitfall of having an outsideexpert design a program in isolation—aprogram that later proves not to fit anorganization's needs. Perhaps even moreimportant, this team approach ensures thatflightcrews are involved in the CRM trainingprogram from the beginning—a critical factorin securing acceptance and commitment byusers.

Who should be trained? The CRM concept hasevolved from an initial narrow focus on thecockpit crew. It has broadened to includeall other groups who interact with thecockpit crew and who are involved indecisions that affect flight safety. Thesegroups include cabin crewmembers, air trafficcontrollers, dispatchers, maintenance people,customer service agents, and even specializedcrisis teams such as bomb threat and hijackteams. Although CRM training has focusedprimarily on cockpit crewmembers, someairlines have begun to develop CRM trainingprograms that include flight attendants,dispatchers, and maintenance people.

Central to the CRM concept is that CRMtraining is for the whole crew, not just forthe individual; that while individualexcellence is always desirable, teamwork is

CRM training is crewtraining, and may includeall groups that work withthe cockpit crew and areinvolved in decisionsthat impact flight safety.

35

The goal of successfulimplementation is thatthe new training"disappears" into theorganization: It becomesan accepted and routinepart of the normaltraining program.

the focus; and that a good blend of resourcesis the best bet for a safe flight.

WHAT?

What should be trained? Early CRM trainingprograms borrowed heavily from managementtraining practices. Conveniently, managementtraining had traditionally dealt with topicslike teamwork and leadership. Managementconcepts were often applied broadly to theaircrew setting, with varying degrees ofsuccess. One characteristic of more recentCRM training programs is a shift from broadconcepts to specific aircrew skills andbehaviors. While no set curriculum isnecessarily appropriate for all airlines, theCRM primary-level skills identified in Figure4 provide one useful framework for CRMtraining. These primary-level skills can beused in developing performance objectiveslater in Step 2: Setting PerformanceObjectives.

HOW?

How should training be introduced?Machiavelli wrote in The Prince that "Thereis nothing more difficult to arrange, moredoubtful of success, and more dangerous tocarry through than initiating changes."Resistance to change can take many forms,from passive disinterest to outrightsabotage. (The term sabotage, in fact, stemsfrom the French sabot or shoe, referring tothe wooden shoes thrown by workers intomachinery to jam gears). Because people tuneout what they are not motivated to hear, manya training program has been scuttled becauseof a lack of user acceptance. This problemis sometimes called lack of "buy-in" or"sign-up."

Usually, introducing change in anorganization invites problems; but notalways. Sometimes the need for change is soglaringly obvious that buy-in is almostautomatic.

In the early 1900s, for example,statisticians at American Telephone andTelegraph identified two powerful growthtrends: telephone use and population growth.Projecting these trends, they forecast that

36

by 1920 every female in the United Stateswould have to be employed as a switchboardoperator in order to meet demand. Within twoyears, AT&T had developed the automaticswitchboard. Everyone from overworkedswitchboard operators to impatient telephonecustomers welcomed the new system. Andtoday, of course, the automatic telephoneswitchboard is one of the key elements of ourso-called information age.

The trends in crew-related airline accidents(shown in Figure 3) are almost as obvious.And the consensus within the airlinecommunity strongly favors CRM training. Buta program known to be beneficial will notnecessarily be accepted. Even the best ofprograms may fail if the introduction of thetraining program is managed poorly. Peopleoften become accustomed to doing things in acertain way, and resistance to change isnormal. But some steps can be taken topromote acceptance of "new" training. Thosesteps are covered later in this chapter under"Implementing CRM Training".

WHERE?

Where should training take place? InitialCRM training can take place in any settingthat is conducive to learning. Manyorganizations hold initial training inoff-site facilities in order to avoiddisruptions. However, the location ofinitial CRM training is less important thanthe process of bringing crew memberstogether.

CRM focuses on crew interaction. And CRMtraining is most effective in groups largeenough to include entire crew units. Laterphases of training are most effective whentrainees are broken out into crew units andtrained in simulators or other cockpitmock-ups.

HOW MUCH?

How much training is required? Initial CRMtraining can be accomplished in as little astwo or three days. But continual CRMtraining should become part of the totaltraining program. CRM skills should beconsidered a major element of the overall

CRM training must beintegrated into the totaltraining program.

37

Performance objectivesspecify what is to belearned.

skills package that produces safe flight.CRM training never really ends. More oncontinual training follows in this chapterunder "Preparing a Training Plan."

2. Setting Performance Objectives.The next step in the development of a CRMtraining program is to set performanceobjectives. Performance objectives are thedesired outcomes of training. They answerthe question, What will trainees be able todo at the end of a training session that theycould not do before training? Objectivesmust be simple, it must be clear if theyhave been met at the end of training. Usableperformance objectives include the specificbehaviors desired and standards for measuringsatisfactory performance. A performanceobjective for classroom training might be:

• Given a written scenario describingaccident No. xxx, crewmembers will be ableto state in writing at least threebarriers to communication present in thissituation.

By expressing objectives in this manner,those people designing the training programare forced to identify exactly what eachlesson should accomplish. Later theseperformance objectives serve as guidelinesfor the evaluation of training. The basicCRM skills were discussed in Chapter 2, CRMSkills, and can be used to develop CRMperformance objectives.

There are different types of learningobjectives. Trainees may learn:

• Intellectual skills, such as problemsolving

• Factual information (knowledge)• Attitudes

• Motor skills

Therefore, any one training session ortraining module may have a mix of trainingobjectives.

38

CRM depends upon support from the entireorganization, not just the aircrew. Thephrase "organizational shell" has been coinedto describe the organizational culture inwhich flight crews operate. One aviationpsychologist writes:

Imagine, ifyou will, abeautifully designed and professionally executed CRMprogram that helpscrewmembers leam andpractice precisely the skills that they need to operate well as ateam In ademanding flight environment. Nowplace that program in an organization where lines ofnight arebadly constructed and constantly changing at the last minute, crews are poorly composed andshort-lived, norms ofconduct reinforce individual order giving and taking rather than team-levelplanning, excellent crew performance goes wholly unrecognized, and crews often are unable toobtain information, technical assistance, ormaterial resources when they need them to proceedwith the work.

To complete agood CRM course in an organization that has...an unsupportive organizationalcontext is like getting all dressed up for adance and having the carbreak down halfway there.Cockpit resource management simply cannot take root and thrive unless organizational conditionsalso foster and support effective teamwork. (Hackman, 1987, p. 37)

A training plan providesa blueprint for CRMtraining development.

The point is well made that performanceobjectives must be set for the organizationas well as the individual.

3. Preparing a Training Plan.The third step in the development of CRMtraining is preparing a training plan. Atraining plan provides a complete descriptionof a course of instruction includingsequenced lessons. Lesson details provide adescription of training objectives, content,methods, training aids, and other elementsrequired for instruction. A sample of anabbreviated training plan is shown on thefollowing page.

A training plan identifies trainingobjectives, provides an outline of coursecontent, specifies training methods, and evenprovides an estimate of the time required foreach topic.

A variety of training methods may be chosento present course content. Since peoplelearn in a variety of ways—by listening, byseeing, by discussing, and by doing—itfollows that there are a number of trainingmethods available. Training methods includelectures, training tapes, seminars,

39

SAMPLE LESSON PLAN

Session Title: Communication Skills

Objectives:

Given a written scenario describing incident No. XXX, the crewmembers are ableto state in writing three barriers to communication present in this situation.

Crewmembers acknowledge communications from others during role-play.Crewmembers use standard terminology throughout a LOFTsimulation.

Topic

Introduction

Verbal andnonverbalcommunication

Recognition

Content

A. Barriers to effectiveaircrew communication

A. Giving and receivingfeedback

B. Acknowledgingcommunications

C. Using standard terminologyD. Nonverbal communications

How to recognize ineffectivecommunication skills in selfand others

Training method

Videotape demonstration

Lecture with viewgraphs

Classroom demonstration

Lecture with viewgraphs

Role-play communicationsexercise

Role-play exercise with selfand instructor critique.

Time

10minutes

20minutes

10minutes20minutes20minutes

30minutes

demonstrations, role-playing, and simulation.In general, it is preferable to use a varietyof training methods in order to enhancelearning and to sustain the trainees'interest.

Lesson developers may find the followingpoints helpful. First, education specialistshave identified nine critical instructionalevents that support learning:

40

1. Gain the trainee's attention.2. Describe the objectives of training.3. Stimulate recall of prerequisite

skills.

4. Present content to be learned.

5. Provide guidance and support.

6. Elicit the performance.

7. Provide feedback.

8. Assess performance.

9. Enhance retention and transfer,

(from Gagne, Briggs, & Wager, 1988)

Second, research shows that people rememberabout 10% of what they hear, 30% of what theysee, and 50% of what they see and hear.

WHAT YOD REMEMBER

What you What you What youhear see see and hear

It follows thatthe more of thetrainee's sensesare involved, thegreater thelearningexperience.Hands-on training(hearing, seeingand feeling) isbetter thanwatching atraining tape. Atraining videotapewith a soundtrackis better than atraining audiotape alone.

Third, it is always desirable to mixinstructional methods and actively involvethe learner. More than young people, adultsare known to respond to training that can bereadily applied in their work. Instructionshould not be limited to lectures andtraining tapes, for example. Role-playing,workshops and other practical exercises that

41

Initial training focuses onintroducing CRM skillsand fostering awarenessof the need for CRMtraining.

Learning new skillsrequires active practiceand feedback.

involve the trainee promote the sense thatthe training is relevant and useful.

Finally, all effective training, includingCRM training, comprises three essentialcomponents:

1. Indoctrination/Awareness

2. Practice and Feedback

3. Operational Reinforcement

1. Indoctrination/Awareness. Initialtraining is conducted in a classroom settingwith the goal of laying a foundation forskills acquisition. One goal of training atthis stage is to convince the crewmember thathe or she can benefit personally from CRMtraining. This goal can be met by reviewingdata on crew-related accidents, by describingcrew coordination problems, by viewingvideotapes of good and poor crew performance,and by assessing personal interaction styles.A second goal is to establish common conceptsand terms for CRM. Discussions,demonstrations, and role-playing in the CRMskills are commonly used methods.

2. Practice and Feedback. Following initialclassroom instruction, trainees are given theopportunity to apply the newly acquired CRMskills. The primary emphasis of CRM trainingat this stage is crew performance.

Changes in knowledge and attitudes sometimesare sufficient to effect changes inperformance, but not always. Something moreis usually required. Practicing skills andreceiving feedback on performance are provento be effective in completing the learning ofdesired behaviors.

Effective practice may be achieved throughhigh, medium, or low fidelity simulations.Realistic high fidelity simulations may beachieved through blending CRM training intoLine Oriented Flight Training (LOFT). LOFTprovides the ideal setting for skillspractice, because it presents a full-crew,full-mission scenario in which CRM skills canbe evaluated along with technical skills.

42

CRM practice may range from full-missionsimulation to role-playing using cockpitprocedures trainers (CPTs) or even crudercockpit mock-ups.

Feedback on performance may also be providedby a number of means, videotape provides aclear and vivid record of performance, and iswidely used. Replay, fast forward, slowmotion, and freeze frame are some of thevaluable videotaping functions that can beused during debriefings.Under the guidance of a skilled instructor,the crew's critique of its own performanceduring debriefing can effectively build CRMskills. Other feedback valuable to theindividual comes from a pilot's fellowcrewmembers. And often the most lastingbenefit to the pilot comes from his ownself-assessment. Each of these assessmentsis vastly improved by the use of videotapingand other audio-visual aids.

3. Operational Reinforcement. All of us havehit ourselves on the finger with a hammer.We experienced what is known as one-triallearning. We immediately grasped the conceptof hammer/finger interaction, and probablyvowed immediately never to do that again.However, the term "one-trial learning" isactually misleading. If we never practiceeven a simple skill like hammering, we aremore likely to hit our fingers again.Sustaining the sophisticated skills of crewcoordination requires continual repetitionand reinforcement. Studies have shown that

._ , A the initial benefits of CRM trainingBehavior that is not dissipate within a relatively short period ifrehearsed and reinforced training is not reinforced over time. Thus,decays over time. CRM training should be included as a regular

part of recurrent training. Since peopletend to forget both facts and skills overtime, continual training should includerefresher seminars reinforced by practice andfeedback exercises such as LOFT.

Training method selection may vary accordingto the component of CRM training beingtaught. The purpose of the indoctrination/awareness component of CRM training is toprovide concepts and terms for crewcoordination problems, and to address CRMskills. Indoctrination/awareness commonlybegins in a classroom setting. Methods of

43

LOFT provides realistic,full-mission simulation toenhance skill transfer.

instruction include lectures, seminars andguided discussions.

The purpose of the practice and feedbackcomponent of CRM training is to apply newlyacquired CRM skills. Accordingly, emphasisis placed on exercises, demonstrations,role-playing, and simulations. Trainees aregiven an opportunity to try out thesebehaviors and to receive feedback on theirperformance.

The purpose of operational reinforcement isto prevent the loss of CRM skills once theyare acquired. CRM training must be includedas a regular part of the recurrent trainingrequirement. Continued reinforcement of CRMskills will require a mix of trainingmethods, including seminars, active practice,and feedback.

The mission of CRM training is to promotechanges in behaviors related to crewcoordination in order to prevent accidents.The bottom line is crew performance.Performance starts with awareness ofappropriate knowledge, attitudes, andbehaviors. But awareness is not enough.Like the athlete with great potential,crewmembers need an opportunity to hone theirperformance through realistic learningexperiences. The practice and feedbackcomponent of CRM training meets this need.

The best training method known for CRMpractice and feedback is simulated lineoperations using flight simulators. LineOriented Flight Training (LOFT) is the bestknown of these programs. LOFT blended withCRM training provides several advantages:

• It embeds CRM into the total trainingprogram.

• It provides a full mission scenario inwhich crews are evaluated for both CRMskills and traditional technical skills.

• It allows crew training in a full crewcontext.

• It invites the use of video recording andplayback during debriefing, providing anexcellent means of evaluating performance.

44

• It is accepted by most crewmembers andprovides an effective means of reinforcingCRM skills.

If LOFT is not available, CRM practice andfeedback can be accomplished with lower-levelsimulations. Realistic demonstrations,problem-solving scenarios, and role-playingexercises, complemented by audiovisual aids,can provide an effective means of reinforcingCRM skills. Keys to effective practiceinclude direct learner involvement, hands-ontraining experience, and immediate feedbackwith the opportunity for further practice.

STEP 2: Implementing CRM Training

Getting key personnelinvolved in the designand development of CRMtraining is one way toenhance commitment.

It is important to develop an implementationplan early in the training developmentprocess. Determine who should be involved inthe implementation; anticipate user reactionsand consider ways to overcome possiblenegative responses; and develop a timetablefor implementation. Finally, introduce thetraining carefully and deliberately. Someorganizations have tried out new trainingprograms on test groups, then asked forinput. They have then revised the programsbefore implementing them on a widespreadbasis.

Seek participation. When a change occurs ina corporate setting, it is crucial toencourage buy-in and discourage resistance.In the extreme, resistance can becomeoutright rejection, sometimes called the NotInvented Here syndrome. Not Invented Here ismost likely when people do not feel involved.Get as many key people as possible involvedin the development of the CRM training. Asearly as possible sign up the opinion leaderswithin your organization and any othersignificant group such as the pilots'association. Seek their participation early,and maintain their involvement while the CRMelement is worked into the total trainingprogram.

Demonstrate program support. In 1961, amajor American corporation established aschool for managers to teach a new approachto company operations. The new approach was

45

based on an innovative marketing philosophycentered on customer needs. After completingthe course, 85% of the management traineesleft the company within a short time. Thereason? The company had not changed, andmanagement had retained its habitual way ofdoing things. In this case, the corporateculture did not support the very change thatupper management had tried to introduce.This example illustrates that all levels inan organization must be on-board or a newprogram may quickly die for lack of support.Thus, it is critical that CRM program supportbe conspicuous from top management through toline operations. Similarly, the support ofthe pilots' association can provide avaluable program endorsement.

Communicate. Commitment can't be assumed; itmust be earned. Therefore, it is vital tocommunicate the reasons for a CRM program.People need to know what CRM is and what itisn't, what it can accomplish, and why CRMtraining is important. Communication can beaccomplished through meetings, seminars,newsletters, and posters. The following is auseful guide to effective communications.

The Seven C's ofEffective Communication

Credibility: The messenger must be seen ascredible and competent by the receiver.

Context: The message should be delivered in acontext that is comfortable to the receiver andinvites his participation.

content: The message must be relevant andimportant to the receiver.

Clarity: The message must be clear and must bedelivered in terms the receiver will understand.Jargon is to be avoided.

Continuity and consistency: The message shouldbe reinforced often and should be consistentfrom all voices within the organization.

Channels: Channels of communication should befamiliar and relevant to the receiver.

Capability: The message should be suited to thecapabilities of the receiver.

46

Check airmen andinstructors must behighly skilled in all areasof CRM performance.

Follow-up. The most valuable guidance comesfrom the student, not from the instructor.Monitor program implementation throughfeedback from CRM trainees. Be prepared toaddress their concerns, and to revise theprogram as appropriate.

The Kev Role of Check Airmen and InstructorsThe success of any CRM training program restson the skills of the people who administerthe training. Thus, it is vital that checkairmen and instructors be carefully chosen.Although check airmen and instructors areoften chosen for their proficiency intechnical matters, proficiency in teachingand evaluating CRM skills is an additionalcredential, not always easily earned.Check airmen and instructors are especiallyimportant for two reasons. First, they aredirectly responsible for instruction,observation, feedback, and programevaluation. Second, they are in a strategicposition to "sell" CRM and are likely to beopinion leaders within the organization.Because of their high profile, any attitudeshort of conspicuous support on their partcan undermine the entire CRM program, whichdepends upon voluntary buy-in at every levelof the organization.

There are at least three ways to encouragecheck airmen and instructors to support CRMand to become proficient in CRM skills. Forone, they can be given CRM training beforeanyone else. For another, they can be givenmore training than anyone else. Finally,check airmen and instructors can be signed upfor the CRM program at an early point. Theycan then provide input into the developmentof the CRM program and develop a sense ofownership of it. That sense of ownership, inturn, sets an example and encourages buy-inby others.

47

STEP 3: Evaluating CRM Training

The consideration of howtraining will be evaluatedshould begin in the earlystages of training design.

Ex-Mayor Koch of New York was known forasking "How am I doing?" as he strolled NewYork's boroughs. This habit was not onlygood for public relations, but it alsoprovided Koch with immediate feedback on hisperformance.

There are several reasons why the question"How are we doing?" needs to be askedcontinually during CRM training:

• To determine if the training meetsobjectives. For example: Aftercompleting the training module onEffective Communications, are traineesactually communicating any better?

• To determine if results from courses meetoverall program goals. For example:After mastering the Communication lessonobjectives for individuals, are crewscommunicating any better?

• To provide feedback to trainees.

• To provide feedback to instructors.

• To review and improve the training programitself.

Preparations for training program (CRM)evaluation should be made during the trainingdesign phase, while training objectives arebeing set. Objectives should be clear andmeasurable. Objectives that are vague ordifficult to measure invite trouble inevaluating the success of the program. Aclear plan for program evaluation should bedeveloped early in the training designprocess because evaluation activities shouldbegin before training is implemented.

As mentioned earlier, it is likely that avariety of outcomes will come from training,including changes in knowledge, attitudes,and behavior. It is appropriate to usemultiple measures of effectiveness inevaluating training. As an example,researchers at the Naval Training SystemsCenter have adopted the following measures ofeffectiveness for a Navy aircrew coordinationtraining program:

48

Tvoe of Evaluation Sample Item/Measure

Pre-Training Assessment 1. Aircrewcoordination is criticaltoflight safety (agree/disagree).

2. List four barriers to effectivecommunication in the cockpit.

Trainee Reaction

Learning

Performance

1.This training was relevant to my flightperformance.

1. List four barriers to effectivecommunication in the cockpit.

1.Crew requests clarification of garbledcommunication during simulatedexercise.

Organizational Outcome 1. Data on incidents involving crewcoordination.

Adapted from Cannon-Bowers et al. (1989).

Pre-training assessment is done beforetraining begins. Its purpose is to provide apre-training baseline measure of skills andattitudes. This pre-training baseline isused as a reference against whichpost-training improvements are measured.Trainee reaction measurements can be taken toassess trainees' reactions regarding trainingprogram relevance, content, methods, or otherfeatures. "The Cockpit Management AttitudesQuestionnaire" is one evaluation tool that isused to assess pre-training and post-trainingattitudes toward crew coordination (SeeAppendix).

Tracking various reactions to CRM training isvaluable for several reasons. First,negative reactions signal a lack ofacceptance by users. But they often point tospecific areas where small adjustments may

49

Evaluation can be usedto assess reactions totraining, learning,changes in performance,and the degree to whichorganizational goals aremet.

Using multiple measuresof training effectivenesspermits a comprehensiveand informative trainingevaluation.

make a big difference immediately. Second,positive reactions probably point the way tofurther program improvements. Third, othergroups' reactions to the training program mayprovide valuable cues for improvements;groups like flight attendants, agents,mechanics, and dispatchers. And finally,reactions tracked over time can reveal muchabout changes within the corporation.Learning is measured against learningobjectives, which are just the flip-side oftraining objectives. The most familiar kindof measurement is probably the question andanswer test. Answers may be oral or written.As an example, one objective might be thattrainees be able to describe barriers tocommunication in the cockpit. Measurementmight take the form of a simple questionrequiring the trainee to describe fourbarriers.

Performance is measured against performanceobjectives. Typically, instructors receivespecial training in conducting performanceevaluations, and then observe and rate crewperformance against performance objectivesduring simulation exercises.

Organizational outcome measures includecomparisons of data on crew-relatedaccidents. Although increased safety is asimple enough ideal, attaining a validmeasure of the effect of CRM training onsafety is very difficult. One reason is thatit is difficult to analyze events that seldomoccur—aviation accidents. Another reason isthat it is often difficult to measure whetherchanges in the accident rate are related toCRM training or to other events, such asequipment upgrades or industry-wide changes.The difficulty in using such comparisons tomeasure training program effectiveness pointsto the need for multiple measures.

Multiple measures of effectiveness are usefulin documenting program successes andshortfalls. For example, trainees may failbecause they have not learned the skills, donot clearly understand CRM concepts, or failto see the training as meaningful orrelevant. Using multiple measures of aprogram's effectiveness allows the evaluatorto isolate specific areas for programimprovement.

50

Finally, evaluation should not be punitive.The purpose of CRM training evaluation is tomeasure program effectiveness and to providespecific guidance for improvements to theprogram. Training evaluation should not beused to assess an individual's fitness forduty.

CRM skills are crew skills. CRM evaluationshould focus on crew or team performance.Individual crewmembers should receive clearfeedback on their own performances, but theemphasis should remain on crew performance.

A Final Note on CRM Training

Setting up a CRM training program is not asdifficult as it might seem to the CRMnewcomer. Using a systems approach theprocess can be broken into three steps:Development, Implementation, and Evaluation.Each of these steps can be further brokendown into manageable units such as NeedsAssessment, Setting Performance Objectives,and Preparing a Training Plan. And theseunits can be further broken down asnecessary.

Furthermore, good CRM training can beaccomplished with a conservative budget byblending it with existing training, usingexisting resources.

A well-conceived CRM training package canblend nicely with existing training and meetwith little resistance. While the CRMelement is perceived as new, it is oftentreated as a stand-alone training product.Over time, as the blending becomes complete,CRM should become indistinguishable fromother training. At that point it has become"seamless", or "invisible" to the trainees,the most desirable state.

Airlines that do not yet have CRM programsare in the enviable position of being able tobuild on the lessons learned from more than adecade of CRM program development. There isnow a wealth of materials available tosupport CRM program development. Eachdeveloping CRM program can be custom-fittedto the unique needs of the organization while

51

drawing on the abundant resources alreadyavailable.

52

Chapter 4: Summary

Commercial aviation is one of the safestforms of transportation, with a safety recordthat is excellent by any standards. Thenumber of commercial jet aircraft in serviceworldwide has climbed steadily over the pastthree decades to a total of 9,530 in 1990,while annual departures have increased to13,298,000 for 1990 (Boeing, 1991).Remarkably, over this same period, the totalaccident rate has declined from over 60accidents per million departures to about 2.5— less than one twenty-fourth the accidentrate in 1959! (see Figure 1) This fortunatetrend can be attributed to advances inequipment technology, to a high level ofindividual technical proficiency, and toimproved operating procedures.

Despite the record, one challenge has notbeen met: Breakdowns in crew performancehave remained the primary factor incommercial jet accidents. Two out of threeaccidents are attributable to flightcrewerror.

In the early years, when equipmentreliability was the biggest problem, theaviation community responded with ingenuityand resolve. Engines and other aircraftcomponents became more reliable, and relatedaccidents declined. Today, with crewperformance the most significant threat toaviation safety, the industry has respondedwith an ambitious program to supporteffective crew coordination and performance:Crew Resource Management training.

One aviation observer has projected that thisindustry focus on crew performance has thepotential to double system safety. Data onthe effectiveness of existing CRM programsindicate that this challenge is being met.

53

54

Glossary

ACTIVE LISTENING: the skill of hearing and understanding otherpeople and checking the accuracy of one's understanding bycommunicating with the sender.

ACTIVE LEARNING: physically performing the essential movementsor a skill or displaying behavior that has been taught (asopposed to passive learning, or simply listening to instruction).ATTITUDE: a way of thinking or feeling; a mental dispositiontowards something that determines that person's response.

BEHAVIOR: a person's observable responses to a stimulus.

COMMUNICATION: the transfer of information and/or messagesbetween or among people by the use of words, letters, symbols, ornonverbal communication.

COMMUNICATIONS PROCESS AND DECISION BEHAVIOR SKILLS: a clusterof CRM behaviors related to effective communications and decisionmaking.

CREW RESOURCE MANAGEMENT: the effective utilization of allavailable resources—hardware, software and personnel—to achievesafe, efficient flight operations.

DECISION-MAKING: the process of selecting a course of actionfrom available options, based on information available at thetime.

FEEDBACK: response messages which clarify and ensure thatmeaning is transferred.

INDOCTRINATION/AWARENESS: the first phase in CRM training, whichprovides a conceptual framework for other phases. It typicallyconsists of classroom instruction focusing on identification ofCRM skills and concepts.

INQUIRY/ASSERTION: the skill of actively seeking out relevantinformation and showing a concern for both self and others'rights.

LEADERSHIP: the ability to utilize appropriate interpersonalskills to motivate, manage, and direct crew activities to achievea task.

LINE-ORIENTED FLIGHT TRAINING (LOFT): a full mission simulationpresented in real-time which is usually videotaped for later crewself-critique.

55

NEEDS ASSESSMENT: the process of determining where training isneeded in an organization, what a trainee must learn in order toperform their job effectively, and who needs training.

ORGANIZATIONAL CULTURE: the larger organizational environment inwhich the flightcrew operates. Structure, standards, and rewardpolicies are part of organizational culture. Management ororganizational support for CRM training is a critical componentof effective program implementation.

PASSIVE LEARNING: training that does not actively engage thetrainee in the instructional process; the trainee is expected topassively "absorb" instructional material.

PERFORMANCE OBJECTIVES: statements which describe the desiredoutcomes of training in specific, behavioral terms.

PLANNING: the ability to establish an appropriate course ofaction for self and others to accomplish a specific goal.

PRACTICE AND FEEDBACK: the second phase of CRM training, inwhich participants actively employ newly acquired CRM skills andreceive feedback on their effectiveness.

REINFORCEMENT: the final phase of CRM training. This phase isongoing and involves ensuring that CRM becomes an inseparablepart of the organization's culture by garnering top managementsupport, identifying and reinforcing effective behavior in normalline operations, and instituting CRM training as a regular partof the recurrent training requirement.

ROLE-PLAYING: a training technique in which trainees are told toimagine themselves in the situations presented by the trainer.Trainees are free to act out different behaviors and reactions aslong as they stay "in role" throughout the session.

SITUATIONAL AWARENESS: an active awareness of internal andexternal conditions that affect flight safety. It includes arealization of current, past and future contingencies that mayaffect flight performance.

STRESS MANAGEMENT: any of a variety of techniques, methods, orgeneral strategies which have been developed to help people copewith the adverse consequences of stress.

SYSTEMS APPROACH TO TRAINING: a systematic approach to trainingdevelopment which provides guidelines for training design,development of instructional activities, implementation, andevaluation of training.

TEAM BUILDING AND MAINTENANCE SKILLS: a cluster of CRM skillsfocusing on interpersonal relationships and effective teampractices.

56

TEAM MANAGEMENT: command and leadership by the captain andsupportive behavior by crewmembers.

TEAM REVIEW: skills involved in pre-mission planning andanalysis, ongoing synthesis and evaluation of information, andpost-mission debriefing.

TRAINING PLAN: an outline of what will take place during atraining session, including training objectives, content,training methods, and training media.

USER ACCEPTANCE: the extent to which a training program isaccepted and endorsed by users.

WORKLOAD MANAGEMENT AND SITUATIONAL AWARENESS SKILLS: a clusterof CRM skills which reflect the extent to which crewmembersmaintain awareness of the operational environment; anticipatecontingencies; and plan and allocate activities that managestress and workload.

57

58

Reference Materials

General Articles

The following references include documents cited in the text,plus other relevant articles and documents of general interest.

Boeing Commercial Airplane Group (1991). statistical summary ofcommercial net aircraft accidents, worldwide operations.1959-1990. Seattle, WA: Product Safety Organization, Boeing.

Billings, C. E., & Reynard, W. D. (1981). Dimensions of theinformation transfer problem. In C.E. Billings & E.S.Cheaney (Eds.), information transfer problems in theaviation system (NASA Technical Paper No. 1875). MoffettField, CA: NASA-Ames Research Center.

Cannon-Bowers, J. A., Prince, C., Salas, E., Owens, J. M.,Morgan, B. B., & Gonos, G. H. (1989, November). Determining.aircrew coordination training effectiveness. Paper presentedat the 11th annual meeting of the Interservice/IndustryTraining Systems Conference, Ft. Worth, TX.

Foushee, H. C. (1984). Dyads and triads at 35,000 feet: Factorsaffecting group process and aircrew performance. AmericanPsychologist. 39. 885-893.

Foushee, H. C. (1982). The role of communications,socio-psychological, and personality factors in themaintenance of crew coordination. Aviation. Space, andEnvironmental Medicine. 53, 1062-1066.

Foushee, H. C, & Manos, K. L. (1981). Information transferwithin the cockpit: Problems in intracockpit communications.In C. E. Billings & E. S. Cheaney (Eds.), Informationtransfer problems in the aviation system (NASA TP-1875).Moffett Field, CA: NASA-Ames Research Center.

Gagne, R. M., Briggs, L. J., & Wager, W. W. (1988). Principles ofinstructional design. New York: Holt.

Hackman, J. R. (1987). Organizational influences. In H. W. Orlady& H. C. Foushee (Eds.), Cockpit resource managementtraining; Proceedings of the NASA/MAC workshop (NASACP-2455) (pp. 23-39). Moffett Field, CA: NASA-Ames ResearchCenter.

Helmreich, R. L. (1984). Cockpit management attitudes. HumanFactors. 26. 583-589.

59

Helmreich, R. L., Chidester, T. R., Foushee, H. C., Gregorich,S., & Wilhelm, J. A. (1989). How effective is cockpitresource management training? Issues in evaluating theimpact of programs to enhance crew coordination (NASA/UTTechnical Report 89-2).

Helmreich, R. L., Chidester, T. R., Foushee, H. C., Gregorich, S.E., & Wilhelm, J. A. (1989). Critical issues in implementingand reinforcing cockpit resource management training(NASA/UT Technical Report 89-5).

International Civil Aviation Organization (1984). Accidentprevention manual (1st. Ed.). ICAO Doc 9422-AN/923.

Jensen, R. S. (1989). Aeronautical decision making - Cockpitresource management (DOT/FAA/PM-86/46). Arlington, VA:Systems Control Technology.

Kelley, W. (1991). Safety tips from the pros. Aviation Safety.11/ (21).

Lauber, J. K. (1984). Resource management in the cockpit. AirlinePilot. f>3t 20-23, 38.

National Transportation Safety Board. (1973). Aircraft accidentreport: Eastern Airlines. Inc. L-1011. N310EA Miami Florida(NTSB/AAR-73/14). Washington, DC.

National Transportation Safety Board. (1979). Aircraft accidentreport: United Airlines Flight 173. DC-8-61. N8082U(NTSB/AAR-79/12). Washington, DC.

National Transportation Safety Board. (1984). Aircraft accidentreport: Air Illinois Hawker Siddlev HS 748-2A. N748LL(NTSB/AAr-84/10). Washington, DC.

National Transportation Safety Board. (1989). Aircraft accidentreport: United Airlines B-737-200 structural failure(NTSB/AAR-89/03). Washington, DC.

National Transportation Safety Board. (1990a). Aircraft accidentreport: United Airlines DC-10-10 engine explosion andlanding at Sioux Citv Iowa (NTSB/AAR-90/06). Washington, DC.

National Transportation Safety Board. (1990b). Aircraft accidentreport: United Airlines B-747-122 cargo door failure andexplosive decompression (NTSB/AAR-90/01). Washington, DC.

Predmore, S. C. (1991). Microcoding of communications in accidentanalyses: Crew coordination in United 811 and United 232. InR. S. Jensen (Ed.), Proceedings of the sixth annualinternational symposium on aviation psychology. Columbus,OH: Ohio State University.

Ruffell Smith, H. P. (1979). A simulator study of the interactionof pilot workload with errors, vigilance, and decisions(NASA Technical Memorandum 78482). Moffett Field, CA:NASA-Ames Research Center.

60

Books and Conference Proceedings

Cooper, G. E., White, M. D., & Lauber, J. K. (Eds.). (1979).Resource management on the flight deck (NASA ConferencePublication 2120). Moffett Field, CA: NASA-Ames ResearchCenter.

Jensen, R. S. (Ed.). (1987). Proceedings of the fourth annualinternational symposium on aviation psychology. Columbus,OH: Ohio State University.

Jensen, R. S. (Ed.). (1989). Proceedings of the fifth annualinternational symposium on aviation psychology. Columbus,OH: Ohio State University.

Jensen, R. S. (Ed.). (1991). Proceedings of the sixth annualinternational symposium on aviation psychology. Columbus,OH: Ohio State University.

Komich, J. N. (1985). An analysis of the dearth of assertivenessby subordinate crewmembers. In R. S. Jensen and J. Adrion(Eds.), Proceedings of the third symposium on aviationpsychology (pp. 431-436). Ohio State University, Columbus,OH.

Orlady, H. W., & Foushee, H. C. (Eds.). (1987). Cockpit resourcemanagement training (NASA Conference Publication 2455).Moffett Field, CA: NASA-Ames Research Center.

Wiener, E. L., & Nagel, D. C. (Eds.). (1988). Human factors inaviation. San Diego: Academic Press.

CRM Training Applications

Bruce, K. D., & Jensen, D. (1986). Cockpit management training atPeople Express. In Orlady, H. W., & Foushee, H. C. (Eds.).Cockpit resource management training (NASA ConferencePublication 2455). Moffett Field, CA: NASA-Ames ResearchCenter.

Christian, D., & Morgan, A. (1986). Crew coordination concepts:Continental Airlines CRM training. In Orlady, H. W., &Foushee, H. C. (Eds.). Cockpit resource management training(NASA Conference Publication 2455). Moffett Field, CA:NASA-Ames Research Center.

Davidson, J. (1986). Introduction to Trans Australia Airlines CRMtraining. In Orlady, H. W., & Foushee, H. C. (Eds.).Cockpit resource management training (NASA ConferencePublication 2455). Moffett Field, CA: NASA-Ames ResearchCenter.

61

Scott-Milligan, F., & Wyness, B. (1986). CRM training and humanfactors training: what Air New Zealand is doing about it. InOrlady, H. W., & Foushee, H. C. (Eds.), Cockpit resourcemanagement training (NASA Conference Publication 2455).Moffett Field, CA: NASA-Ames Research Center.

Sellards, R. (1986). Cockpit resource management at USAir. InOrlady, H. W., & Foushee, H. C. (Eds.), Cockpit resourcemanagement training (NASA Conference Publication 2455).Moffett Field, CA: NASA-Ames Research Center.

Vandermark, M. J. (1989). A comprehensive crew management programfor the America West Airlines pilots and attendants. In R.S. Jensen (Ed.), Proceedings of The Fifth Annual Symposiumon Aviation Psychology (pp. 640-645). Columbus, OH: OhioState University.

Wilkinson, D. (1989). Cockpit resource management training - TheQantas experience. In R. S. Jensen (Ed.), Proceedings of TheFifth Annual Symposium on Aviation Psychology (pp. 456-461).Columbus, OH: Ohio State University.

Wilson, T. (1989). Aircrew team management - The Australianexperience. In R. S. Jensen (Ed.), Proceedings of The FifthAnnual Symposium on Aviation Psychology (pp. 462-467).Columbus, OH: Ohio State University.

Yocum, M., & Monan, W. (1986). CRM training in corporate/regionaloperations (Group V). In Orlady, H. W., & Foushee, H. C.(Eds.), Cockpit resource management training (NASAConference Publication 2455). Moffett Field, CA: NASA-AmesResearch Center.

62