pedestrian safety countermeasures deployment and evaluation: las vegas case study

Pedestrian Safety Countermeasures Deployment and Evaluation: Las Vegas Case

StudyShashi Nambisan

Director, InTrans & Professor of Civil EngineeringIowa State University (shashi@iastate.edu)

Srinivas Pulugurtha, The University of North Carolina at CharlotteMukund Dangeti, University of Nevada, Las VegasVinod Vasudevan, University of Nevada, Las Vegas

FHWA’sPedestrian Safety Web Conference

Washington, DCMay 28, 2009

Goals• Improve pedestrian safety, minimize risk• Identify, develop, deploy, and evaluate

countermeasures• Case Study: Las Vegas metro area, Nevada

2

Introduction• Significant growth for 20+ years• Wide, fast street grid network

High posted & operational vehicle speeds• Widely used transit system• High risk conditions for pedestrians• Demographics

Population ~ 1.8 million Diversity: age, race

• 85 percent of the crashes involved locals

3

Pedestrian Crashes (2003 – 2006) 4

Methodology• Identify candidate locations

GIS based analysis Site characteristics Problem characteristics

• Develop, deploy, & evaluate countermeasures • Measures of effectiveness

5

Study Design• Before and after Studies• Comparative studies (with control group)• Data collection ( ~18,000 pedestrians)• Statistical analyses

ParametricNon-parametric

6

Study Locations• Top priority / high risk locations

Crash index and crash rank• Site selection: 18 locations

Includes 4 control locations Excluded the resort Corridor (The “Strip” and its

proximity)• Different jurisdictions

City of Las Vegas City of North Las Vegas Clark County Nevada Dept of Transportation (State)

7

Study Locations

Control PointsHigh Pedestrian Risk Locations

R!

Major Streets0 0.7 1.4 2.10.35Miles /

! !R

R !9 10

13 14 15 16

1817

12 11

! R7 8

!!R

123

R !6 5 4

!!

SAHARA AV

CAREY AV

FLAMINGO RD

TROPICANA AV

WASHINGTON AV

JONES BL VD

OWENS AVNEL LIS BL VD

CHARLESTON BLVD

DECAT UR

BL VD

VEGAS DR

STEWART AV

LAKE MEAD BLVD

PARADISE RD

BR UC E ST

MAR YL AN D

P KWY

MART IN L KIN

G BLV D

VEGAS VALLEY DR

SMOKE RANCH RD

PECO

S R

D

DESERT INN RD

BONANZA RD

TWAIN AV4TH ST

TWAIN AV

US 95

PEC OS R

D

DESERT INN RD

PE COS R D

8

Selection of Countermeasures• Site characteristics

Geometric conditions Operating conditions Light conditions Demographics Land-use

• Costs

9

Countermeasures• Engineering based countermeasures• ITS based countermeasures• Others

10

Advanced Warning Signs / Yield Markings 11

High Visibility Crosswalk Treatment

12

In-Roadway Knockdown Signs 13

Portable Speed Trailer 14

Turning Vehicles Yield to Pedestrians

15

Danish Offset and Median Refuge16

Pedestrian Activated Flashers 17

Automatic Pedestrian Detection and Smart Lighting

18

Pedestrian Buttons that Confirm “Call”

19

Pedestrian Channelization 20

ITS No-Turn on Red Blank out Signs

21

Pedestrian Countdown Timers with Animated Eyes

22

Measures of Effectiveness / Statistical Tests• Pedestrian

Using the crosswalk Captured / diverted Looking for cars before crossing Trapped in the middle of the street Pedestrian-vehicle Conflicts Pedestrian waiting for signal to cross Delay

• Driver Yielding behavior, distance Blocking crosswalk Speed

23

Speed Trailer Site Information 24

Speed Trailer and Vehicle Speeds

35.0

40.0

31.5 31.9

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

40.0

45.0

Direction of Traffic

Ave

rage

Spe

ed (m

ph)

Before After

54.6 kph

50.7 kph

64.3 kph

51.3 kph

25

Speed Trailer: Vehicle Speeds Analysis

MOE

Baseline vs. Stage 1 Baseline vs. Stage 2

DeltaMeanSpeed

P-value H0

DeltaMeanSpeed

P-value H0

H0: Vbefore= Vafter vs. Ha: Vafter< Vbefore

Eastbound mph

(kmph)

5.5(8.9) <0.001 Reject 8.1

(13.0) <0.001 Reject

Westbound mph

(kmph)

6.5(10.5) <0.001 Reject 3.7

(6.0) <0.001 Reject

26

Speed Trailer: Analysis of Pedestrians(Safety) Measures of Effectiveness

Baseline Stage 1 Stage 2

Sample = 165 Sample = 47 Sample = 156

Percent Percent Percent

% pedestrians who look for vehicles before beginning to cross 80 100 100

% pedestrians who look for vehicles before crossing 2nd half of street 85 100 100

% pedestrians trapped in the roadway 41 34 37

27

Highly Effective CountermeasuresDescription CostAdvanced Yield Markings for Motorists Low

In-roadway Knockdown Signs Low

Pedestrian Countdown Signals with Animated Eyes

Medium

Danish Offset High

Median Refuge High

Portable Speed Trailer High

Pedestrian Activated Flashing Yellow High

28

Moderately Effective CountermeasuresDescription CostPedestrian Call buttons that Confirm Call (Visible/Audible confirmation)

Low

Turning Vehicles Yield to Pedestrians Low

ITS No-Turn on Red Signs Medium

ITS Automatic Pedestrian Detection Devices

High

29

Countermeasures with Low Effectiveness

Description CostWarning Signs for Motorists Low

High Visibility Crosswalk Treatment Medium

Pedestrian Channelization High

Smart Lighting High

30

Summary• Significant overall benefits

Pedestrian Driver

• Permitting & deployment considerations• Administrative / jurisdictional hurdles• Vendor / procurement difficulties• Education needs: pedestrians, motorists

31

Acknowledgments• US Dept of Transp., Federal Highway Admin • Nevada Dept of Transportation• Nevada Office of Traffic Safety• Regional Transp Commission of So. Nevada • Clark County, Nevada• City of Las Vegas• UNLV TRC: students, staff

32