Evaluation of Flashing Yellow Arrow Traffic Signals in Indiana

42 Pages·2015·3.62 MB·English

JOINT TRANSPORTATION
RESEARCH PROGRAM
INDIANA DEPARTMENT OF TRANSPORTATION
AND PURDUE UNIVERSITY
Evaluation of Flashing Yellow Arrow
Traffi c Signals in Indiana
Robert A. Rescot, Shuo Qu, Rebecca Noteboom, Ahmad Nafakh
SPR-3725 • Report Number: FHWA/IN/JTRP-2015/08 • DOI: 10.5703/1288284315530
RECOMMENDED CITATION
Evaluation of flashing yellow arrow traffic signals in Indiana
Rescot, R. A., Qu, S., Noteboom, R., & Nafakh, A. (2015). (Joint
Transportation Research Program Publication No. FHWA/IN/JTRP-2015/08). West Lafayette, IN: Purdue University.
http://dx.doi.org/10.5703/1288284315530
AUTHORS
Robert A. Rescot, PhD, PE
Assistant Professor of Civil Engineering
Department of Mechanical Engineering
Purdue University Calumet
(C2o1rr9e)s 4po3n3d-1in6g0 8Author
[email protected]
Shuo Qu
Civil Engineering Graduate Research Assistant
Department of Mechanical Engineering
PReubrdeuccea U Jn. iNveortseibtyo Coamlumet
Civil Engineering Undergraduate Research Assistant
Department of Mechanical Engineering
PAuhrmdaude UJ.n Nivaefraskithy Calumet
Civil Engineering Undergraduate Research Assistant
Department of Mechanical Engineering
Purdue University Calumet
JOINT TRANSPORTATION RESEARCH PROGRAM
The Joint Transportation Research Program serves as a vehicle for INDOT collaboration with higher education
institutions and industry in Indiana to facilitate innovation that results in continuous improvement in the planning,
design, construction, operation, management and economic efficiency of the Indiana transportation infrastructure.
https://engineering.purdue.edu/JTRP/index_html
Published reports of the Joint Transportation Research Program are available at: http://docs.lib.purdue.edu/jtrp/
NOTICE
The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of
the data presented herein. The contents do not necessarily reflect the official views and policies of the Indiana
Department of Transportation or the Federal Highway Administration. The report does not constitute a standard,
specification, or regulation.
COPYRIGHT
Copyright 2015 by Purdue University. All rights reserved.
Print ISBN: 978-1-62260-348-0
ePUB ISBN: 978-1-62260-349-7
TECHNICAL REPORT STANDARD TITLE PAGE
1.   Report No.  2.  Government Accession No.  3. Recipient's Catalog No. 

FHWA/IN/JTRP‐2015/08     

4. Title and Subtitle  5. Report Date 

Evaluation of Flashing Yellow Arrow Traffic Signals in Indiana  March 2015 
6.  Performing Organization Code 

7. Author(s)  8.  Performing Organization Report No. 

Robert A. Rescot, Shuo Qu, Rebecca Noteboom, Ahmad Nafakh  FHWA/IN/JTRP‐2015/08 

9.  Performing Organization Name and Address  10. Work Unit No. 

Joint Transportation Research Program 
Purdue University 
550 Stadium Mall Drive 
West Lafayette, IN 47907‐2051  11.  Contract or Grant No. 
SPR‐3725

 12.  Sponsoring Agency Name and Address  13.  Type of Report and Period Covered 

Indiana Department of Transportation  Final Report 
State Office Building 
100 North Senate Avenue 
Indianapolis, IN 46204 
14.  Sponsoring Agency Code 

15.  Supplementary Notes 
Prepared in cooperation with the Indiana Department of Transportation and Federal Highway Administration. 

16.  Abstract 

The evaluation of flashing yellow arrow signals for widespread implementation was evaluated. Through the collection of field driver 
performance data, survey data, crash data, at two test sites in the state, it was concluded that this is a worthwhile practice to be 
considered for a larger scale deployment. The return on investment includes both increased safety, and improved mobility. Given 
Indiana’s widespread usage of span and catenary signal supports, installation could be simplified to place a larger four section flashing 
yellow head in a horizontal orientation while leaving adjacent through lane three section signal heads in a vertical alignment, and not 
decrease the standard of care provided to the public, given proper engineering judgment. 

17.  Key Words  18.  Distribution Statement 

flashing yellow arrow, FYA, permissive left turns, traffic signals  No restrictions.  This document is available to the public through the 
National Technical Information Service, Springfield, VA 22161. 

19.  Security Classif. (of this report)    20.  Security Classif. (of this page)  21. No. of  Pages  22.  Price 

Unclassified  Unclassified  38   
Form DOT F 1700.7 (8‐69)
EXECUTIVESUMMARY datashowedthatamajorityofdriverssurveyedunderstoodwhattodo
invarioussituationsordidnotgiveafail-criticalresponse.Afail-critical
EVALUATIONOFFLASHINGYELLOWARROW responsewasonethatwouldlikelyleadtoacrash,suchasproceeding
TRAFFICSIGNALSININDIANA withtheright-of-wayonaredindication,orstoppingonasolidgreen
arrow. The only concerning case was when a vertical flashing yellow
signalheaddisplayedasolidyellowleft-turnarrowwhiletheadjacent
throughlaneshadasolidcirculargreensignalindication.Inthiscasethe
Introduction
fail-criticalresponseratewas11%.
Anationalreviewofmediareportswasalsoconducted,anditwas
The evaluation of the use of a flashing yellow arrow (FYA) for
identified that if proactive communications are not provided by
permissiveleftturnsinIndianawasundertakentoprovideguidanceto
INDOT,misinformationcanbeperpetuatedbythemedia.Membersof
theIndianaDepartmentofTransportationonthistopic.Thesesignal
theStudyAdvisoryCommitteedevelopedtoolstohelpprovideINDOT
typeshavebeenshownbyotherstatesandstudiestoreducecrashesand
withtoolsreadytobedeployedtolocalmediaoutlets.
improve mobility at the same time. This win-win opportunity is the
Crashdatawasrequestedfortheseintersectionsfrom2009through
resultofincreaseddriverawarenessduringthepermissiveleft-turnphase
thepresent.Itisfirstimportanttonotethatinthebeforecondition,left
at a signal where a driver must yield the right-of-way to oncoming
turns were protected only, and then when the FYA signals were
traffic. Increased mobility results from the elimination of the yellow
installed,thisintroducedapermissiveleftturn.Theliteraturesuggested
trap, thus allowing lead/lag signal phasing. This can be particularly
thatthecrashratewouldincreaseduetothisincreaseinmobility.The
importantforsignalcoordinationprojectsalongcorridorssothatgreen
actualcrashdatashowedthatwhencomparedtoprevioustrends,the
bandscanbebetteraligned.
flashingyellowsignalsdidproduceanincreaseinprojectedcrashrates.
InthestateofIndiana,mosttrafficsignalsupportsystemsinstalled
However,theincreasewaseitherbeloworwithinguidanceprovidedby
on the state highway system are span and catenary style instead of
theliterature.Thepost-installationcrashesateachsitewerecarefully
cantileveredmastarms.ThisisnotablebecausetypicalFYAsignalhead
scrutinized,andeleventotalhadoccurredthroughJuly2014.
arrangementsconsistofaverticalstackoffoursignalsections(fromtop
tobottom:solidredarrow,solidyellowarrow,flashingyellowarrow,
solidgreenarrow),insteadofbeingthreesectionsinthecaseofbotha
Implementation
protectedonlysignal,oraprotected-permissivefive-section‘‘doghouse’’
stylesignalhead.Thesesignalsonlyattachtothecatenarywiresatthe
InlookingatcoststoinstallFYAatintersections,thereareseveral
endsofthesignalhead,andtheadditionofanextrasectioninvertical
keyparameters.Thefirstisthesignalcontrolcabinethardware.Forthis
heightcouldcreateaverticalclearanceissue.Theonlywaytoresolve
hardwareexpense,aslongasthemalfunctionmanagementunitsareof
thisclearanceissuewouldbetothenraisethecatenarywires.However,
theMMU2specification,theycanbeupgradedforlittletonocostby
notallverticalsupportpoleshaveenoughextraverticalspacetopermit
meansofafirmwareupdate.Thiscouldbedoneaspartoftheroutine
such a raising, and this would require increased installation time.
annualrecertificationprocess.Ifacontrollerisnotcompliantwiththe
A possible solution for such a situation was evaluated in Vincennes,
MMU2 specification, then a possibly significant expense may be
Indiana,wherebythefour-sectionFYAsignalheadwasinstalledina
incurred.Thesecondcostisthatforanextrasectiontobeaddedtoa
horizontal orientation, while the other signal heads remained in a
signalhead.Thethirdcostwouldbetoupgradeexistingfieldwiresto
verticalorientation.
haveaminimumofeightconductorwires.Thecostofpullingadditional
wiresthroughthesignalconduitwouldbedwarfedbyanycostsrelated
Findings to collapsed or damaged conduit needing replacement. Based on the
AASHTO2010HighwaySafetyManual,thecomprehensivecostofa
Acomparativeanalysiswasconductedbetweenatypicalinstallation signalpropertydamageonlycrashis$7,400and$44,900forapossible
inCenterville,Indiana,andthemodifiedsetupinVincennes.Datawas injury crash. The reduction of at least one possible injury crash or a
collected at both intersections to study driver behavior when a solo combinationofseveralpropertydamageonlycrashesisallthatwould
vehiclewasapproachingaflashingyellowarrowandwouldhavetoyield be required to have a positive return on investment in the first year
forthepermissiveleftturn.Additionalcontrollocationsatfivesection alone. Continued crash reductions into the future would significantly
doghousesignalsineachcitywereselected.Theresultwasthattherewas increasethebenefitreceivedbytheState.
notanystatisticaldifferenceobservedindriveracceleration/deceleration Therefore,asaresultofthisresearch,itisclearthatalargerscale
betweenanyofthefoursites. implementationofFYAsignalheadsshouldbeconsidered.Therewas
Tosupplementthedriverbehaviordata,surveyswereconductedin alsonoreasontobelievethat,givenproperengineeringjudgment,the
both Vincennes and Richmond (neighboring city to Centerville). The placementofaFYAsignalheadinahorizontalconfigurationadjacent
survey participants were asked if they would go, yield, or stop when to vertical through lane signal heads would reduce the safety and
shown various images of left-turn scenarios. The result of the survey mobilitybenefitsprovidedbytheFYAsignalhead.
CONTENTS
1.INTRODUCTION........................................................................... 1
2.BACKGROUND............................................................................ 4
3.PUBLICOUTREACH........................................................................10
4.HARDWARELIMITATIONS .................................................................11
5.FIELDTESTSITES .........................................................................12
6.DATACOLLECTION........................................................................13
6.1 CrashData..............................................................................13
6.2 DriverPerformanceData ...................................................................13
6.3 SurveyData .............................................................................15
7.DATAANALYSIS ..........................................................................15
7.1 CrashData..............................................................................15
7.2 DriverPerformanceData ...................................................................16
7.3 SurveyData .............................................................................17
8.CONCLUSIONS ............................................................................18
9.ACKNOWLEDGMENTS .....................................................................19
REFERENCES ...............................................................................19
APPENDICES
AppendixA.State-by-StateImplementationsofFYA .................................................21
AppendixB.DriverSurveyImages...............................................................22
AppendixC.INDOTOperationsMemorandum13-04.................................................25
AppendixD.INDOTDesignMemorandum........................................................31
LISTOFTABLES
Table Page
Table2.1 BenefitsofFlashingYellowBallPermittedIndications—Washington 4
Table2.2 EngineeringAssessmentEvaluationMatrix—SafetyandOperations 6
Table2.3 EngineeringAssessmentEvaluationMatrix—ImplementabilityandHumanFactors 7
Table2.4 EngineeringAssessmentEvaluationMatrix—Versatility 8
Table2.5 AllowableCombinationsofPlacement,IndicationArrangement,andPhasingforPotentialIndicationType 9
Table2.6 SummaryofAverageCrashRate 9
Table2.7 RankOrderbyDisplayType—AverageCrashesper100,000Left-Turns6OpposingVehicles 9
Table2.8 RankOrderbyDisplayType—AverageLeft-TurnCrashesperYear 9
Table2.9 RankOrderbyDisplayType—AverageLeft-TurnCrashesper100Left-TurnVehicles 10
Table7.1 CrashDataSummary 15
Table7.2 CrashCauseSummaryatFYATestSites 16
Table7.3 SummaryofVehiclePerformanceData 17
Table7.4 ResultsofTwoSampleT-TestsforDifferencesinMeanAcceleration 17
Table7.5 ResultsofTwoSampleT-TestsforDifferencesinMeansofR2RegressionValues 17
Table7.6 SummaryofSurveyResponsestoResearchQuestions 18
Table7.7 SurveyResponseSummary 18
LISTOFFIGURES
Figure Page
Figure1.1 Comparisonofaprotected/permittedleft-turnlanesignaltoathroughlanesignal 1
Figure1.2 TypicalPPLTdisplaysusedintheUnitedStates 1
Figure1.3 UniquePPLTdisplaysusedintheUnitedStates 2
Figure1.4 Typicalyellowtrap 3
Figure1.5 FatalandinjurycrashesatsignalcontrolledintersectionsinIndiana 3
Figure1.6 Crashreductionsobservedatintersectionafterimplementingtheflashingyellowarrow 4
Figure2.1 Temporaryflashingyellowadvisorysign 4
Figure2.2 Auxiliarysigntype1 5
Figure2.3 Auxiliarysigntype2 5
Figure2.4 NCHRP20-07tableofresultsfordriverresponseatFYAsignals 6
Figure3.1 INDOTFYAcommunityhandoutcard 10
Figure3.2 Eastbound45thStreetatCalumetAvenueinMunster,Indiana 11
Figure3.3 SouthboundMississippiStreetat83rdStreetinMerrillville,Indiana 11
Figure5.1 INDOTFYAsupplementalsign 12
Figure5.2 Centerville,Indiana,FYAtestsitelookingwestboundtowardRichmond 12
Figure5.3 Vincennes,Indiana,FYAtestsitelookingeastbound 12
Figure5.4 WarningcardforcorrectMMU 13
Figure6.1 Fielddatacollectionsetup 13
Figure6.2 Fielddatacollectionasviewedfromadistance 13
Figure6.3 Vincennesdistance-timedata 14
Figure6.4 VincennesFYAspeed-timedata 14
Figure7.1 CentervilleFYAsamplevehicle#6trajectory 17
1.INTRODUCTION Lens Color and Left-Turn Indication
Area Used
Arrangement Protected Mode Permitted Mode
Providing a consistent driver expectation has been a
fundamentalprincipalofmoderntrafficengineering,andis MUTCD 4-Section
the foundation for the Manual on Uniform Traffic Control Horizontal Used in
Texas, Nebraska
Devices (MUTCD). However, traffic signals routinely
and others
provide mixed messages at locations having permissive left
turns. At such locations, the through lanes are allowed to
proceedwiththeright-of-wayonasolidcirculargreen(CG) MUTCD 5-Section
ballsignalphase,howeverwhenthesamesolidgreenballis Horizontal Used in
Texas, Nebraska
displayedforapermissiveleftturnitnolongerindicatesan
and others
assignment of right-of-way as shown in Figure 1.1. Thus a
‘‘Left Turn Yield on Green’’ sign is typically added to the
mastarmofthesignaltoclaritythisuncertaintytodrivers.
MUTCD 5-Section
Oneofthesolutionsforthisinconsistencyistosubstitutea
Vertical Used in
flashingyellowarrowinplaceofasolidgreenballtoindicate Texas and most
apermissiveleftturn.Byitsverydefinitionthecoloryellow Western States
(amber), when displayed by a traffic signal, indicates a
cautionarycondition,whichisconsistentwithhowadriver
shouldrespondtoapermissiveleft-turnsituation.Beginning
Variation of 5-
in March 2006, the Federal Highway Administration
Section Cluster
provided interim approval for the flashing yellow arrow to
beused,anditwasthensubsequentlyincorporatedintothe
2009MUTCD.
The flashing yellow arrow (FYA) traffic signal is a MUTCD Typical 5-
specialized signal display for protected/permissive left-turn Section Cluster
control(PPLT).Itutilizesaflashingyellowarrowtoinstead
ofatraditionalsolidcirculargreenballindicationduringthe
permissive phase with the intent to improve the safety Figure 1.2 Typical PPLT displays used in the United States
performance of the intersection. Figure 1.2 shows the most (Brehmeretal.,2003).
common PPLT display arrangements used in the United
States. Figure 1.3 illustrates other unique traffic signal
displaysthathavebeenusedinrecenttimes(Brehmer,Kacir adjacentthroughlanesreceivethecircularyellowindication
Noyce,&Manser,2003). for their change interval. Because of the change in signal
The yellow trap is one of the major problems of the indication, the left-turner mistakenly believes that the
traditional PPLT control displays. The yellow trap occurs opposing traffic also has the yellow change interval also
whenasignalchangesfromthepermissiveleft-turnintervals and so attempts to make the left turn, resulting in a
in both directions to a lagging protected movement in only dangerous condition at the intersection in these situations.
onedirection,asshowninFigure1.4.Adriverattemptingto Theyellowtrapoccursbecausetheopposingtrafficdoesnot,
makealeftturnonthepermissive circulargreenindication in fact, receive a yellow change interval but instead has a
becomestrappedintheintersectionwhentheircirculargreen circulargreenindicationinthethroughlanesandaprotected
indication turns yellow for the change interval (for the left-turn arrow indication. This condition creates the
through traffic). During the change interval, the left-turn potential for conflict between an unsuspecting left-turning
driver who is attempting to clear the intersection sees the vehicle and opposing, non-stopping, through traffic
(Brehmeretal.,2003).
There is a business case for considering the implemen-
tationofflashingyellowPPLTsignals.Accordingtopage21
LEFT TURN
YIELD oftheIndiana2010StrategicHighwaySafetyPlan‘‘Crashes
ON GREEN
at the intersection of two or more roadways in Indiana
produceoneinfourofallsevereoutcomecrashesandabout
one in five fatal crashes’’ (Brehmer et al., 2003).
Not the Same Message Implementation of the flashing yellow arrow in Indiana
represents an opportunity to reduce such crashes and save
lives. It also improves operational flexibility by permitting
lagging left turns. Such lagging left turns would otherwise
create a yellow trap, and thus opens up many additional
opportunities to improve mobility through intersection
coordination along corridors. Implementation of the
flashingyellowarrowmaybedoneaspartofsignalvisibility
Figure 1.1 Comparison of a protected/permitted left-turn lane improvement projects with highway safety improve-
signaltoathroughlanesignal. ment program (HSIP) funds from the Federal Highway
JointTransportationResearchProgramTechnicalReportFHWA/IN/JTRP-2015/08 1
Lens Color and Left Turrn Indicatioon
Area Used 
Arrangemment Protected Mode Permitted Mode
Maryland
Washington State 
No Longer in Operatioon
Cupertino, CA
Michigan
SSeattle, WA
Green or Yellow Bi-
Modal Lens
Sparks, NV
Reno, NV
Indicates Flashing
Figure1.3 UniquePPLTdisplaysusedintheUnitedStates(Brehmeretal.,2003).
Administration. Additionally, in time it may also possibly Recognizing that each crash has real costs, directly and
leadtoeliminationofsupplementalleft-turnsignalsignsand indirectly to the State, reducing these numbers holds the
thus reduce the number of maintenance items in the field. potential for real gains in economic output. At injury and
A reduction in the number of maintenance items would be fatal crashes, there are direct costs for law enforcement,
expected to yield a cost savings over a period of years into damaged roadside hardware, and clean up. Indirect costs
the future. Thus Indiana stands to gain substantially from include congestion, decreased worker productivity, and
this research, by saving lives, and creating opportunities to decreasedearningswhichresultsinadecreaseintaxestobe
integrate external (federal) funding; truly a win-win for the collected.
state. Insearchingformethodstoreducethesecrashes,onesuch
As previously mentioned, the 2010 Strategic Highway possibilityistheflashingyellowarrow.However,therearea
Safety Plan for Indiana states that intersection crashes number of possible combinations for how this could be
accountfor25%ofinjurycrashesand20%ofallfatalcrashes implemented, with varying implications. For example, the
in the state. Specifically at signalized intersections these NCHRP report 493 (project 3-65) entitled ‘‘Evaluation of
crashes have remained relatively steady for the period Traffic Signal Displays for Protected/Permissive Left-Turn
between2004and2009asseeninFigure1.5,withanaverage Control’’identifiedeightdifferentprotectedpermittedphasing
ofjustunder463injurycrashesperyearand55fatalitiesper schemes and 21 various hardware mounting combinations
year (Brehmer et al., 2003). This steady pattern of crashes options(Brehmeretal.,2003).
raisesthequestionofhowcantheStateseektoproactively As indicated in TRB web-only circular #123 the
decreasesuchcrashes. typical crash reduction observed at signalized intersections
2 JointTransportationResearchProgramTechnicalReportFHWA/IN/JTRP-2015/08
Opposing Through
Signal
1 All Red
2 Protected Left Turn
Clearance Interval (End Protected
3
Left-Turn)
4 Permissive Phase
5 Change Interval (Yellow Trap)
Opposing Through Phase Indication
6
Still Green
Figure1.4 Typicalyellowtrap(Brehmeretal.,2003).
implementing the flashing yellow arrow was approximately The safety and mobility benefits that are inherent with
30% at the study sites (as shown in Figure 1.6) (Brehmer FYA are only possible if drivers correctly understand the
et al., 2003). If one applies this reduction factor to the meaning of the signal displays. Kerrie L. Shattler et al.
statewide total of injury and fatal crashes (463+555518) preparedastudyaddressingthisissue.InherstudyShattler
therecouldbeapotentialforpreventing155crashesacross created a driver comprehension survey to evaluate driver
thestateperyearwiththistechnology. responses to various combinations of supplemental signage
and adjacent signal indications for both PPLT five section
700 signalheadsandverticalfoursectionFYAsignalheads.The
resultsofthissurveyincluded
600
54
500 56 60 50 53 54 . ‘‘Nosignificantdifferenceswerefoundincorrectdriverresponses
400 oftheFYAandCG,regardlessofwhetherornotasupplemental
signwasusedattheFYAapproach.However,theanalysisofthe
300 537 fail-criticalresponsesrevealedsignificantlyhigherincorrect‘‘go’’
200 429 484 429 446 452 responses for the CG scenario, compared to the FYA with
supplementalsign.’’(Schattler,Rietgraf,Burdett,&Lorton,2013)
100 . ‘‘Regardlessofthecoloroftheadjacentthroughtrafficsignal
(green or red), the provision of the supplemental sign at
0
2004 2005 2006 2007 2008 2009 the FYA approaches significantly improved drivers’
understanding of the correct ‘‘Yield’’message. This finding
Injury Crashes Fatal Crashes was further confirmed by the fail-critical responses, which
Figure 1.5 Fatal and injury crashes at signal controlled showedthattheFYAwithsupplementalsignhassignificantly
intersections in Indiana (Indiana Department of Transportation, lower fail-critical ‘‘go’’ responses than the FYA without a
2010). supplementalsign.’’(Schattleretal.,2013)
JointTransportationResearchProgramTechnicalReportFHWA/IN/JTRP-2015/08 3