AnalysisofAccidentStatisticsI.BACKGROUNDToensurethatsafetyobjectivesaremet,adistinctionmustbemadebetween:(1)locationswhicharehazardousasidentifiedbasedonaccidentexperiences,and(2)locationsandelementsthatarepotentiallyhazardousduetotheirgeometricsorphysicalfeatures.Alocationcanbeidentifiedashazardousbytheoccurrenceofanabnormalnumber,rate,orseverityofaccidentsoveragivenperiodoftime.A.TimePeriodandSegmentLengthConsiderationsAccidentbasedproceduresareusedtoidentifylocationsdefinedashazardousbasedonpastexperience.Theseproceduresinvolvethereviewandanalysisofsystemwideaccidentinformation.•Tocomparetheaccidentexperienceofseverallocationsfairly,theperiodoftimeoverwhichaccidentsarecountedandthelengthofroadwaysectionshouldideallybethesameateachlocation.•Ifnot,anaccidentratemaybecomparedbetweenlocations,expressedforacommonunitofexposure(i.e.,accidentspermillionvehiclemiles,oraccidentspermillionenteringvehicles).1.AnalysisPeriodConsiderations•Accidentdataforthemostrecent1to3-yearperiodisnormallyusedandisgenerallysufficient•2or3yearanalysisperiodsaremoreappropriateatlocationswithlowtrafficvolumes,wherea1-yearperiodmaynotprovidesufficientinformation.•Accidentdatashouldonlybeusedwhentherearenomajorchangesinfacilitycharacteristicsorlanduse.AnalysisofAccidentStatistics-2-PermitWritersWorkshop2.AnalysisRoadwayLengthConsiderationsTheroadwaynetworkcanbedividedintospotsand/orsegments.Isolatedcurves,bridges,andintersectionsareexamplesofspotlocations.Segmentsaretypicallydefinedbyaparticularlength(e.g.,300feet,0.1mile,1.0mile,etc.)orasthesectionofhighwaybetweentwodefinedspots.Whenselectingalengthforspotsorsegments,thefollowingpointsareconsidered:•Segment(longroadway)lengthsshouldbenoshorterthantheminimumdistanceusedbypoliceofficerstodescribeanaccidentlocation.Forexample,ifaccidentsarereportedtothenearest0.1mile,thentheminimumsegmentlengthshouldbe0.1mile.ODOT’saccidentdatabaseisreportedforevery0.01mile.•Forareaswhereaccidentreportingissubjecttoerrorsorlessaccuracy(i.e.,ruralareasorareaswherefieldreferencemarkersarefarapart),longersegmentlengthsshouldbeusedforanalysispurposes.•Spot(shortroadway)lengthsshouldincludetheareaofinfluencearoundahazard.Forinstance,theanalysisofintersectionsshouldincludeallaccidentsthatoccurwithinaspecifiedradius,suchas250feet,fromthecenteroftheintersection.Drivewayaccidentsoccurringwithin250feetfromthecenterofanintersectionshould,therefore,beincludedinthecountofaccidentsforthatintersection.Logically,itshouldbethestoppingsightdistance,plusthequeuelengthatintersections.AnalysisofAccidentStatistics-3-PermitWritersWorkshopII.MethodsforIdentifyingHighAccidentLocationsMethodsforanalyzingthehazardousnessoflocationsincludethefollowing:•Spotmapmethod•Accidentfrequencymethod•AccidentratemethodA.SpotMapMethodThesimplestmethodforidentifyinghazardouslocationsistoexamineanaccidentspotmap.Themapwillshowthespotsorsegmentshavingthegreatestnumbersofaccidents.Thisisaneffectivewaytogetapictureoftheaccidentclustersinsmallareas.B.AccidentFrequencyMethodThefrequencymethodrankslocationsbythenumberofaccidents.Thelocationwiththehighestnumberofaccidentsisrankedfirst,followedbythelocationwiththesecondhighestnumberofaccidents,andsoon.Thismethoddoesnottakeintoaccountthedifferingamountsoftrafficateachlocation.Therefore,thefrequencymethodtendstorankhighvolumelocationsashigh-accidentlocatio...
优质VIP
