文档下载 文档下载导航
设为首页 | 加入收藏
搜索 请输入内容:  
 导航当前位置: 文档下载 > 所有分类 > 自然科学 > 数学 > 2104数学建模美赛A 交通流 安全性 参考资料 交通量
侵权投诉

2104数学建模美赛A 交通流 安全性 参考资料 交通量

2104数学建模美赛A 交通流 安全性 参考资料 交通量

DataDescription

FusionofAccidentandTraf cFlowData

TheaccidentdatawereobtainedfromtheTraf cAccidentSur-veillanceandAnalysisSystem TASAS maintainedbytheCali-forniaDepartmentofTransportation Caltrans1993 .Thedata-basecontainsthosecollisionsthatoccurontheCaliforniaStateHighwaySystemforwhichtherearepolicereports.MostofthecollisionsincludedintheTASASdatabasewereinvestigatedinthe eld,butsomewerereportedafterthefact,usuallyforinsur-ancereasons.Thedatabasedoesnotcovercollisionsforwhichtherearenopolicereports.Sincethefocusisoncollisionsthatinvolvedvehiclestravelingonthemainlanesofurbanfreeways,wewereconcernedonlywithwhatarede nedas‘‘highway’’collisionsintheTASASdatabase.Forcalendaryear1998,9,341suchcollisionsarerecordedinthedatabaseforsixmajorfreewayroutesinOrangeCounty,California:InterstateRoute5,StateRoute22,StateRoute55,StateRoute57,StateRoute91,andInterstateRoute405.

Dataontraf c owduringthetimeperiodleadinguptoeachaccidentwasmatchedtotheaccident.Thesedatacomefromanarchiveddatabaseof30-sobservationsfrominductanceloopde-tectorsburiedatintervalsalongthefreeways.Thesedetectorsprovideinformationontwovariablesforeach30-sinterval:Thenumberofvehiclesthatpassovertheloop count andthepro-portionoftimethattheloopiscoveredbyavehicle occupancy .Althoughthesetwovariablescanbeused underveryrestrictiveassumptionsofuniformspeedandaveragevehiclelength,andtakingintoaccountthephysicalinstallationofeachloop toinferestimatesofspacemeanspeedsatapoint,weavoidmakinganysuchassumptions,anduseonlythesedirectmeasurementsinouranalyses.Weassumeonlythattheratioofcounttooccupancyhasamonotonicrelationshiptospacemeanspeed.Aftertestingdif-ferentlengthsoftimeformonitoringoftraf cconditions,wedeterminedthatweneededapproximately30minof30-sobser-vationsattheloopdetectorstationclosesttothelocationoftheaccidenttoestablishstablemeasuresoftraf cconditionspriortotheaccident.

Thetimeofeachaccidentisnotknownwithprecision.Aninspectionoftheaccidenttimes,presumablyobtainedfromeye-witnessaccountsdocumentedinpolicereports,revealsthat85.6%ofthe9,341collisionshavereportedtimesinminutesthatfallpreciselyonthe125-minintervalsthatcompriseanhour.Becauseofthisobviousreportingbias,reportedaccidenttimesaretreatedaslikelybeingroundedtothenearest5-mininterval.Sinceitisimportantinthisstudythatthetraf cdatarepresentpreaccidentconditions ratherthanconditionsarisingfromtheaccidentitself ,theperiodofobservationsusedintheanalysisiscutoff2.5minbeforethe‘‘nominal’’accidenttimetohelpre-moveany‘‘causeandeffect’’ambiguitiesassociatedwiththeap-parentround-offofreportedtimes.Consequently,foreachacci-dent,pre-accidenttraf cconditionsaremeasuredbyupto55sequential30-sloop-detectorobservations,beginning30minbe-forethenominalaccidenttime.

Ateachmainlineloop-detectorstation,datatypicallyarecol-lectedforeachfreewaylane;theminimumnumberoflanesatanymainlinefreewaysectioninOrangeCountyin1998wasthree.Tostandardizetraf c owdataforallcollisionsindependentofthenumberoffreewaylanesinvolved,datawerecompiledforthreelanedesignations: 1 theleftlane,alwaysbeingthelanedesig-natedasthenumberonelaneaccordingtostandardnomenclatureofnumberinglanesinsuccessionfromthemediantotheright

shoulder; 2 aninteriorlane,beinglanetwoonthree-andfour-lanefreewaysectionsandlanethreeon ve-andsix-lanesec-tions;and 3 therightlane,alwaysbeingthehighestnumbered right-most lane.

Missingdataprovedamajorproblemindealingwiththeloop-detectordatausedinthisstudy.Completedataforall55intervals a27.5-minperiod wasavailablefor24.5%ofthestations;an-other11.4%ofthestationshadmissingdataforoneormoreofthe55timeslices.Theremaining64.1%oftheloop-detectorsta-tionsreportednodataatallfortheentire27.5-minperiod.Pre-sumably,theselatterstationswereinoperativeatthattime,ortherewassomeotherprobleminretrievingthedata.

Filteringofobservationsbycontentwasstillnecessaryfortheloop-detectorstationswitheitherfullorpartialdata.Wereviewedalldatasequencesbasedontimeseriesdeviations,deviationsacrosslanes,andlogicalrulesderivedfromfeasiblevolumeandoccupancyrelationships i.e.,frompropertiesofplausiblefunda-mentaltraf c owdiagrams .Basedonthesetests,approximately16%oftheavailable30-sloop-detectorobservationswereiden-ti edasbeingpotentiallyinvalid.Insituationswhereone30-sobservationwasmissingorout-of-boundsbutwherethedatafortheadjacenttimesliceswerevalid,thedataforthemissingtimeslicewereinterpolatedfromtheadjacentobservations.

Implementationofthe lteringandinterpolationoperationsresultedinasampleof1,191collisionswithafull27.5minofostensiblyvalidloop-detectordataforthedesignatedthreelanesattheclosetdetectorstation.Thisrepresents12.8%ofthe9,341highwaycollisionsonthesixmajorOrangeCountyfreewaysthatarerecordedintheTASASdatabasefor1998.Forthis nalsample,theaveragedistancefromtheaccidentlocationtotheclosestdetectorstationis270mandthemediandistanceis190m.Fully78%ofthe1,191collisionswerelocatedwithin400m 0.25miles ofthedetectorstation,95%werelocatedwithin800m,and99%within1,200m.

AccidentCharacteristics

Availableinformationregardingthecharacteristicsofeachcolli-sionincluded: 1 thenumberofparties usuallyvehicles in-volved; 2 movementsofeachvehiclepriortocollision; 3 thelocationofthecollisioninvolvingeachparty; 4 theobject s struckbyeachvehicle;and 5 theseverity,asrepresentedbythenumbersofinjuredandfatallyinjuredpartiesineachinvolvedvehicle.Noinformationwasavailabletousconcerningdriversorvehiclemakesandmodels.ThecharacteristicsusedherearelistedinTable1.Foreachofthesecharacteristics,contingencytablechi-squaredtestsrevealedthatthereisnostatisticallysigni cantdifference atthe95%con dencelevel betweenthesubsetof1,191accidentsforwhichwehavetraf c owdataandthecomplementarysubsetof8,150accidentsonOrangeCountyfree-waysforwhichwehavenotraf cdata.

WeatherandLightingConditions

Includedinthedocumentationofeachcollisionisinformationonlighting,weather,andpavementconditions.Only13%ofallfree-wayaccidentsinOrangeCountyin1998occurredduringcondi-tionsofwetroads.Abreakdownoftheaccidentsbytheseenvi-ronmentalconditionsisdisplayedinTable2.Withtheexceptionof three dusk-dawnaccidentsonwetroads,thereareatleast30accidentsforeachcombinationofweatherandlighting,anumberjudgedtobeasuf cientcellsizeforanalyses.Thethreewetdusk-dawnaccidentsweredroppedfromtheanalyses,leavingsevensegmentsde nedbythecrosstabulationofTable2.

JOURNALOFTRANSPORTATIONENGINEERING©ASCE/JULY/AUGUST2003/343

上一页第2页

你可能喜欢

  • 道路通行能力
  • 元胞自动机
  • 美国数学建模
  • 美赛论文
  • 数学建模美赛
  • 篮球教练
  • 交通安全

热门文档

相关文档

站点地图 | 文档上传 | 侵权投诉 | 手机版
本站所有资源均来自互联网,本站只负责收集和整理,均不承担任何法律责任,如有侵权等其它行为请联系我们.
文档下载 Copyright 2013 doc.guandang.net All Rights Reserved.  -  email