Impact of data source on travel time reliability assessment.

PDFTech ReportUMD-2013-01DTRT12-G-UTC03Travel timeData collectionProbe vehiclesBluetooth technologyPerformance measurementReliabilityReal time informationHigh occupancy vehicle lanesMarylandMid-Atlantic Universities Transportation CenterHaghani, AliZhang, YanruHamedi, MasoudUniversity of Maryland (College Park, Md.). Dept. of Civil and Environmental EngineeringUnited States. Dept. of Transportation. Research and Innovative Technology AdministrationUS Transportation CollectionTravel time reliability measures are becoming an increasingly important input to the mobility andcongestion management studies. In the case of Maryland State Highway Administration, reliabilitymeasures are key elements in the agency\ue2\u20ac\u2122s Annual Mobility Report that is used for both transportationplanning and operations. Recent advancements in vehicle tracking technologies have provided bothprivate sectors and transportation agencies with multiple technologies for travel time data collection.This paper investigates the effect of data source selection on travel time reliability assessment. One yeardata from two independent sources, probe and Bluetooth, on two major freeway corridors in Marylandare used for travel time reliability analysis. Bluetooth sensors are permanently installed on multiplepoints of these two corridors and probe data are provided by INRIX. A statistical method is applied tocompare reliability measures obtained from both sources. Several travel time reliability performancemeasures are studied. Results show that some reliability measures are more sensitive to the data sourcethan the others. It is also shown that performance measures for HOV and general purpose lanes must becalculated separately