Mining vehicle classifications from the Columbus Metropolitan Freeway Management System.

PDFTech ReportFHWA/OH-2015/1State Job Number 134696Vehicle classificationFreeway management systemsLoop detectorsTraffic surveillanceData miningTraffic dataOhioOhio. Dept. of Transportation. Office of Statewide Planning and ResearchCoifman, BenjaminHuang, BoRedmill, KeithWu, MoOhio State UniversityUnited States. Federal Highway AdministrationUS Transportation CollectionVehicle classification data are used in many transportation applications, including: pavement design,environmental impact studies, traffic control, and traffic safety. Ohio has over 200 permanent count stations,supplemented by many more short-term count locations. Due to the high costs involved, the density ofmonitoring stations is still very low given the lane miles that are covered. This study leveraged the deployeddetectors in the Columbus Metropolitan Freeway Management System (CMFMS) to collect and analyzeclassification data from critical freeways where the Traffic Monitoring Section has not been able to collect muchclassification data in the past due to site limitations. The CMFMS was deployed in an unconventional mannerbecause it included an extensive fiber optic network, frontloading most of the communications costs, and ratherthan aggregating the data in the field, the detector stations sent all of the individual per-vehicle actuations (i.e.,PVR data) to the traffic management center (TMC). The PVR data include the turn-on and turn-off time forevery actuation at each detector at the given station. Our group has collected and archived all of the PVR datafrom the CMFMS for roughly a decade. The PVR data allows us to reprocess the original actuationsretroactively. As described in this report, the research undertook extensive diagnostics and cleaning to extractthe vehicle classification data from detectors originally deployed for traffic operations.The work yielded length based vehicle classification data from roughly 40 bi-directional miles of urbanfreeways in Columbus, Ohio over a continuous monitoring period of up to 10 years. The facilities span I-70, I-71, I-270, I-670, and SR-315, including the heavily congested inner-belt. Prior to this study, these facilitiespreviously had either gone completely unmonitored or were only subject to infrequent, short-term counts