Methods for estimating bicycling and walking in Washington state.

PDFTech ReportWA-RD 828.1Agreement GCB 1699, Task 3BicyclingWalkingEstimatingData collectionVehicle miles of travelPerson miles of travelPedestrian countsTraffic countsWashingtonWashington (State). Dept. of Transportation. Research OfficeNordback, KristaSellinger, MichaelOregon Transportation Research and Education ConsortiumUnited States. Federal Highway AdministrationUS Transportation CollectionThis report presents the work performed in the first and second phases in the process of creating a method tocalculate Bicycle and Pedestrian Miles Traveled (BMT/PMT) for the state of Washington. First, we recommendimprovements to the existing Washington State Bicycle and Pedestrian Documentation Program to provide data forBMT/PMT estimates, including expanding the program geographically and installing permanent automated bicycleand pedestrian counters to complement the short duration count program. The method to estimate BMT/PMT relieson the assumption of a stratified random sample drawn from the set of all roads and paths divided into 16 groups.These groups are based on three spatial attributes, which were gathered from a review of the literature:\ue2\u20ac\ua2 Level of urbanism (2 categories): Urban and Rural\ue2\u20ac\ua2 Facility type (2 categories): Highway/Arterial and Other\ue2\u20ac\ua2 Geographic/climatic regions (4 regions): Coast Range, Puget Lowland, Cascades, Eastern WashingtonThis report describes the first steps being taken toward the goal of computing this metric. Count data from Seattle,Olympia, and the State\ue2\u20ac\u2122s Count Program have been gathered. To account for temporal variation, seasonal, daily andhourly adjustment factors have been computed based on one year of count data collected from the Fremont Bridgein Seattle. The short duration count sites have been grouped by the attributes described above, though most fall intojust two groups: Puget Lowland Urban Arterial/Highway and Puget Lowland Urban Local/Collector/Path. Little orno data are available in most of the other groups. The roads in the state have also been divided into these 16 groupsin order to compute total centerline miles for each group. This report outlines a sample-based method that could beused to compute BMT/PMT for the state and identifies both the data available for such a computation as well as thedata gaps. It also suggests other methods that could also be used to estimate BMT/PMT to compare to the count-based method