Geospatial Analysis of Bicycle Network “Level of Stress”, Bicycle Crashes and the Geo-coded Pavement Conditions for Risk Factors
PI: Haizhong Wang (OSU), Haizhong.Wang@oregonstate.edu
Dates: 07/01/2013 – 8/31/2015
Final Project Report: PacTrans-35-OSU-Wang
Safety remains a problem on U.S. roadways, with more than 32,000 fatalities, 2.2 million injuries and 6 million crashes each year. Less than two percent of motor vehicle crashes deaths are bicyclists. The loss of 677 lives in bicycle/motor vehicle crashes in the U.S. in 2011, although lower than the 830 fatalities in 1995, is still on the rise just a few years ago. Cities and counties in the United States have made small progress promoting bicycling by developing painted bike lanes, separate bicycle-only highways, bike share programs and incentives for businesses that encourage employees to bike to work. Recent research proposes evaluating urban bicycle treatments of this kind by how to reduce bicycle crashes and the stress-level for cyclists on road networks (Mekuria, Furth and Nixon 2012). This report proposes a four-step classification system for streets, roads, highways and bike paths based on research into consumer categories of bicycle users. The system ranges from stress “level one” streets where vehicles travel under 30 miles an hour and provide for good vehicle visibility such that small children can safely bike on them, to stress “level four” streets with 40 plus mile per hour vehicle traffic only traversed by professional cyclists. This proposed study merges existing research on spatial pattern of bicycle crashes and pavement conditions with recently proposed criteria for evaluating bicycle network “level of stress” connectivity and how it is connected to the casual patterns of bicycle crashes. This research combines demand based household travel data from the 2011 Oregon Household Activities Survey (OHAS) and Oregon crash database with supply based geographic information system (GIS) analysis of the City of Corvallis, Oregon’s bicycle infrastructure network to evaluate the impact of bicycle infrastructure investment on household behavior, and the statewide crash database. Multiple measures of connectivity to bicycle networks are evaluated, including level of service directly around households, number of destinations for given trip purposes within specified ranges of the households, and the number of bicycle lane miles connected to the household through low-stress networks. The research team will utilize demographic information from the OHAS dataset to minimize issues of self-selection into specific, bicycle friendly units by individuals of certain ages and family structures and GIS-based Oregon statewide crash database.
