Bicycle Safety Analysis: Crowdsourcing Bicycle Travel Data to Estimate Risk Exposure and Create Safety Performance Functions

Abstract

Around 700 bicycle fatalities happen every year, and the number has steadily increased since 2009. At the same time Seattle, Wash., and Portland, Ore., are promoting cycling as a healthy transportation mode, so bicycle safety has become a more urgent concern for the Pacific Northwest area. The Highway Safety Manual provides an evidence-based approach – safety performance function (SPFs) to evaluate safety for common traffic, but not, however, for bicycles. Therefore, a data-driven and evidence-based bicycle safety evaluation method is needed specifically for the Pacific Northwest region. Based on the first bicycle SPFs created by Krista Nordback in 2013, we used STRAVA bicycle count data (a type of crowdsourced bicycle travel data), other traffic count data, and bicycle crash data to establish Pacific Northwest SPFs in terms of bike count and crash frequency for road intersections. The SPFs demonstrated the relationship between crash frequency and traffic and bike volumes. The intersections with higher traffic volumes had higher bicycle crash frequencies. In addition, this project improved the usability of a GIS tool that had been created during a previous PacTrans project to estimate bicycle exposure. States DOTs and other agencies can use the SPFs to screen and identify previous bicycle black spots in the Pacific Northwest region in order to optimize safety investments.