Using crowdsourced data to estimating exposure for bicycling safety

Authors: Colin Ferster*, University of Victoria, Trisalyn Nelson, Arizona State University, School of Geographical Sciences & Urban Planning, Laberee Karen, University of Victoria, Department of Geography, Winters Meghan, Simon Fraser University, Faculty of Health Sciences
Topics: Transportation Geography, Spatial Analysis & Modeling, Hazards, Risks, and Disasters
Keywords: Crowdsourcing, VGI, big data, active transportation
Session Type: Paper
Day: 4/9/2020
Start / End Time: 3:20 PM / 4:35 PM
Room: Director's Row E, Sheraton, Plaza Building, Lobby Level
Presentation File: No File Uploaded


Exposure matters for understanding bicycling safety in cities. Spatially and
temporally detailed exposure data is needed to understand bicycling safety,
yet bicycle count data is limited in cities. We present an approach for
estimating and correcting for exposure for bicycling incident hotspot maps
in Ottawa, Ontario, Canada. We obtained incidents from official (police
reports) and crowdsourced (BikeMaps.org) data, and we obtained exposure from
crowdsourced data (Strava Metro). We present raw and corrected bicycling
incident hotspot maps that offer complementary views of safety burden (where
the most incidents happen) and risk (where there are many incidents relative
to the number of bike trips). We found that correcting for exposure moved
safety hotspots from high comfort bike infrastructure (such as physically
protected cycle paths) to major streets with no bike infrastructure or
low-comfort bike infrastructure (such as shared lanes on major streets and
painted bike lanes that are not physically separated from traffic). We found
higher rates of incidents in winter and on weekdays. In contrast, we found
evidence of many bicycling trips on weekends, often on multi-use trails, and
very few incidents indicating that these are safe times and places for
bicycling in Ottawa. Crowdsourced exposure data can help provide a more
detailed understanding of bicycling safety in cities.

Abstract Information

This abstract is already part of a session. View the session here.

To access contact information login