Bicycle Route Choice Data Collection using GPS-Enabled Smartphones

Abstract

The proliferation of consumer-grade smartphones with global position satellite (GPS) location capabilities opens a new data collection method for researchers. When confronted with a lack of data on bicycle routes preferred by local cyclists, the San Francisco County Transportation Authority (SFCTA) developed a freely downloadable iPhone/Android smartphone “app” called CycleTracks to collect actual bicycle routes traversed by city cyclists. Cooperation with local bicycle advocacy groups, along with social media and email campaigns, encouraged use of the app by regular citizenry. Several rounds of pre-release testing showed that making the app start up quickly, and minimizing battery usage during recording, were critical to getting good data. Once installed on a user's smartphone, a single “tap” would start and stop recording a bicycle trip; after completing a trip, the app automatically uploaded the track to a central database/web server, via the phone's built-in data plan. Approximately 5,000 usable bicycle trips were collected from hundreds of users in the region. Demographic data was optionally provided by some users, and showed a bias toward frequent cyclists, and toward male users (even more than cycling is already male-dominated in the region's most recent household travel survey). A bicycle route choice model developed using the data revealed sensitivity to slope, presence of bike lanes and/or bike route designations, trip purpose, and gender. The bike route choice model is now being integrated into San Francisco's regional travel model.