Deciphering urban cycling: Analyzing the nonlinear impact of street environments on cycling volume using crowdsourced tracker data and machine learning

Abstract

Cycling mitigates urban development-related traffic and environmental issues and benefits human health. However, exploring the nonlinear associations between urban environmental factors and cycling remains challenging. Moreover, the potential of crowdsourced data like Strava Heatmap for cycling research has rarely been validated. Using Melbourne as a case study, we assessed the association between urban environmental attributes and cycling amount through street view images and artificial intelligence techniques. The results indicate that proximity to blue spaces is the most significant factor in promoting cycling amount. Additionally, road network density, sky openness, and distance to green spaces each have an optimal threshold. Lastly, built environment features, landscape features, and perceived environment are all associated with cycling amount, validating the inclusion of both subjective and objective environmental measures in cycling research. These findings provide insights and empirical evidence for policymakers in designing bicycle-friendly urban environments.