Emergence of scaling in dockless bike-sharing systems for bike choice behavior

Dockless bike-sharing platforms, which offer better accessibility and flexibility for users, play an important role in the transition towards green transportation. However, unlocking their full potential requires a deeper understanding of mobility patterns of both users and shared dockless bikes. While fundamental laws of human mobility have been extensively studied, we are still lacking a comprehensive analysis on mobility patterns of shared conveyances. By treating each bike as an individual entity, we analyze mobility patterns of bikes at both individual level and collective level. For exploration and visitation dynamics, bikes exhibit different scaling exponents than cyclists, which also vary across cities. In terms fo gyration and average travel distance, patterns of bikes can be even qualitatively different across cities, which also shows distinctions from mobility patterns of users. Despite these quantitative and qualitative variations, mobility patterns of bikes at both individual and collective levels can be well explained by bike choice behavior of users, which follows a Zipfian law. This suggests that, rather than choosing among bikes with the best conditions, users have a non-negligible probability of choosing bikes with less good conditions. Despite great diversities across cities, such Zipfian pattern remain stable. Our framework holds broad implications in shared mobility, which can contribute to develop a greener, healthier, and more sustainable transportation.