Street Morphology And Travel By Dockless Shared Bicycles In Beijing, China

App-based, dockless bicycle-sharing systems (DBSSs) have proliferated across cities in China since their inception in 2016, managed by the major Internet service providers and with central government support, in an effort to help grow nonmotorized transport. Although offering more generalized use and extended range, DBSS exhibits great variability in local use rate, suggesting that control of certain local environmental attributes could be instrumental in supporting more widespread bicycle use. This article is concerned with the relationship between street morphology defined at the street segment level and related shared bicycle use rate. Mobike trip data for a 10-day period in May, 2017 - 3.09-m trip records with origins and destinations - were accessed. Multilevel regression analysis shows that generated trips and travel distance are positively related to both betweenness and maximum convex hull radius but are negatively related to diversion ratio. All street morphological metrics show significant effects across spatial scales. Number of shared bicycle trips is positively associated with floor area ratio but negatively associated with intersection density. Land use mix is positively related to generated trips but negatively related to biking distance, while destination accessibility shows the inverse relationship. The results provide insight into where trips using DBSS are likely to be found, given the street pattern, but also how their wider and more generalized use could be promoted using urban planning.