Leveraging passively generated big data

Hans-Heinrich Schumann, He Haitao, Asya Natapov, Mohammed Quddus

Understanding e-scooter riders’ behaviours as well as the appropriate utilisation of the rich data produced by the mode is a challenge, with negative implications for their successful integration in the transport and urban planning ecosystem. Linking the theory of natural movement with big data analytics, this project evaluates the impact of urban morphology, socio-demographic information, and infrastructure on e-scooter rider’s behaviour. Based on findings from three case study locations, the project demonstrates that urban morphology parameters to e-scooter flow and route choice models significantly improves their predictive power, analyses the time-dependent behaviour of e-scooter riders, and quantifies the multi-modal benefits of dedicated cycling infrastructure, finding that it reduces perceived e-scooter travel distance by at least 51%.