Simulating and Evaluating Rebalancing Strategies for Dockless Bike-Sharing Systems

Abstract: Following the growth of dock-based bike sharing systems as an eco-friendly solution for transportation in urban areas, Dockless systems are revolutionizing the market for the increased flexibility they offer to users. Bike redistribution is a common approach to improve service, and there exists extensive research considering static and dynamic rebalancing strategies for dock-based systems. We approach the dockless problem by defining abstract stations from trip start and end location frequency. This paper offers an optimizing Mixed Integer Program framework to model the effects of various bike repositioning strategies for dockless systems. We process 30 days worth of Singapore-based dockless bike data from September 2017 to extract trips. Pairing our mixed integer program with a demand model built from the processed data, we unveil trends between fleet size, lost demand, and magnitude of repositioning proper to the repositioning strategy employed. We also show that increasing repositioning potential does not always improve service performance.