Rebalancing stochastic demands for bike-sharing networks with multi-scenario characteristics

Bike-sharing networks have become a carbon-emission and environmentally friendly form of transportation in recent years. However, the asymmetric demand patterns user behaviour, both temporally spatially, inevitably lead to an imbalance distribution shared bikes cities, thereby becoming greatest obstacle networks’ development. Based on real-world data cycling trips, we analyse challenging problem imbalanced bike from entire-city perspective, establishing that static rebalancing for whole city is stochastic variable with multi-scenario characteristics. On this basis, develop integer programming model consider multiple vehicles time-varying rental costs, alleviate distribution, while also analysing intrinsic properties such model. We further propose chance constraint model, optimising bike-sharing network through implementation various genetic algorithms employ block crossover mutation operators. reveal inability deterministic models addressing demands operational bike-sharing. In meantime, supported simulation, demonstrate proposed approach can resolve effectively efficiently, ensuring delivery high-level service across entire metropolitan city.