Travel behavior simulation based bike-sharing station distribution

Abstract

Efficient bike-sharing station planning requires balancing user demand, accessibility, and cost constraints in dynamic urban environments. Traditional planning methods often rely on aggregate demand data, overlooking the intricate, non-linear interactions between individual user behaviors and station distribution. This study employs an agent-based MATSim model to explore the impact of hub design efficiency on station size, scale, and user travel patterns. By transforming macroscopic trip data into disaggregated individual trips, we simulate real-world travel behaviors in Vienna’s bike-sharing system. We assess three staged implementation plans (40%, 70%, and 100% deployment) to evaluate demand shifts, hub utilization, and profitability. Our findings reveal how strategic station placement influences mode choice, last-mile connectivity, and network efficiency. This research contributes to equitable and cost-effective bike-sharing expansion by providing insights into optimizing station configurations while adapting to evolving urban mobility needs.