Simulation of intermodal shared mobility in the San Francisco Bay using MATSim

Abstract

This paper proposes a generic shared mobility simulator based on an agent-based model - MATSim. The implemented module can simulate both human and engine powered vehicles, docked/station and dockless/free-floating services, and multiple operators. The module is employed in a case study based in the San Francisco Bay Area that focuses on potential benefits of docked bike-sharing services to serve as a first/last mile solution for public transit. The initial results indicate that the current BayWheels docked bike-sharing service can potentially reduce travel time for 1049 765 trips through intermodality compared to current transit travel times. By focusing our analysis on these trips that could benefit from bike-sharing as a first/last mile solution we present the quality of service indicators for different levels of adoption and two fleet sizes.