Integrating Dynamic Ride-Sharing into an Agent-Based Traffic Simulation: A Sensitivity Analysis

This article presents an in-depth analysis of the potential of dynamic ride-sharing (DRS) systems using the agent-based simulation framework MATSim. A sensitivity analysis is performed based on multiple scenarios with different parameters of a newly developed dynamic ride-sharing extension, the potential of using DRS. One key finding is that the number of people who are willing to use DRS service significantly impacts its utilization among users. To ensure a positive reception, adjusting the DRS departure time to allow for a one-hour or more time window is essential. Additionally, our research highlights the importance of not excluding short trips from DRS options. Moreover, we observed that offering incentives to DRS drivers substantially increases the service’s adoption rate. However, this can lead to unintended consequences, such as a shift from active modes of transportation to DRS. Our study also revealed an increase in Vehicle Kilometers Traveled (VKT) after introducing DRS in the scenarios. However, a notable share of these VKT results from unmatched DRS drivers for which no statement can be made if they would genuinely opt for this mode. In the "maximum scenario" we identified a 1.8% share of DRS riders, representing a realistic potential for DRS under favorable conditions.