Simulation-based optimization of service areas for pooled ride-hailing operators

摘要

Dynamic ride hailing with passenger pooling has become a popular form of urban transport and is a growing sector around the globe. The area where these services operate is often limited to densely populated inner city districts, whereas non-pooled options are often available in larger areas. In this paper, we introduce a simulation-based methodology that allows to optimize the service area of a ride hailing service using an agent-based simulation and apply it to the taxi demand of Berlin, Germany. Three different criteria are used for the optimization, which take the average vehicle occupancy, the revenues collected per area or both into account. The results show that for the given parameters a service area that focuses on an extended central area and some areas around may be profit-maximizing for operators.