Benefits of flexible boarding locations in on-demand ride-pooling systems

On-Demand Ride-Pooling services can provide a convenient mobility mode while decreasing vehicle kilometers in a transportation system. Nevertheless, the ultimate impact of such a service depends on a variety of different parameters and settings defining the service. In this study, a simulation framework is proposed which is designed to investigate of the influence of boarding locations on such a service, i.e. the locations in a street network, where an operator is allowed to perform boarding processes of customers. This study (1) investigates the influence of different boarding location densities and (2) proposes an optimization framework which is able to optimize boarding locations dynamically given that customers are required to walk short distances to/from their pick-up/drop-off location. The simulation framework is tested on a case study for Munich, Germany. Results indicate that dynamic boarding location optimization can increase the number of customers that can be served by up to 4% compared to a quasi door-to-door service while decreasing fleet kilometers traveled by up to 4.8%. Although customers have to walk additional 1.5 min compared to the quasi door-to-door service, the aggregated travel time (ride time and walking time) increases only marginally.