This thesis explores the design, implementation, and evaluation of an on-demand ride-pooling system within a multimodal microscopic traffic simulation environment. As cities continue to face challenges such as congestion, emissions, and growing mobility demands, there is an increasing need for flexible and efficient transport solutions. Ride-pooling, where multiple passengers with similar origins and destinations share a vehicle, offers the potential to support more sustainable urban mobility. This research investigates how such systems can be effectively modeled using microscopic simulation, while also incorporating efficient vehicle-request matching methodologies. The research begins with an extensive literature review covering on-demand mobility systems, shortest route algorithms, vehicle routing problems, and previous studies on ride-hailing and ride-pooling. The review provides essential insights into system implementation, including matching strategies, objective function perspectives and key performance indicators (KPIs). Notably, it also highlights that the modeling of on-demand mobility within microscopic simulations has remained largely unexplored. Two vehicle-request matching algorithms, spontaneous and batch assignment, are presented in detail and illustrated with examples. Building on this, the application of an on-demand ride-pooling system in the Vissim microsimulation tool is demonstrated. The steps necessary to reflect the dynamic nature of ride-pooling in such a simulation are outlined, including an external control via the COM interface. A further contribution is the development of an interface that enables an automatic connection between a camera-based traffic data collection system with the simulation, simplifying the essential data entry in the simulation. Extensive simulation experiments are performed for a typical weekday on the campus of the University of the Bundeswehr Munich, providing a realistic test environment. A broad set of KPIs including service rate, sharing rate, mean waiting time, mean detour time, total mileage, mean mileage, vehicle occupancy, mileage index, and empty mileage ratio, are systematically analyzed. Several simulation runs evaluate the impact of operational parameters on system performance, while others investigate the temporal evolution of KPIs throughout the entire simulation day. Additional scenarios assess the effects of travel demand, cost function weighting, and batch periods. Furthermore, comparisons between ride-hailing and ride-pooling operations, as well as between different matching strategies, are conducted. The findings confirm that the developed matching methodologies can be effectively applied within a microscopic simulation, and offer valuable insights into the operational performance of the on-demand ride-pooling system. In conclusion, this thesis provides both methodological and practical contributions to the modeling and evaluation of ride-pooling services within a microscopic traffic simulation. The insights and methodologies derived from this research are particularly valuable for researchers and practitioners who aim to implement on-demand systems in microsimulation frameworks. While the proposed methodology has been applied in a small-scale model, extending this approach to larger areas presents a promising direction for future research.     «

This thesis explores the design, implementation, and evaluation of an on-demand ride-pooling system within a multimodal microscopic traffic simulation environment. As cities continue to face challenges such as congestion, emissions, and growing mobility demands, there is an increasing need for flexible and efficient transport solutions. Ride-pooling, where multiple passengers with similar origins and destinations share a vehicle, offers the potential to support more sustainable urban mobility. T...     »