Mobility on Demand

Mobility-on-demand (MOD) systems aim to optimize transportation efficiency and accessibility through dynamic, often shared, vehicle services. Current research emphasizes improving service scheduling and vehicle routing using advanced algorithms like column generation enhanced by graph neural networks and reinforcement learning, often incorporating considerations for real-time constraints and demand prediction. These advancements are crucial for enhancing the sustainability and resilience of urban transportation networks, particularly in addressing challenges like climate change mitigation, equitable access, and efficient resource allocation in both routine and crisis situations. The field is also actively developing robust simulation platforms for testing and validating new algorithms and operational strategies.