Scalable Ride-Sourcing Vehicle Rebalancing with Service Accessibility Guarantee: A Constrained Mean-Field Reinforcement Learning Approach
Researchers have developed a new approach using constrained mean-field reinforcement learning to optimize vehicle rebalancing for ride-sourcing platforms. This method models vehicle interactions with the overall distribution rather than individual vehicles, significantly reducing computational complexity and allowing scalability to tens of thousands of vehicles. The approach also incorporates an accessibility constraint to ensure equitable service distribution across different geographic regions, balancing demand fulfillment with fair supply coverage. AI
IMPACT This research could lead to more efficient and equitable operations for ride-sharing services by optimizing vehicle distribution.