Collaborative Vehicle Routing
Collaborative vehicle routing (CVR) optimizes the delivery of goods or services by coordinating multiple vehicles, aiming to minimize costs, emissions, and congestion while ensuring fairness among participating entities. Current research emphasizes developing efficient algorithms, including reinforcement learning and mathematical programming models like tabu search and mixed-integer programming, to solve the complex routing and resource allocation problems inherent in CVR, particularly in dynamic environments with time windows and varying priorities. These advancements are significant for improving logistics efficiency across various sectors, from autonomous vehicle networks to disaster response and hospital operations, by enabling better coordination and resource utilization.
Papers
Fair collaborative vehicle routing: A deep multi-agent reinforcement learning approach
Stephen Mak, Liming Xu, Tim Pearce, Michael Ostroumov, Alexandra Brintrup
Coalitional Bargaining via Reinforcement Learning: An Application to Collaborative Vehicle Routing
Stephen Mak, Liming Xu, Tim Pearce, Michael Ostroumov, Alexandra Brintrup