Solving Vehicle Routing Problem Using Reinforcement Learning
摘要
With the growing demand in the logistics sector and the use of technology, vehicle routing problems (VRPs) have become very useful in solving real-life scenarios. Dynamic vehicle routing problem (DVRP) differs from classical vehicle routing in that all information is not known beforehand but arrives periodically during the execution of the planned route. In this paper, we propose reinforcement learning (RL) to solve the vehicle routing problem and how to model DVRP by RL.