RP-DQN: An application of Q-Learning to Vehicle Routing Problems
This addresses routing optimization for logistics and transportation, with incremental improvements in method application.
The paper tackles complex vehicle routing problems by applying Q-learning with an improved state representation, achieving state-of-the-art performance for autoregressive policies on the CVRP and being the first to apply machine learning methods to the MDVRP.
In this paper we present a new approach to tackle complex routing problems with an improved state representation that utilizes the model complexity better than previous methods. We enable this by training from temporal differences. Specifically Q-Learning is employed. We show that our approach achieves state-of-the-art performance for autoregressive policies that sequentially insert nodes to construct solutions on the CVRP. Additionally, we are the first to tackle the MDVRP with machine learning methods and demonstrate that this problem type greatly benefits from our approach over other ML methods.