RP-DQN: An application of Q-Learning to Vehicle Routing Problems.
Ahmad BdeirSimon BoederTim DerneddeKirill TkachukJonas K. FalknerLars Schmidt-ThiemePublished in: CoRR (2021)
Keyphrases
- vehicle routing problem
- reinforcement learning
- vehicle routing problem with time windows
- metaheuristic
- tabu search
- state space
- routing problem
- learning algorithm
- test instances
- benchmark problems
- traveling salesman problem
- benchmark instances
- waste collection
- guided local search
- multi depot
- optimal policy
- travel time
- particle swarm optimization
- np hard
- variable neighborhood search
- memetic algorithm
- greedy randomized adaptive search procedure
- knapsack problem
- combinatorial optimization
- search strategies
- constraint satisfaction
- path relinking
- optimal solution
- knn
- neighborhood search