Where the Action is: Let's make Reinforcement Learning for Stochastic Dynamic Vehicle Routing Problems work!
Florentin D. HildebrandtBarrett W. ThomasMarlin W. UlmerPublished in: CoRR (2021)
Keyphrases
- stochastic dynamic
- vehicle routing problem
- reinforcement learning
- vehicle routing problem with time windows
- action selection
- metaheuristic
- action space
- tabu search
- waste collection
- routing problem
- multi depot
- test instances
- benchmark problems
- traveling salesman problem
- benchmark instances
- neighborhood search
- state space
- variable neighborhood search
- memetic algorithm
- combinatorial optimization
- markov decision processes
- knapsack problem
- guided local search
- optimal policy
- particle swarm optimization
- np hard
- greedy randomized adaptive search procedure
- path relinking
- search algorithm
- optimal solution