Computing optimal (R, s, S) policy parameters by a hybrid of branch-and-bound and stochastic dynamic programming.
Andrea VisentinSteve PrestwichRoberto RossiS. Armagan TarimPublished in: Eur. J. Oper. Res. (2021)
Keyphrases
- branch and bound
- stochastic dynamic programming
- approximate dynamic programming
- optimal solution
- linear program
- search algorithm
- lower bound
- column generation
- influence diagrams
- combinatorial optimization
- upper bound
- sensitivity analysis
- search space
- dynamic programming
- branch and bound procedure
- continuous state
- control policy
- experimental design
- step size
- asymptotically optimal
- linear programming
- policy iteration
- average cost
- control policies
- neural network
- reinforcement learning
- optimal policy
- state dependent
- decision making
- worst case
- finite horizon
- objective function
- decision problems
- markov decision processes
- metaheuristic