Optimal Policies for a Capacitated Two-Echelon Inventory System.
Rodney P. ParkerRoman KapuscinskiPublished in: Oper. Res. (2004)
Keyphrases
- optimal policy
- lot sizing
- multistage
- decision problems
- markov decision processes
- state space
- reinforcement learning
- dynamic programming
- infinite horizon
- finite horizon
- multi item
- np hard
- long run
- production inventory
- finite state
- state dependent
- markov decision process
- serial inventory systems
- initial state
- average reward
- control policies
- average cost
- sufficient conditions
- lost sales
- average reward reinforcement learning
- policy iteration
- dynamic programming algorithms
- data mining
- partially observable markov decision processes
- reward function
- mixed integer
- search algorithm