Asymptotic Optimality of Semi-Open-Loop Policies in Markov Decision Processes with Large Lead Times.
Xingyu BaiXin ChenMenglong LiAlexander L. StolyarPublished in: Oper. Res. (2023)
Keyphrases
- markov decision processes
- open loop
- asymptotic optimality
- optimal policy
- asymptotically optimal
- lead time
- holding cost
- closed loop
- lost sales
- infinite horizon
- average cost
- order quantity
- sufficient conditions
- control system
- finite state
- inventory systems
- decision problems
- state space
- finite horizon
- markov decision process
- dynamic programming
- reinforcement learning
- reward function
- supply chain
- partially observable markov decision processes
- control policies
- long run
- single item
- average reward
- policy iteration
- single stage
- inventory level
- random variables
- state dependent
- service level
- lot sizing
- multistage
- markov decision problems
- initial state
- arrival rate
- action space
- reinforcement learning algorithms
- sample path
- total cost
- queue length
- special case