Optimal Policies and Approximations for a Serial Multiechelon Inventory System with Time-Correlated Demand.
Lingxiu DongHau L. LeePublished in: Oper. Res. (2003)
Keyphrases
- optimal policy
- inventory level
- lost sales
- inventory systems
- production inventory
- single stage
- infinite horizon
- markov decision processes
- state space
- decision problems
- state dependent
- finite horizon
- dynamic programming
- long run
- periodic review
- stochastic demand
- reinforcement learning
- multistage
- inventory models
- inventory management
- finite state
- inventory control
- customer demand
- single product
- lead time
- policy iteration
- sufficient conditions
- dynamic pricing
- sample path
- production rate
- policy evaluation
- order quantity
- single period
- markov decision process
- fixed cost
- dynamic programming algorithms
- average reward
- holding cost
- single item
- asymptotically optimal
- initial state
- machine learning
- average reward reinforcement learning
- demand distributions
- serial inventory systems
- periodic review inventory
- base stock policies
- ordering cost
- partially observable markov decision processes
- reward function