A Deep Q-Learning Approach for Continuous Review Policies with Uncertain Lead Time Demand Patterns.
Jianpin ZhouShuliu ZhangYingtang LiPublished in: ISCID (1) (2018)
Keyphrases
- lead time
- holding cost
- lost sales
- optimal policy
- demand distributions
- inventory systems
- supply chain
- periodic review
- inventory control
- single item
- stochastic lead times
- replenishment policy
- service level
- order quantity
- infinite horizon
- lot sizing
- planning horizon
- inventory level
- reinforcement learning
- stochastic inventory
- assembly systems
- total cost
- stochastic demand
- random variables
- state space
- single product
- lot size
- single stage
- inventory models
- customer demand
- base stock policy
- decision making
- learning algorithm
- sample path
- function approximation
- expected cost
- model free
- revenue management
- long run
- single period
- action space
- inventory costs