Fully dynamic reorder policies with deep reinforcement learning for multi-echelon inventory management.
Patric HammlerNicolas RiestererTorsten BraunPublished in: Inform. Spektrum (2023)
Keyphrases
- inventory management
- reinforcement learning
- multi echelon
- supply chain
- lead time
- optimal policy
- lost sales
- assembly systems
- markov decision process
- vehicle routing
- inventory control
- state space
- total cost
- optimal control
- supply chain management
- control policies
- holding cost
- robust optimization
- inventory level
- markov decision processes
- dynamic programming