State-Dependent Parameter Tuning of the Apparent Tardiness Cost Dispatching Rule Using Deep Reinforcement Learning.
Byungwook MinChang Ouk KimPublished in: IEEE Access (2022)
Keyphrases
- parameter tuning
- state dependent
- optimal policy
- reinforcement learning
- steady state
- markov decision processes
- wafer fabrication
- state space
- parameter settings
- infinite horizon
- queueing networks
- holding cost
- markov chain
- production scheduling
- multistage
- long run
- queue length
- single server
- arrival rate
- dynamic programming
- machine learning
- stationary distribution
- finite state
- learning algorithm
- asymptotically optimal
- initial state
- inventory level
- simulated annealing