Reinforcement Learning based on Stochastic Dynamic Programming for Condition-based Maintenance of Deteriorating Production Processes.
Hasan RasayFarnoosh NaderkhaniAmir-Mohammad GolmohammadiPublished in: ICPHM (2022)
Keyphrases
- stochastic dynamic programming
- production processes
- reinforcement learning
- approximate dynamic programming
- continuous state
- production process
- function approximation
- dynamic programming
- sufficient conditions
- robot navigation
- linear program
- experimental design
- state space
- action space
- influence diagrams
- partially observable markov decision processes
- step size
- reinforcement learning algorithms
- finite state
- optimal control
- state dependent
- control policy
- learning algorithm
- machine learning
- model free
- production system
- planning problems
- control policies
- markov decision processes
- markov chain
- multi agent