Introducing an AI based framework for Maintenance Decision-Making (MDM) and Optimization
In our recent paper ‘Reinforcement learning for maintenance decision-making of multi-state component systems with imperfect maintenance’ the AI-PROFICIENT team proposes an artificial intelligence (AI) based framework for maintenance decision-making (MDM) and optimization of multi-state component systems with imperfect maintenance.
In specific, Van-Thai Nguyen, Phuc Do, Alexandre Voisin and Benoit Iung’s framework consists of two main phases:
- The first aims at constructing Artificial Neural Network (ANN) based predictors for system’s reliability and maintenance cost. Results show that ANN is suitable to the reliability, maintenance cost forecasting.
- The second refers to the use of Deep Reinforcement Learning (DRL) algorithms to optimize maintenance decision, which can deal with large scale applications. As analysed, DRL is a potentially powerful tool for MDM and optimization, considering the effects of imperfect maintenance.
You can read the full paper here.