Condition-based or predictive maintenance techniques have proven to be effective to improve production performance and systems availability. The main topic of this communication is related to stochastic modelling and optimization of condition-based maintenance. We consider a continuous state s ystem. Its state is modelled as a continuous random variable which can only be observed by inspection. In this communication, we propose a predictive maintenance policy which jointly indicates, at each inspection and according to the observed system state value, both whether a preventive maintenance is necessary and when the next inspection should be done. The objective is to minimize the long run average cost incurred by inspections, preventive maintenance and unexpected break-downs.
展开▼