Assessing uncertainty from data collection to maintenance optimization

In this paper, we will propose a framework to perform the optimization of periodical maintenance tasks for a production line, with a specific viewpoint on uncertainty issues from the modelling step to the analysis of numerical results. From a structured log file of operational data, we build a reliability-based model (block diagram) that is used to optimize the parameters of the maintenance policies through Monte Carlo simulations. The model is determined by using every data source available (Computerized Maintenance Management System hierarchy and failure mode classification especially).