Physical Asset management : DAta, models, health aWAre-decisioN – PADAWAN

The aim of this project is to develop mathematical methods to help manage instrumented industrial assets undergoing maintenance. It aims to exploit multiple data sources in all their complexity, taking into account their heterogeneity, quality and cost. It proposes an integrated approach linking stochastic modeling, statistical calibration and decision-making. The project will be based on general probabilistic modeling, taking into account the characteristics of industrial asset degradation processes: multidimensional degradation, imperfect maintenance effects, usage and environmental conditions, individual unobserved heterogeneity. It will develop appropriate statistical inference methods: estimation, goodness-of-fit testing and model selection for missing or censored data, with measurement error, under general observation schemes. It will propose dynamic decision-making approaches for prescriptive maintenance, but also for optimizing the monitoring and use of the assets under consideration, making the best use of available data and monitoring the evolution of the health status of the systems under consideration, while taking into account various constraints. Finally, the proposed methods and tools will be implemented in open source software.