Chair Risk and Resilience of Complex Systems | LGI | CentraleSupélec
Predictive maintenance – from sensor measurements to prognostics, to maintenance planning
Dr. Mihaela Mitici, Assistant Professor
Algorithmic Data Analysis group, Faculty of Science
Utrecht University, Netherlands
Time: Thursday, May 04, 2023, 10:00 – 12:00 am CET
Online link: Click here to join the meeting on Microsoft Teams
Abstract In recent years, the health condition of assets is often monitored by sensors. The obtained condition-monitoring data has incentivized the development of data-driven Remaining-Useful-Life (RUL) prognostics. These RUL prognostics are further informing maintenance planning (predictive maintenance). In this presentation I will discuss the case of predictive maintenance for aircraft engines. Using Convolutional Neural Networks, probabilistic RUL prognostics are developed. These prognostics are the basis to identifying optimal maintenance times, following the theory of renewal reward processes. The results show that this predictive maintenance model leads to low maintenance costs and number of failures, without wasting the life of the assets. Overall, the proposed framework provides a roadmap from sensor measurements to data-driven prognostics to predictive maintenance planning.
Mihaela Mitici has a PhD in Stochastic Operations Research, Department of Applied Mathematics, University of Twente, the Netherlands. During 2016-2022 she was an Assistant Professor at Aerospace Engineering Faculty, TU Delft. Her expertise is in Operations Research, with a focus on stochastic processes, decision-making under uncertainty, applied probability theory, machine learning. Her main application domains are predictive maintenance and  
mobility. Her work has been awarded Best Paper Award 2nd prize at the 2022 Prognostics and Health Management Europe (PHMe) Conference, Thomas L. Fagan Award at the 2021 Reliability and Maintainability Symposium (RAMS), and Best Innovation Award at the 2021 AGIFORS Aircraft Maintenance Operations Special Session.
Chaire on risk and resilience of complex systems
Laboratoire Génie Industriel (LGI)
3 rue Joliot-Curie F-91192 Gif-sur-Yvette France