Jump Markov Linear Systems for deterioration modeling and Remaining Useful Life estimation
Résumé
Deterioration modeling and Remaining Useful Life (RUL) estimation plays nowadays an important role in implementing a predictive maintenance program since it helps to predict early the failure time caused by an incipient deterioration, and thus to provide sufficient time for maintenance crew to act before the actual failure eventually occurs. One well-identified challenge in deterioration modeling is that equipment does not always deteriorate following only one single mechanism, instead several deterioration modes can co-exist in competition due to the dynamics and variability of the operating and environmental conditions. In the literature, this co-existence issue has been recently studied but only discrete-state models with no transition allowance between the deterioration modes have been considered. Real equipment, however, can experience and switch between different deterioration mechanisms during its whole life. This paper proposes thus the use of the Jump Markov Linear Systems (JMLS) model to develop a solution to the above limitation. To illustrate the proposed approach, a numerical study is given.