A simulation-based remaining useful life prediction method considering the influence of maintenance activities

As the key of the prevalent prognostics and health management, remaining useful life prediction has attracted considerable attentions during the past decades. However, almost all of the existing remaining useful life prediction methods were implemented under the premise that the deteriorating systems were not maintained over the whole life cycle. For the deteriorating systems experiencing maintenance activities during their life profiles, this paper presents a simulation-based remaining useful life prediction method taking the influence of maintenance activities into account. Specifically, the Wiener process with jumps is employed to model the degradation path of a deteriorating system, where the jump parts are used to characterize the influence of maintenance activities on the system degradation. The parameters in the degradation model are estimated by the maximum likelihood estimation method. To acquire the remaining useful life distributions of the deteriorating system, we design a simulation-based algorithm on the basis of the Markov Chain Monte Carlo method. Accordingly, the interested statistics associated with the remaining useful life can be obtained numerically. Finally, a numerical example is provided to show the implementation of the newly proposed remaining useful life prediction method.