A mixed-effects model of two-phase degradation process for reliability assessment and RUL prediction

Abstract

Degradation modeling plays a critical role in reliability assessment and remaining useful life (RUL) prediction of products. In some practical applications, the degradation processes of products consist of two distinct phases, and hence traditional models are inadequate to characterize the degradation processes. To overcome this problem, we propose a mixed-effects model based on the Wiener process to capture the two-phase degradation pattern. More importantly, this novel model accommodates two significant aspects: phase correlation and unit heterogeneity. Using the new model, we develop an offline reliability assessment method using run-to-failure data from a population of products, and an online RUL prediction method for an operational product. Finally, a simulation study and a real case of batteries demonstrate the superiority of the proposed model and method.