Reliability prediction using multivariate degradation data

Abstract

This paper presents a general degradation modeling and analysis approach involving multiple degradation measures. We first explore the correlation between the degradation measure and the failure event by introducing a probabilistic measure. A logistic function is used to define this probabilistic measure and assess the likelihood of a failure event given the degradation level. When the stochastic stress condition is considered this probabilistic measure is an approximation of the likelihood of a failure event. We then propose a state-space model to describe the evolution of the degradation process by incorporating both the degradation dynamics and random stress effects. Finally the degradation dynamics is used to predict the reliability function. The practical use of the proposed method is demonstrated by a case study.