Degradation change point and its application in CBM decision

The relationship between the mean degradation measurement and time is usually monotonically and non-linearly increasing or decreasing. There exists a time point after which the degradation rate rapidly changes. We call this point as degradation change point. It is useful to identify the degradation change point since the probability of a functional failure will quickly increase after an item is used beyond the change point. This paper presents a data-driven approach to find the degradation change point. The approach fits a piecewise model to the observed degradation data and the change point is derived from the fitted model. Two real-world examples are used to illustrate the proposed approach. The usefulness of the degradation change point is also discussed in a CBM setting.