Asset health reliability estimation based on condition data

This paper presents a method using recursive Bayesian analysis to estimate asset health reliability. The method enables reliability analysis and prediction using degradation condition monitoring data from an individual asset, rather than using failure data from a population of assets. A scheme was devised to assist the implementation of the method in both normal and abnormal zones during the whole asset operation cycle. An experiment on bearing life testing was employed to validate the method.