A hybrid prognostics and health management approach for condition-based maintenance

Condition-based maintenance (CBM) is an efficient proactive maintenance strategy based on actual conditions obtained from in-situ, non-invasive tests, and operating measurement. In recent year, prognostics and health management (PHM) has emerged as one of the key enablers for achieving efficient system-level maintenance and for lowering life-cycle costs. This paper overviews methodology of physics-of-failure (PoF) approach and categorizes data-driven approach for the PHM application, summarizes their advantages and disadvantages respectively, and presents a hybrid prognostics approach which incorporate both the advantages of PoF and data-driven approaches for condition-based maintenance.