PHM Integration with Maintenance and Inventory Management Systems

Abstract

Prognostic techniques are intricately tied to the physics of incipient-fault-to-failure progression, and hence most prognostics research has focused on developing techniques for a range of components such as rotating machinery parts. The research and development of such techniques has relied on the theories of material science, structural mechanics, domain expertise, as well as empirical studies such as accelerated run-to-failure testing. Even after prognostic models have been developed and operationally validated for various components of a system, the challenge remains how prognostic assessments from individual components of a system (such as an aircraft engine) should be used to make maintenance and logistics decisions. In this paper, we describe an integration process where the primary focus is on bridging the gap between the individual component prognosis and the system-level reasoning required to support maintenance and inventory management decisions. The research involves integration of component health assessment with an information fusion mechanism that operates in conjunction with a higher-level reasoning engine which utilizes system-level structural and functional dependencies. The higher-level reasoning engine generates a system availability analysis that leads directly to actionable tasks for the inventory and maintenance management decision support systems. The inventory management decision support system involves predicting the spares requirements, and when this is integrated with remote health monitoring and intelligent diagnostics and prognostics, it can assess different sparing allocation schemes, and maximize system availability within budget constraints.