Designing for health: a methodology for integrated diagnostics/prognostics

A critical element of system readiness is the effectiveness of integrated diagnostics and prognostics. Errors in detection and isolation of failures cause unnecessary maintenance actions requiring additional troubleshooting time and replacement of could not duplicates (CNDs). Integrated diagnostics/prognostics is achieved through a systems engineering closed loop process from start to finish. Every step from conceptual phase through program shut down thrives on an integrated approach to insure maximum coverage of faults and unambiguous isolation while minimizing false alarms and re-test okays (RTOKs). An integrated health management system (HMS) methodology, proposed in this paper, connects functionality, failure modes and diagnostics/prognostics under one umbrella providing a conduit for tight traceability from requirements through design, analysis, integration, verification and validation, factory testing, and fielding while encouraging maturation through data collection. This framework minimizes errors between diagnostic/prognostic analyses and actual performance, maximizes test vertically, and paves the way for a lower risk fielded product. Iterative analyses and trade studies during the concept and development phases optimize diagnostic/prognostics approaches and architecture. Analyses and simulations during the development phase optimize health sensor selection, sensor placement, test effectiveness, and test strategies (i.e. embedded vs. external, test flow, test type etc.). Analyses and test data provide the foundation for the fidelity and accuracy of a diagnostic/prognostic reasoner, which selects test sequence, filters information and reports faults and their location. A data collection system facilitates diagnostic/prognostic effectiveness trending and identifies target areas for improvement as a function of maturation during the deployment, operations and support phases. This paper describes how all these segments weave together to create a seamless closed loop diagnostic/prognostic process that fits into the new DoD 5000.2 instruction: integrated defense acquisition, technology and logistics life cycle management framework