Condition based maintenance for complex distributed systems

Condition Based Maintenance (CBM) solutions are traditionally challenging to implement for today's complex distributed systems. By their very nature, these systems pose several technical obstacles. The systems to be monitored are distributed, often at remote locations, requiring a data collection network infrastructure that is secure, robust to intermittent connectivity and scalable. Heterogeneous “leading indicator” data is collected from various sources: discrete forms of data such as status, state, mode, system error reporting and inputs from other software systems; parametric data such as environmental sensors and system sensors; and manually collected data such as operator observables and maintenance actions performed. Disparate sources and forms of data also pose a challenge for analysis. Thus, in order to implement a CBM solution for complex distributed systems, it must be based on three core pillars: smart sensors capable of collecting heterogeneous data types, scalable and generically applicable predictive analysis methodologies, and a secure network infrastructure. Mikros is currently deploying a CBM+ system for combat systems on the U.S. Navy's Littoral Combat Ship (LCS). In this CBM+ system application, smart sensors are used to collect heterogeneous data from Navy combat systems. Data is collected in IEEE SIMICA standard format and transferred securely from LCS ships deployed around the world to a central server in the U.S. The Prognostics Framework®, a model-based prognostics reasoning engine, is used to analyze all data to produce prognostic alarms, identify maintenance action needs, report Remaining Useful Life (RUL) of key components, and provide a comprehensive health management capability for the LCS fleet. In summary, heterogeneous data collection made possible through smart sensor technology, model-based prognostics, and a secure network infrastructure provide a flexible and extensible framework to implement CBM for complex distributed systems. Without these core capabilities, CBM falls short of its goals to proactively support maintenance needs, increase system readiness and reliability and reduce overall life-cycle costs of today's complex distributed systems.