Intelligent Maintenance Support Toolset for Reliability Improvement

Scientific Monitoring Inc. (SMI), in partnership with GE Aviation (GE), proposes to develop a set of maintenance support tools for automated analysis of reliability data. These tools can extract data from a military data warehouse (MDW), analyze the data, update models, flag failure trends, and recommend proper actions to maintainers and designers/ engineers. This toolset will use existing measurements of engine and C&A components (or LRUs, discover (or mine) cases, relationships, and modeling evidences from the performance, the usage, and the maintenance data of the LRUs, focus on ground-based, non-real time applications and be applicable to all types of accessory components, and readily applicable to flight control components. The objective of the proposed Phase II project is to demonstrate the ability of generating more timely and accurate maintenance recommendations through the use of the analytical tools designed in Phase I. The three analytical tools to be demonstrated are: the reliability anomaly detector (RAD), the competing risks tool (CRT), and the probabilistic learning tool (PLT). The demonstration will use a combination of actual and simulated engine and maintenance data in a relevant operating environment. Actual data will be drawn from the F414 MDW, and simulated data will be created to form a certain desired test cases. The relevant operating environment will be similar to the execution and user environment of the Navy/GE maintenance program. The goal of the Phase II project is to mature the technology/tools to TRL 5/6 so that one or all of the tools can readily be fielded in the Navy/GE maintenance and reliability analysis process to benefit both legacy and propulsion systems.