Automated Test Equipment Data Analytics in a PBL Environment

Abstract

The Performance Based Logistics (PBL) approach to platform sustainment is greatly enhanced when informed by high quality information about the current state of critical fleet assets, and a reliable estimate of anticipated future needs. Aircraft platform sustainment stakeholders have long used information from data analytic software to inform PBL teams and make more efficient and cost optimized decisions on operations, maintenance, and supply chain actions. These analytics consume platform operational data sources, and fleet asset parametric data to provide a wide range of information including fault diagnostics, failure prognostics, and part order demand forecasts. The branches of the U.S. Armed Forces operate and maintain a “fleet” of Automated Test Equipment (ATE) used to evaluate and diagnose critical Line Replaceable Units (LRUs) removed across a wide range of vehicle platforms. These critical test platforms generate comprehensive log files for test procedures including: self-diagnostic tests, calibration, and LRU Unit Under Test (UUT) evaluations. Generally speaking, this data is not finding its way back to a central repository where it can be analyzed, processed by automated analytic processes, and used for platform analysis and decision making. This paper describes a design for Automated Test Equipment test log analytics to provide enhanced information to test platform Performance Based Logistics. Examples are provided to show how results for a fleet of test instruments can be aggregated into central repository appropriate for human and machine learning processes. When a representative dataset is compiled, models can be trained achieve analytics goals of increasing sophistication from simple anomaly detection, through fault isolation diagnostics, to projections of future maintenance and supply chain needs. Also covered is how these test log based analytics can be combined with information extracted from other operational data sources including UUT test findings, test station maintenance logs, and part orders to provide additional test platform benefits. Finally, the implementation of test log analytics has potential benefits to the UUT platforms as well. This includes providing a path to accelerated component diagnostics through smart Test Program Sets (TPSs) that self-optimize based upon an understanding of historic test results.