A Predictive Analytics-based Resource Demand Estimation and a Software Framework for Agile Planning

QSI, leveraging its extensive experience in diagnostics and prognostics of complex NASA, military and commercial systems through its commercial software toolset TEAMS (Testability Engineering and Maintenance System) and research on power quality monitoring and mission planning, seeks to develop a data-driven, self-learning software framework, for context-based (i.e., mission tempo, resource, weather, location and task-based) assessment of critical quantities (e.g., fuel, ammunition, water, food, medical supplies, repair parts, etc.) over time. The software framework, once integrated with the TEAMS toolset, will monitor mission activity (planned or ongoing) data streams to estimate power demands of mission systems, as well as detect and classify mission events (using a trained model from TEAMS-DMin (Data miner) that is automatically embedded in TEAMS-DE (decision engine)), forecast context-sensitive resource demands to meet the COA needs, and provide alerts when the demands exceed the capabilities for altering schedules, selecting alternative COAs, or reducing demands from low priority loads. While the specific focus of demonstration will be on predicting the energy and power needs for systems in a FOB based on forecasts of weather and tempo of operation, the software architecture is generic, modular and extensible in that it can be used to predict all aspects of mission-resource requirements.