Data-Driven & Physics Informed Machine Learning for Advanced Diagnostics & Prognostics (PPMX)

PredictiveIQ intends to show how its Small Data Machine Learning (ML) technologies can significantly improve Prognostics/Predictive Maintenance (PPMx) predictions of remaining useful life (RUL) of components and systems used in Combat Vehicles. Small Data ML is powered by Data-Driven ML methods and Physics Informed Machine Learning (PIML) approaches. The PredictiveIQ PPMx solutions address three major Army PPMx problems: The Data Problem: PredictiveIQ's Small Data ML significantly reduces the amount of data required for PPMx systems to be truly Predictive (What is likely to happen?) and Prescriptive (Hoe to act in Respomse). PredictiveIQ's PIML frameworks also enable the creation of virtual sensors enabling measurements in locations where it would be difficult or impossible to place physical sensors. Decisions in Contested Logistics Environments: PredictiveIQ's Small Data ML enables on Edge and on Platform training, enabling the deployment of this technology in a distributed, no-contact, combat environment. Persistent Platform Modernization for Sustainment: PredictiveIQ's Small Data ML complies with Modular Open Systems Approach (MOSA) This modular feature offers allows to be mission-specifically tuned and continually improved, thereby creatingĀa degree of uncertainty on our adversaries, and therefore leading them to either not engage or to miscalculate.