Digital-twin and AI-driven Expert Reasoning Engine for TPS

As the Army transitions towards a warfighting doctrine of Multi-Domain Operations (MDO) for large scale combats, it is essential for the tactical units to achieve a much higher accuracy and resolution of the health assessment of their equipment. Rapid determination of No Evidence of Failure (NEOF) at the tactical level  as well as rapid determination of failed components and consequent loss of functionality of the equipment at the tactical unit without the need for reach back to the depot is essential for success of the MDO doctrine. The increased ability towards the determination of equipment health and its functional assessment as well its maintenance and repair at the tactical unit is thus a key success factor for equipment availability and sustainability while incurring a reduced logistics footprint.  QSI seeks to design and develop a scalable methodology for smart, adaptive and purpose-built TPS generation using digital twin-driven diagnostic reasoning and use test descriptions, procedures and setup information from existing TPS to structure the test program itself.  The test execution sequence and strategy are derived from the digital twin and its associated AI-based causal reasoning engines such as QSI’s TEAMS tools, while the specific details of the test execution, its setup and execution related instruments and parameters are extracted from the TPS repository using AI-based techniques such as Retrieval Augmented Generation (RAG) and finely tuned Large Language Models (LLMs). TEAMS digital twin and causal reasoning engine will leverage the UUT’s health-related and diagnostic data that are generated onboard as the basis for an optimal insertion point for the TPS execution. The TEAMS software can readily generate the current health of the different assemblies and components of the system including the UUT.