PREDICTRONICS CORP. — Department of Defense SBIR Phase II: N193-A01

Description

Advancements in artificial intelligence (AI) and machine learning are paving the way for more robust implementations of predictive solutions, especially for applications such as predictive maintenance. This technology can potentially impact the Navy’s operations and result in lower maintenance cost and higher fleet availability. In the Small Business Innovation Research (SBIR) Phase 1, the analytics methodology proposed and developed by the team provided promising results with three public data sets, as well as giving a glimpse of this potential technology through a software prototype of its user interface. In this SBIR Phase 2, one of the goals is to have a fully functioning software that will be evaluated on a Navy use case, such as aircraft engines or rotorcraft health and usage monitoring systems (HUMS). The other goal is to develop the ability to incorporate labeling from both a human-in-the-loop as well as other data sources, including maintenance or inspection records, would provide for a more automated way of labelling the new anomalous patterns. Such new capability will enable the solution to be more robust and result in the generation of less false alarms over time.