See how the DoD is using AI for detecting maintenance on rotary wing aircraft to ground vehicles.
The Center for Government Contracting Director, Jerry McGinn, and Senior Fellows, Richard Beutel and Benjamin McMartin, examine the current state of and future opportunities for predictive maintenance as a case study for the adoption of Artificial Intelligence (AI) and machine learning (ML) as a military capability. Collaborating with the military departments, the Defense Innovation Unit and the Joint Artificial Intelligence Center launched a series of projects starting in 2017 exploring AI/ML predictive maintenance solutions on platforms ranging from the E-3 Sentry and F-16 aircraft to UH-60 and AH-64 rotary wing aircraft to ground vehicles.
The results of these initial projects were generally promising, but the complexity and heterogeneity of maintenance data and numerous other factors made it clear that the utilization of AI/ML techniques in this area would require continued iteration and user feedback. Promisingly, the services have continued their investments to develop better data sets, better algorithms, and better tools towards the predictive maintenance objective.
Based on the pilot efforts examined in this white paper, predictive maintenance is developing into a promising use case for AI technology in DoD. The authors conclude with a number of recommendations on data, metrics, prioritization, building trust across the maintenance enterprise, and other areas to help DoD remain on this promising trajectory.
