REDSTONE ARSENAL, Alabama – The U.S. Army Aviation and Missile Command G-3 partnered with the Joint Artificial Intelligence Center (JAIC) in July 2018, when the JAIC was initially formed.
The partnership was designed to inject machine learning and artificial intelligence into some of the G-3 processes and was a way to help the JAIC understand some of the challenges that Army aviation was facing related to sustainment and readiness.
“This partnership began in January of 2019 when the JAIC started a predictive maintenance (now known as Joint Logistics) mission initiative and the UH-60 was chosen as the platform for evaluation,” said James Snyder, AMCOM G-3 Sustainment Support Branch chief. “We continue to partner with the JAIC from that time forward working on predictive maintenance solutions and focusing specifically on teardown analysis of UH-60 components at Corpus Christi Army Depot [in Texas].”
The JAIC is focused on broad enablement of artificial intelligence capabilities by providing infrastructure, developing pathfinder capabilities and offering technical expertise for Department of Defense users to pursue their use cases.
“We have several key engagement areas, known as mission initiatives including joint logistics, joint warfighting, business process transformation, warfighter health, joint information warfare and threat reduction and protection,” said Chris Shumeyko, JAIC product manager. “We are also providing a cloud-based environment that will provide infrastructure for developers in the field to pursue their own AI capabilities.”
According to Snyder, this partnership allows the sharing of resources between the two commands.
“Utilizing the JAIC expertise in AI/ML with the AMCOM sustainment expertise allows for the analysis of current data sets geared toward sustainment improvements, “said Snyder. “More importantly, it sets the stage for development of future capabilities that will allow AI to identify faults, provide trend analysis and offer solutions to sustainment issues proactively.”
Snyder stated that injecting AI/ML into the process means using proven methods that continue to learn and understand requirements based on received data.
“It allows for an automated approach to solving problems quicker and with greater accuracy than manual methods,” said Snyder. “The ability for a program to learn from the data it receives and adapt to current operational requirements is key to successfully solving sustainment challenges.”
Recently, the JAIC funded the development of an AI/ML tool known as AutoFire for G3, which was developed by the Algorithmic Warfare Provisional Program Activity Office.
“The basis of the program is to conduct root cause analysis of depot-level engine teardowns to provide the primary reason for failure of the component,” said Snyder. “The implementation of Autofire will reduce the need for the manual board process and provide results quicker. As teardown analysis is transitioned into the Maintenance Consolidated Database, the need for an automated process becomes even more critical to provide actionable information in a timely manner.”
“When it comes to logistics and maintenance, there is an overwhelming amount of data available — anything from aircraft sensor data to maintenance forms and part records,” said Shumeyko. “Ordinarily, subject matter experts play a huge role in understanding this data and identifying trends that may affect the readiness of the Army’s vehicle fleet. However, as the amount of data grows, you either need more experts to comb through that data or possible warning signs of problems may get missed. By injecting AI/ML, we’re not replacing these experts, but rather providing them with tools that can find hard-to-spot trends, anomalies or warning signs in a fraction of the time. Our goal is to increase the efficiency of the experts.”
At CCAD, the PMx initative is specifically focused on teardown analysis of UH-60 components.
“H-60s are powered by two T700 turbojet engines,” said Shumeyko. “When these engines fail in the field, maintainers will sometimes send those engines back to CCAD for further inspection or repair. This is an opportunity to identify the root cause of failure. For example, a maintainer in the field may report that an engine was producing low-torque indications. At the depot, that engine may get torn down, isolating each component for inspection, and a report generated. Then subject matter experts would analyze that data and determine why the engine experienced low torque — foreign object damage, perhaps.”
This partnership is important as it aligns with the Department of Defense’s emphasis on AI/ML implementation. AMCOM’s partnership with JAIC places G-3 in a position to lead and steer efforts.
“AMCOM’s lead position with these efforts is critical to the aviation enterprise benefiting from these processes early in the DOD and DA planning cycle,” said Snyder.