Follow Datanami:
January 21, 2020

DoD Looks to Scale Predictive Maintenance

illustrate dogfight for DARPA AI program

Source: Lockheed Martin

In its efforts to make greater use of commercial technologies, a Pentagon innovation office formed to streamline government contracting has expanded an predictive maintenance effort designed to keep front-line aircraft ready for duty.

The Defense Innovation Unit (DIU) created in 2015 to link the military with technology vendors recently awarded a five-year, $95 million contract to to boost aircraft readiness. The company said it will provide an AI-based software application that uses machine learning algorithms to monitor aircraft systems. The goal is to spot critical subsystem failures before they occur and help predict the parts and maintenance required to keep aircraft flying.

The AI platform also would serve as a logistics tool, identifying the type of part required to fix an airborne system and where that part can be acquired from DoD’s far-flung logistics network. The ability to anticipate parts failures is seen as a way of saving time and money associated with unscheduled maintenance.

With front-line fighter aircraft like the F-35A costing about $80 million a copy, operations and maintenance spending continues to soar. As a result, the Pentagon comptroller recently estimated annual O&M costs at $292 billion. “Given these numbers, even a fractional increase in aircraft mission capability can save billions,” Ed Abbo‚’s president and CTO, said in announcing the DIU contract.

The company’s aircraft maintenance application runs on its AI suite designed to allow users to deploy and scale applications on public or private clouds. The flight readiness effort would allow maintenance crews to combine large volumes of siloed data. Those data sets could then be used by machine learning algorithms to improve maintenance operations.

While the system will initially be used by DoD for predictive maintenance, said its platform could also be used for applications ranging from data fusion to improved logistics.

The Air Force has used the company’s AI platform to boost readiness of the F-35 fighter along with E-3 Sentry (AWACS) surveillance aircraft. The DIU contract would expand the predictive maintenance effort to all three military services.

DIU selected in November 2017 to demonstrate its AI and Internet of Things platform to provide a predictive maintenance for the Air Force’s AWACS and F-16 aircraft. The DoD-wide contract announced last month represents an effort to scale AI technologies across the U.S. military.

That initiative parallels expanding AI initiatives in the civil aviation sector. For example, Delta Airlines recently announced an expanded predictive maintenance application that will incorporate machine learning and modeling. The carrier claims its the Skywise predictive maintenance application helped reduce maintenance-related cancellations to just 55 in 2018.

Based in Silicon Valley, DIU’s goal is to prototype and field new military capabilities based on commercial technologies within 36 months.

Recent items:

Predictive Maintenance Drives Big Gains in Real World

Big Data Predictions: What 2020 Will Bring

The Trade War, Supply Chain Risk, and AI