JAIC is Using Data to Predict Supply Chain Shortages Caused by COVID-19

JAIC is Using Data to Predict Supply Chain Shortages Caused by COVID-19

The project exemplifies the direction the center is taking to make AI operational across the Defense Department.

The Defense Department’s Joint Artificial Intelligence Center has a supply chain data integration project underway to predictively identify supply shortages and complications caused by COVID-19 and to better inform how the military can best respond to the pandemic.

The JAIC launched the project, called Project Salus, in March to help the Northern Command and National Guard find where critical supplies were limited by leveraging vendor data and government datasets and turning them into information that DOD could operationalize in responding to crises like the COVID-19 pandemic.

As the JAIC looks to operationalize that project toward fighting COVID-19, it's undergoing its second series of sprints to transition Project Salus to the national response.

“Our latest efforts in the response to the COVID-19 situation has been Project Salus, where we developed a data aggregation platform in multiple predictive models primarily focused on providing intelligent humanitarian assistance decision support,” JAIC Chief AI Solution Architect and Acting Chief for Data Science and Engineering Nathaniel Bastian said of the project at an AFCEA and George Mason University online symposium last Wednesday.

Project Salus made significant progress over the past two months in redirecting its work toward COVID-19 support.

“We got something fielded, working some of the best companies in the United States with some of the best AI, machine-learning talent in the world that have been part of those companies, and we did it as an integrated product team,” JAIC Director Lt. Gen. Jack Shanahan said at an AFCEA event last Thursday.

Bastian explained that key members of the integrated product team consists of a mission, project and product managers, a data science and AI engineer, infrastructure engineer, cloud architect, test and evaluation engineer and a policy and acquisition subject matter expert.  

The team has been able to so far offer initial AI and machine-learning prototypes to responders to provide feedback, and the project is also undergoing review to ensure it complies with DOD policy.

Project Salus also exemplifies the longer-term strategy that the JAIC has taken in moving DOD AI principles and expertise to the frontlines of DOD’s mission and in making JAIC the DOD’s AI center of excellence, Bastian added.

“We are striving to be this … center of excellence that’s connecting mission system users, our warfighters, with our customers, which are our program offices, to build AI-enabled capabilities that are human-centered and that can be trusted and used and adopted,” Bastian said.

Standard