Wednesday, April 29, 2020

Meet 'Salus' - The Pentagon's New COVID-Hotspot, Panic-Buying Predicting AI

A new predictive technology developed by the Pentagon can anticipate product shortages could help military personnel move supplies to retailers or hospitals with geographical precision before a shortage develops during a public health crisis. 
The Joint Artificial Intelligence Center (JAIC) developed the new prototype artificial intelligence tool that predicts and addresses shortages of food, water, medicine, and other essential goods in a geographical region during crisis times. "You have to be looking a little in the future," said Nand Mulchandani, chief technical officer at the JAIC, who recently spoke with Defense One
Called Salus, the new tool combines data from the Census Bureau, Medicare, hospitals, and can forecast community spreading of a virus in a geographical region, as well as pulling data from retailers and warehouses of product inventories, and develop a predictive model of where shortages could potentially be seen, right down to specific zip codes or even stores. 
The military is working toward a "common view and a predictive capability to truly understand where the next problem sets are going to be and bringing to bear all of the logistical capability," said Mulchandani, who spoke with Roll Call in early April. 
The new tool has already been field test with communication systems of Northern Command and National Guard, which are both supporting FEMA's efforts in combating the coronavirus spread. 
Mulchandani told Defense One that the AI tool is flexible in determining different types of problems that could occur by altering the data that is piped in. 
He said Salus was first tested by determining and developing a model of where shortages of ventilators, masks, and other medical supplies could be seen. 
"The next question really was resource allocation like food," he said. (Read More)