The Defense Innovation Unit (DIU) and the Joint Artificial Intelligence Center (JAIC) conducted two separate pilots to test the capability of two machine learning platforms to identify and address unmatched transactions and other accounting errors in the Department of Defense’s financial management enterprise.
Summit2Sea led the first pilot, which used the Sagemaker ML platform from Amazon Web Services and financial data from six of 24 defense agencies backed by the DOD’s comptroller’s office, DIU said Thursday.
Vertosoft and DataRobot collaborated with the Army Financial Management and Control team to test their artificial intelligence platform during the second pilot.
The first pilot enabled each of the six DOD agencies to build a separate model when it comes to resolving UMTs using six different use cases. The agencies were able to reduce the time of finding and addressing these erroneous transactions from two hours to two minutes per UMT on average.
DataRobot and Vertosoft evaluated their AI platform across four use cases and found that the technology achieved 99 percent accuracy when it comes to detecting errors in transactions. The two ML platforms ran on a government cloud environment, called ADVANA.
“The plan is to add more use cases over time which will improve the UMT coverage,” said Eric Dorsey, a contractor and DIU program manager for the effort.
“With the large number of UMT generated in the Army FM&C, we expect them to save millions of dollars of labor every year. Our hope is that other DoD entities such as the Navy, Air Force, Marines and components will be able to leverage this project in the future,” added Dorsey.
Summit2Sea and the team of Vertosoft and DataRobot have moved their technologies to the production phase.