SAS, a major business intelligence and analytics software provider, announced that the U.S. Department of Agriculture is using SAS analytics to root out fraud among its $74 billion food stamp program.
More specifically, the USDA's Food and Nutrition Service (FNS) is using SAS predictive analytics to fight fraud in its Supplemental Nutrition Assistance Program (SNAP).
The SNAP program assists more than 46 million Americans. Despite the percentage of fraudsters being small, annual losses can still reach as high as $750 million by some estimates. Enter SAS with its analytics software, which helps spotlight suspicious behavior by benefits recipients.
"SNAP fraud rates are remarkably low—barely over one percent," said Karen Terrell, vice president of SAS Federal, in a statement. "Use of this cutting-edge technology demonstrates USDA's zero-tolerance policy when it comes to misuse of these vital benefits that help needy Americans put food on the table."
FNS is using SAS software to detect suspicious behaviors that indicate a beneficiary may be illegally selling SNAP benefits for cash—a scam known as SNAP trafficking. State agencies can then flag those individuals for further investigation.
Predictive analytics goes beyond mere statistics and reports on past activity to provide the best assessment of what is likely to happen in the future. For example, the system might analyze a series of claims for lost electronic benefit transfer (EBT) cards tied to the same address or neighborhood. Investigators would then be alerted to direct their attention to clients who may be selling their benefits at an area store and can focus on whether these beneficiaries are illegally redeeming their EBT cards for cash.
Moreover, FNS is exploring use of SAS predictive analytics as part of a business-process re-engineering test under way in several states. FNS will develop an assessment of the SNAP activity in each state and apply the SAS predictive modeling technology to analyze recipient behavior to uncover suspicious activity, SAS officials said.
This process combines data, statistical algorithms and machine-learning techniques to predict the likelihood of future outcomes, such as fraud, based on historical data. This enables states to quickly identify and take enforcement action against those suspected of SNAP trafficking.
"Over the years, FNS has developed best practices for using technology and analytics to combat fraud, waste and abuse in its program," said Terrell. "With this project, FNS is taking a strong leadership role, using what they've learned to help the states go after bad actors. This initiative is a model for the states and other federal benefit programs to root out fraud and improve program integrity."
SAS has provided analytics solutions for the federal government for nearly four decades. Its software is used throughout the federal workplace, including civilian agencies, the Department of Defense and the intelligence community.
Of the company's approximate $3 billion in 2014 revenue, 15 percent came from government. Government is SAS' second largest industry segment, with banking the largest with 26 percent of SAS sales.