Retail chains can now receive real-time reporting on self-checkout theft and other scan avoidance—as well as prevent false alerts at the self-checkout—with StopLift Checkout Vision Systems’ SmartAssist artificial intelligence (AI) technology.
SmartAssist AI prevents false alerts and interventions, immediately flags unscanned merchandise at the self-checkout, and alerts the attendant before the customer leaves the checkout, says StopLift. The system also can identify an item not meant for purchase but placed in the bagging area, such as a purse or reusable shopping bag, so that no alert will be sent and no beep will occur.
The technology works in the entire spectrum of self-checkouts and POS systems. NCR Corp., a global self-checkout technology company, has incorporated StopLift’s AI capabilities into its self-checkout solutions to address retailers’ concerns over shrink and to provide what it says is a frictionless checkout.
Working with retailers on four continents, StopLift says it already has detected and confirmed more than 3.2 million scan avoidance incidents at thousands of manned and self-checkouts in retail chains, and enabled retailers to improve cashier and self-checkout attendant training. StopLift’s patented video analytics technology visually determines what occurs during each transaction to immediately distinguish between legitimate and fraudulent behavior at the checkout.
“SmartAssist enhances weight-based security to handle a broader spectrum of cases,” said Malay Kundu, founder and CEO of StopLift. “Until now, security at the self-checkout has been performed solely by weight scales. Unfortunately, weight scales take time to settle, and legitimate purchases do not always match their expected weight in the database. As a result, conventional security measures often produce mistaken alerts which frustrate customers, leading some retailers to disable security and expose themselves to theft.”
StopLift’s video analytics enable self-checkouts to detect theft and unintentional scan avoidance that circumvent the weight sensor. For example, the system can detect merchandise left in the shopping cart or bagged outside of the bagging area without scanning. With real-time alerts, the attendant is notified right away and can take immediate action before the customer leaves the checkout.
By utilizing StopLift’s video analytics to “double check” potential alerts, SmartAssist can prevent unwarranted interventions. This protects the shopper from the annoyance of having their transaction interrupted, the embarrassment of having the self-checkout light flash, and the frustration of waiting for an attendant to come and clear it, the company says.
“This means no more customer complaints of, ‘The machine beeped at me,’” Kundu said.
“Fewer false-positive alerts make their way to the attendant, who can now act properly on legitimate alerts and cover a greater number of self-checkout stations,” says StopLift. “This means greater productivity gains and associated labor savings for the retailer.”