Your Browser is Out of Date

ShareDemos uses technology that works best in other browsers.
For a full experience use one of the browsers below
Dell EMC Ready Solutions for AI: Retail Loss Prevention

view all chapters

Executive summary
Business challenge Audience We value your feedback

Business challenge

  • Recent data from the retail sector shows that 2019 worldwide annual revenues exceeded $25 trillion dollars (https://www.statista.com/statistics/443522/global-retail-sales/). In the United States alone, the retail sector employs over 15 million workers in more than 1 million retail establishments. The 2012 US Census found that retail establishments vary in size and activity from $10,000 in annual sales to $25,000,000 and more. Despite the size differences and diversity of the retail sector, all retail businesses share the impact from lost revenue that occurs when merchandise goes missing. Retail organizations have long recognized the financial impact of inventory that is deemed missing or lost. A 2017 Money magazine article estimated that inventory loss from stores cost the U.S. retail industry nearly $48.9 billion for the previous year (https://money.com/shoplifting-fraud-retail-survey/). Most of the financial impact of inventory loss in the retail sector occurs in stores at the point of sale (POS). 

    Preventing scan fraud activities such as scanning errors and UPC barcode switching at the POS is critical to reducing this major source of inventory loss. Dell Technologies in partnership with Malong Technologies and NVIDIA offer a solution that uses artificial intelligence (AI) to reduce inventory loss that is caused by either accidental or intentional behavior during checkout. The solution is powered by an in-store Dell EMC PowerEdge server that runs state-of-the-art product recognition technology from Malong RetailAI Protect. The solution can be used in most retail stores to reduce inventory loss during POS use at both self-checkout (SCO) and employee-staffed checkout lanes. Cameras that are positioned around an SCO kiosk capture real-time video of product items as they are scanned. The products are cross checked with purchases that are recorded by the POS system. NVIDIA T4 GPUs, part of the NVIDIA Tesla product line, perform video decoding and product recognition. The GPUs provide the throughput and fast response times that a retail checkout system requires. PowerEdge servers have a well-integrated management suite that makes them ideal as edge nodes in distributed retail environments. The Malong software offering is designed for distributed deployment, monitoring, and management through Microsoft Azure IoT Hub and the NVIDIA EGX stack.