According to Market Watch, by 2020 the global food and grocery retail market size is expected to reach $12.24 trillion. With so much money at stake, big-box retailers are using the latest advancements in machine learning, computer vision and deep learning algorithms to reimagine every aspect of our neighborhood grocery store.
From startups to giants like Amazon, now’s the time for other AI companies to enter and disrupt the trillion-dollar supermarket industry and rethink how we buy food in the future. Let’s look at three ways AI is transforming the modern supermarket.
Inventory forecasting and food waste reduction
According to the United Nations, roughly one-third of the food produced in the world for human consumption gets lost or wasted every year.
The supply chain for fresh foods offers unique challenges that contribute to excessive food waste. Fresh food is perishable, easily damaged, difficult to track/quantify, and varies in quality. A staggering 45 percent of all fruit and vegetables are wasted. With a relatively short shelf life, much of the waste occurs at the end of the supply chain, during distribution and consumption.
Afresh Technologies uses its proprietary AI technology to help grocers forecast demand, replenish inventory and gain consumer insight through decision models. With its deep learning algorithms, Afresh processes millions of data points to provide demand forecasts at the item-level. The AI application recommends items needed to replenish inventory, and this procurement process is automatically integrated into a grocer’s workflow.
Unlike current supply chain technologies – which assume products are always the same – new AI algorithms are being designed to account for perishable products like fresh produce. Deep learning can then forecast demand at the end of the supply chain in real-time to reduce food waste.
Theft and shrinkage prevention
Retail crime such as internal theft, shoplifting and fraud is a serious problem facing grocery stores. While inventory shrinkage costs the U.S. retail industry over $45 billion each year, grocers allocate only 0.36 percent of sales to reducing shrinkage.
New AI solutions are helping loss prevention departments combat theft and fraud at the checkout counter. StopLift uses AI technology developed at MIT to detect various forms of “shrink” at the point of sale. By inserting vision algorithms to existing security cameras, the AI algorithm analyzes video footage to immediately distinguish legitimate and fraudulent behavior. This can range from operational errors, cart detection for unpaid items, cashier fraud and sweethearting – the unauthorized giving away of merchandise without charge. In all, any product at the point of sale that is not scanned at the checkout is flagged by the AI vision algorithm and considered a loss.
Piggly Wiggly, an American supermarket based in Southern and Midwestern regions of the U.S. reported losses between $6,000 to $10,000 monthly at one of its outlets due to checkout shrinkage. Since installing the StopLift AI, shrinkage has declined to $1,000.
For the everyday shopper, the biggest advancement of AI will be the convenience of cashier-free payments. Amazon Go is the first retailer to use computer vision, deep learning algorithms and sensor fusion to automate the retail transaction in a brick-and-mortar convenience store.
Customers can purchase products without being checked out by a cashier or a self-checkout station. Cameras on the ceiling and store shelves with built-in weight sensors detect which items customers took. When a customer takes items off the shelf, it’s placed in a virtual cart tied to a customer’s Amazon account. Once the customer leaves the store, the transaction is automatically processed.
Rather than outfitting entire stores with expensive cameras and smart shelves, new startups allow other smaller supermarkets to compete with Amazon Go. Caper developed an artificial intelligence powered shopping cart that leverages computer vision to avoid checkout lines.
Customers simply pick up items and drop them into the cart. Built-in sensors identify the item and calculate the amount on interactive screens fitted on the carts. It also recommends items based on the products inside the cart and showcases promotions and deals. By interacting with customers at the cart level, Caper suggests supermarkets can increase basket size by up to 18 percent.
From modern to future supermarkets
Despite the rise and popularity of online shopping, customers will continue to shop at local supermarkets to purchase food. Artificial intelligence is playing a pivotal role in shaping grocer’s supply chain to reduce food waste, prevent theft to minimize shrinkage, and reimagining the customer experience by eliminating checkouts and long lines. And with the development of 5G, it’s likely that the food supply industry will increase their use of technologies like connected sensors, computer vision and other monitoring devices to track inventory in real-time.