How AI-powered supermarkets reduce friction, food waste

AI is being used across retail for many different applications, starting from checkout applications and continuing with supply chain efficiencies, order forecast and optimization, and store assortment management. Shoppers are usually comfortable with whatever makes life easier and their store experience better, regardless of whether or not they understand the AI technology behind it. 

Such is the case with frictionless, walk-out checkout solutions that use AI. For example, Trigo is implementing a retail store operating system called StoreOS that combines AI with smart sensors to make smarter brick-and-mortar stores. 

“As more grocery and food retailers merge their digital and physical offerings, StoreOS runs applications that allow retailers to do things like track inventory in real time, minimize out-of-stock items for in-store and online shopping, optimize the layout of stores and limit shoplifting,” Shay Ziv, Trigo’s vice president of marketing, told Retail Leader Pro. “The system gathers anonymized data from ceiling cameras and shelf sensors to generate real-time automatic analysis of interactions between people and products in a store.”

The first application to run on Trigo’s StoreOS is EasyOut, which turns traditional grocery stores into frictionless supermarkets where shoppers can walk in, select their items and walk out without having to queue at checkout or scan any items. Payments and receipts are settled digitally. EasyOut then feeds into EasyStock, which provides real-time visibility of store inventory status, improving stock-replenishment processes and providing smart out-of-stock alerts that lower customer dissatisfaction and increase efficiency and sales, Ziv explained. 

This year, Wakefern Food Corp. launched an autonomous convenience store powered by Trigo at the retailer’s corporate headquarters. Additionally, German grocer REWE opened the country’s largest hybrid-autonomous supermarket, where customers have the option to shop and pay for their purchases in a traditional way or autonomously without a checkout process. Last December, Trigo and REWE opened Germany’s first fully automated supermarket.

Opportunities for AI

AI with smart sensors can minimize waste by determining when perishable and other expired products need to be discarded, reduce shrink from shoplifting and decrease out of stocks.

“Computer vision AI offers retailers the opportunity to increase their revenue and provide a better customer experience on many different levels, whether it is through improving efficiency, product availability or better marketing tools,” Ziv said. 

Using computer vision AI, retailers also can customize offerings, both for a single customer and for the whole store. 

“For the customer, using AI, the retailer can offer personalized promotions, in-store or online,” Ziv said. “At the store level, using AI the retailer can customize the store assortment based on the in-store shopper behavior. More generally, AI is used in many applications that might make the shopping journey quicker and better, such as frictionless checkout or reduction of out of stock.”

As far as store management, using computer vision AI, retailers can get a real-time picture of on shelf-inventory, understanding which items are running short and need to be replenished, reducing out-of-stocks, and improving customer satisfaction, as well as improving their operations and efficiency. 

“This technology can also help improve the ordering forecast, using in-store customer behavior, leading to a higher availability not only in the store but also in the warehouse,” Ziv said. 

For example, computer vision AI systems can alert retailers about items that are misplaced, whether it is a single item, or a whole shelf that is not compliant with the store planogram, she adds.

Additionally, using computer vision AI and machine learning, retailers can improve their carbon footprint by optimizing store hours, monitoring stock levels to avoid overstock and understock and tracking expiration dates — significantly reducing waste. In regard to reducing food waste, AI can predict which products are likely to go to waste and proactively donate or repurpose them, Ziv said.

“AI can analyze data from suppliers to identify the most sustainable options, reducing the carbon footprint of the supply chain,” she added. 

Ziv said the future of AI improvement will reside with: 

• Improved customer experience: AI can provide personalized recommendations and offer real-time support, improving the customer experience and driving sales.
• Increased efficiency: AI can automate routine tasks and provide insights into operations, reducing costs and increasing efficiency.
• Better inventory management: AI can predict demand, manage stock levels and optimize the supply chain, reducing waste and improving overall efficiency. 
• Improved data analysis: AI can process vast amounts of data, providing insights into customer behavior, market trends, and supply chain performance, enabling retailers to make informed decisions. 
• Personalized marketing: AI can analyze customer data to provide personalized marketing and advertising, increasing customer engagement and driving sales. 

“Retailers using AI will be able to manage their store more efficiently, and improve their offering to customers, leading to better financial results and more returning customers,” Ziv said.

What’s next: As AI grows across retail, two issues arise: consumer data privacy and consumer understanding of AI usage. As evidenced by the growth and trendiness of ChatGPT, consumers are intrigued by AI, but they aren’t necessarily aware of the potential consequences and misinformation that comes with the emerging tech. ChatGPT’s success brings AI into daily conversation and activities for consumers, which lowers the barrier of entry for retailers.