Their model uses a variety of data, according to Sam’s Club officials. Things like local temperatures (warmer weather often means buying fewer pancakes); whether the Sunday football game is home or away (home games can mean more pies are required); How popular are pecan pies this year (more pecan pies can translate to fewer pumpkin pie sales).
These and other data points are linked to an AI model they’ve built. He makes recommendations to each store manager, such as how many pies should be on hand in their stores per hour. Last year Sam’s Club sold enough pumpkin pies to fill 450 football fields, officials said. (They refused to give an exact number).
Precisely forecasting demand, the officials added, is necessary, because competition to keep customers is intense, and profit margins narrow.
“If members don’t get what they need, they won’t renew with us,” said Pete Rowe, vice president of technology for Sam’s Club and member of the store whose family buys both pumpkin pie and pecans for Thanksgiving this year. “It is important for us and our model to make sure.”
Pumpkin pie or pecan pie? With these recipes, you don’t have to choose.
In recent years, sophisticated AI models have become commonplace in grocery stores. Driven by the pandemic and supply chain challenges, it’s rapidly changing the grocery-buying experience: from AI-powered shopping carts that automatically recognize items you’ve picked up to chef bots that generate recipes based on your purchases.
The rise is due to a combination of factors, according to grocery experts. Stores now have access to huge amounts of data, including from third-party brokers and shopper loyalty programs. Computer processing power is cheaper and faster. Machine learning models, the programs that computers use to learn and adapt on their own, have evolved. The epidemic has played a big role.
In pre-pandemic times, stores used software to help manage inventory, staffing and predict when merchandise would be available, said Gary Hawkins, chief executive of the Center for Retail and Technology. After the pandemic hit, “supply chains exploded, demand exceeded,” Hawkins said, and grocery stores were unprepared and needed smarter systems.
He added, “It literally blew all the models, simply because they weren’t advanced enough.” Very quickly, especially the big guys said, ‘We need something better here. “”
In April of 2019, Walmart launched an intelligence research lab where cameras and sensors are attached to algorithms to monitor how well stocked shelves are. In March, Kroger launched an artificial intelligence lab where the technology can track the freshness of vegetables. Ketchup maker Kraft Heinz is now using machine learning to track demand for its products leading up to events like the Super Bowl. Amazon opened a fully automated Whole Foods this year that uses deep learning software to let customers shop and checkout without the need for a cashier. (Amazon founder Jeff Bezos owns The Washington Post.)
Startups have also proliferated. New York-based Caper Cart builds AI-powered shopping carts that automatically recognize what customers pick up and check out. Shelf Drive in Seattle tells stores how many items they need per day. Hivery, based in Australia, has a template to advise grocers about where to put produce on shelves.
“AI is making its way into almost every technology-related capability,” Hawkins said.
This World Cup is taken from and fed by artificial intelligence
Dominic D’Agostino, a 30-year-old member of Sam’s Club in Dayton, Ohio, said he had no idea the company used such sophisticated technology to forecast demand for pumpkin pie.
Though he’s not a fan of the dish, and likely wouldn’t bring any of it to his sister’s house for the holidays—”the only pie I really like is pizza,” he said—”D’Agostino is intrigued, and somewhat concerned, that AI is being used in this way.” .
“It’s scary,” he said in an interview. “She’s also great.”
Rowe said that the Sam’s Club made the decision to use AI shortly before the pandemic. The series used software to guide its operations, but felt it could do better.
In years past, for example, Rowe said “we’d make a lot of pumpkin pies and a lot of croissants and this [would lead] For our partners are wasting their time and we also have to get rid of the stock.
Now, the company is using machine learning to predict inventory for everything it makes in-house, like pies and roast chicken. They also have “autonomous floor scrubbers” — or self-driving robots — that scan shelves and send alerts to employees to decide which items to restock first when delivery trucks arrive.
Rowe said it helped the store become more than 90 percent accurate in predicting demand, and he wants it to be even higher.
Google engineer who believes the company’s artificial intelligence has been achieved
Although AI is attractive, it also has risks. Researchers from the University of Arkansas said algorithms are running out of pools of customer data, which raises privacy risks. It can also lead to bias.
“Even if race or gender is not an official entry into an AI algorithm,” they wrote, “the AI application may compute race/gender from other data and use this to ‘price up’ specific demographics.”
Others note that AI is not a universal solution, and stores may waste money buying fancy software just to keep up with the hype.
“You can’t be overly fond of the shiny object element of AI,” Mike Hanrahan, former CEO of Walmart’s Intelligence Research Lab, said in a tech post. “There are a lot of shiny things out there that do things that we think are broadly unrealistic and maybe, in the long run, aren’t beneficial to the consumer.”