The fastest growing of Britain’s big four supermarkets, Morrisons, on Thursday reported another rise in quarterly sales, and Germany based machine learning company Blue Yonder is claiming some of the credit.

Commenting on the way its software reduced overstocking and understocking at the supermarket’s stores, Blue Yonder’s founder Professor Michael Feindt said:

“Only those retailers who understand the importance of advanced machine learning algorithms and Big Data will survive and thrive.”

Blue Yonder recently won an Institute of Grocery Distribution award for its innovative work with Morrisons.

The supermarket brought in Blue Yonder after it decided it needed a new system to replenish its stores. Previously it was relying on manual orders and experiencing problems with product availability.

Morrisons was looking for a capital light solution to integrate with existing infrastructure while facilitating removal of legacy systems, and providing a high level of automation to reduce manual intervention.

Crucially, the company says, the system needed to offer highly accurate customer demand predictions at an individual store/product/day basis to increase product availability and reduce lost sales, using self-learning technology to ‘leapfrog’ Morrisons against the rest of the market.

Morrisons immediately identified Blue Yonder as an industry leader in AI and Machine Learning innovation, and the Blue Yonder Replenishment Optimisation solution as the ideal technology.

“The system is capital light, utilising cloud technology and store specific historic sales data to forecast stock requirements. It is reducing costs and stock levels, while also saving time for colleagues, and providing a better offer for customers,” David Potts, CEO at Morrisons said.

In just 12 months, Blue Yonder optimised ambient and long-life replenishment among 26,000 stock keeping units (SKUs) across 130 categories in all of Morrisons’ 491 stores.

Morrisons and Blue Yonder agree the dramatic improvement in replenishment capabilities is due to the Blue Yonder’s solution’s use of cutting-edge AI technology. Not only can AI automatically forecast orders daily per store and per SKU to meet customer demand, its self-adjusting capabilities mean demand can be optimised down to product level – factoring in influences including store demand patterns, seasonality, weather, and promotions – they say.

In this way, they say, the solution can model what is going to happen, not just what has happened, and because the solution is cloud-based, it is scalable. For Morrisons, this means the grocer can respond quickly to changes in customer demand, and in the wider market.

The Blue Yonder technology has already improved shelf gaps by up to 30% and reduced stockholding in store by two to three days, Morrisons says.

Morrisons will soon be introducing Blue Yonder Replenishment Optimisation across all product groups, except fashion.

Prof. Dr. Michael Feindt, particle physicist, ex-CERN scientist and founder of Blue Yonder said: “Along with stock machine learning benefitting retailers, businesses can also use AI to automate tasks that may seem laborious to staff, improving their standing in the company and job satisfaction, as well as giving them more time for their main tasks.”

He continued: “Retail is dealing with many industry changes: the age of millennials has shifted the rules of shopping, requiring options of 24/7 shopping on any platform, delivered when they want it. Combine this with digital transparency of pricing, and the ease of comparing products online, and the retail sphere is a very different place with margins getting ever tighter. This makes it vitally important that retailers are able to utilise artificial intelligence solutions and deliver even better customer experiences, with the most effective replenishment and pricing strategies, while also making it mission-critical to reduce waste.”