Machine learning can help supermarkets better manage the costly problem of having too much or too little fresh food in stock, argues a new report from McKinsey & Company.
The technology could provide grocers a competitive advantage in the battle between independent supermarkets, incumbents, new international entrants and grocery delivery start-ups.
The authors note most traditional supply-chain planning systems take a fixed, rule-based approach to forecasting and replenishment.
“Retailers are constantly having to make difficult trade-offs when placing orders with fresh-food suppliers: order too much, and the food goes to waste; order too little, and you lose sales and erode customer loyalty. But with demand fluctuating daily, how can retailers know the right amount to order?”
Machine learning in supply chain planning can be used to automate manual processes and improve accuracy, while learning over time.
The authors comment, “Retailers that use machine learning technology for replenishment have seen its impact in many ways — for instance, reductions of up to 80 per cent in out-of-stock rates, declines of more than 10 per cent in write-offs and days of inventory on hand, and gross-margin increases of up to 9 per cent.”
A machine learning algorithm can make demand forecasts based not just on historical sales data but other influential factors such as advertising campaigns, store-opening times, local weather and public holidays.
“Advanced algorithms currently used by leading retailers already analyse more than 50 parameters. And the calculations are done at a much more granular level than standard systems are able to do: retailers can determine the effect of each parameter on each SKU in each store (and in each distribution center, where relevant) on a daily basis,” the report says.
For example, in the graph below the verticals bar show that stocking four pineapples in the store on a particular day means the store is likely to sell them all, reducing the risk of wasting food. However if more than four people wanted to buy a pineapple the store would miss out of revenue. The green curve on the graph shows the expected value of costs for each stock level, taking into account potential loss of revenue due to out-of-stocks, as well as potential markdowns and waste. In this case, the algorithm identifies a stock level of nine units as optimal.
The system can align individual ordering decisions with the retailer’s strategic goals and KPIs. Eg, if the retailer is more concerned about margins than revenues, the algorithm will adjust decisions accordingly.
Getting suppliers to accept orders based on a new forecasting system will be a new hurdle in supermarket-supplier negotiations.