<noscript id="yq6u8"></noscript>
<samp id="yq6u8"><xmp id="yq6u8">
<center id="yq6u8"><wbr id="yq6u8"></wbr></center>
<optgroup id="yq6u8"></optgroup>

Ahold Delhaize CEO warns for 'harmful' algorithms

Frans Muller (Ahold Delhaize) warns for biased algorithms
Photo: Ahold Delhaize

Artificial intelligence is becoming increasingly important in the supermarket industry, but biased algorithms can cause great damage. Ahold Delhaize wants to pay more attention to this, CEO Frans Muller says.


Extensive automation

Algorithms can be biased by their design or because of the data with which they are 'trained', Muller explains. This can happen when the developing teams are not composed in a balanced way: that is why Ahold Delhaize is now paying more attention to ensuring that these teams have the right staffing. In terms of diversity, gender, background and world orientation, they need to form a representative picture of the retailer's diverse customer base.


With 55?million customers a week in the United States, Europe and Asia, bias should be eliminated as much as possible, the CEO told participants at the Erasmus Centre of Data Analytics (ECDA) annual congress, on which ICT website Computable reports.


Rich data saved Ahold

The customer journey is becoming more data-driven, and artificial intelligence is playing an increasingly important role for the supermarket. "But the algorithms could not prevent certain shelves from becoming empty at the start of the coronavirus crisis. The pandemic led to an increase of no less than five hundred percent in online orders in the Netherlands. One in three customers started using the app. The number of transactions both in stores and online increased sharply. Without 'rich data' and good algorithms, Ahold Delhaize would never have been able to handle this so well", Muller demonstrated.


The CEO expects a lot from far-reaching automation in the supply chain, in which algorithms ensure the replenishment of stocks. The retailer has learned a lot from the corona crisis and now claims to be better prepared for a next pandemic. "But at the same time, we still have a number of things to learn."