A technology start-up is seeking five bakeries to take part in a pilot scheme using artificial intelligence (AI) to reduce surplus.

CatsAI uses 10,000 data points – including roadworks, school holidays and the weather – to predict how much of each product a bakery should produce on any given day. Bakers need to give CatsAI access to EPoS data and the “brain” will do the rest.

The results are delivered via email and bakers are provided a range for each item – 10 to 15 white tin loaves for example – meaning they can make an informed decision based on batch sizes, ingredients and so on. The system claims to offer a sales uplift of up to 20% and a reduction in waste of up to 80%.

While technology such as this is becoming increasingly common in larger operations including supermarket in-store bakeries, it has yet to be widely adopted by smaller businesses.

“There’s a complete gap in the market for small and medium-sized enterprises in getting access to really affordable technology that has been the preserve of the big corporates,” CatsAI founder Paul Staples told British Baker. He started the company alongside Stephen Kinns, Stuart Ware-Lane and Tim Cuddeford, who all have a background in financial services and the banking sector.

So far CatsAI has trialled the technology with bakeries in Windsor and Essex, with the results described by Staples as an “immediate success”.

“We identified just under a 20% increase in sales that they could achieve through stocking correctly, and there was a 72% reduction in the waste of production by correlating baking cycles with what the consumers are buying, where, when and how,” Staples said of the Windsor trial.

It is now seeking five more bakeries for further trials, with a nationwide rollout planned for 2019.

Bakeries involved in the trial will be charged a “nominal fee” to cover costs.