It’s an innovation that saves money—and the emissions that cause climate change. How an artificial intelligence food-waste product transforms kitchens
In a bustling restaurant kitchen, chefs chop, stir, taste and—throw away. Tossing out leftovers and unused bits of food used to go unnoticed. But not anymore, thanks to a device called Winnow Vision that uses machine vision, artificial intelligence and machine learning to identify the type of waste and to generate a report the restaurant can use to reduce it.
The Winnow system takes photos of food as it’s thrown away. Using these images, the machine trains itself to recognise what’s in the bin more accurately than humans. Over time, the system enables kitchens to register food waste automatically.
“Commercial kitchens are wasting twenty to twenty-five percent of their volume materials,” says Kevin Duffy, Winnow’s co-founder. “Winnow's artificial intelligence now makes food-waste tracking so easy and accurate that it should become the standard in every commercial kitchen.”
One third of the world’s food production is wasted. According to the UN, the resources consumed in producing this wasted food is equal to 3.3 billion tons of CO2 emissions and water equivalent to three times the volume of Lake Geneva. In developed countries, food waste occurs mainly at the end of the food supply chain, at consumption. This includes households, as well as “made to stock” kitchens, Winnow’s target market, which includes restaurants, hotels, cruise ships and staff canteens.
Sustainability in artificial intelligence food savings
Winnow’s first product was a manual tool named Waste Monitor. Daily reports by this tool can help kitchens make smarter decisions, saving up to half the food that might otherwise be wasted.
The company has now launched its second-generation product, Winnow Vision, which integrates machine vision and artificial intelligence. Over time, Winnow Vision becomes smarter. Its predictive ability saves users time and reduces human error. Eventually, it reaches full automation, giving kitchens pinpoint accuracy without any input or energy from the kitchen staff. The time the machine needs to attain predictive automation depends on the volume of data collected. The more systems there are capturing data, the quicker the path to automation.
“This is machine learning technology,” says Maria Lundqvist, an economist at the European Investment Bank, which is financing Winnow. “It is not static. The more you use it the more effective it becomes.”
Artificial intelligence food waste software in Romania
Winnow has offices in six countries, one of them in Cluj, Romania, an emerging information-technology hub. The Cluj office serves as the company’s research and development centre. Aiming to increase its staff and further develop its technology, Winnow signed a loan of €7.5 million with the European Investment Bank, the EU bank.
The loan is made under the European Growth Finance Facility, which supports innovative companies. It’s expected to help Winnow boost its workforce to 88 in Cluj, of which 66 will be software developers.
“Financing the R&D is financing skilled people,” says Francisco Alves Da Silva, the loan officer at the European Investment Bank who worked on the deal. “Apart from fostering skilled job creation, the loan is also in line with the European Investment Bank’s sustainability commitment. It also supports R&D in a cohesion country.”
Artificial intelligence food waste reduction becomes a reality
Based in the UK, Duffy founded Winnow in 2013 with another American entrepreneur, Marc Zornes. The company’s technology is being used in over 1,000 commercial kitchens in more than 40 countries.
“This cutting edge technology is simple to use and can increase awareness about food waste and also save money and emissions,” says European Investment Bank economist Lundqvist. “It is a win-win situation.”
Here’s how the win-win is working out, with chefs using Winnow technology saving:
- Over $33 million a year in reduced food purchasing costs
- 23 million meals a year from what would otherwise have been trash (that’s one every two seconds)
- 36,000 tonnes of CO2 emissions avoided per year
Duffy reflects that “it is hard not to be proud. We have achieved quite a bit. But as a founder, my ambition for the impact that we should have does not stop here. We want to move the market to recognize the size of the problem and the benefits of providing a solution.”