- Arla Foods has new artificial intelligence software to better estimate how much milk 1.5 million cows belonging to 10,300 farmers will provide in the future. The Denmark-based cooperative said in a release about 440 million pounds of milk annually can now be used more efficiently and sustainably using this AI tool.
- Before developing the software, company personnel would have to make manual calculations on Excel spreadsheets. The task can now be done in a few hours and is 1.4% more accurate than the previous system, the company said.
- The software is being used in all of Arla's European markets, including Denmark, Germany, Sweden, the U.K., Belgium, Luxembourg and the Netherlands.
Like many dairy producers today, Arla is doing whatever it can to maximize efficiencies and make sure profits and sustainability are at the highest levels possible. As the largest dairy company in the U.K. — and with thousands of farmer-owners as members — it doesn't make sense to continue doing things in a labor-intensive way when there are high-tech tools available.
The AI forecasting tool isn't the only forward-looking approach the dairy cooperative is taking. In 2017, Arla automated some of its cheese-making procedures in Denmark and started using robot arms equipped with sensors to milk cows. With plant-based competition stealing consumers in the global marketplace, the company can't afford to be inefficient.
Given the current market, it makes sense for companies like Arla to try to maximize profit. In Europe, plant-based dairy alternatives are forecast to increase at a compound annual growth rate of 7.2% between now and 2024, according to Mordor Intelligence.
Numbers are similar in the United States, where total dairy sales reported by the Dairy Farmers of America — representing about a third of producers — fell by about $1.1 billion last year, according to Eater. While some of that was attributed to a drop in average milk prices, sales of beverages made from plants and nuts jumped 9% in 2018, to a total of $1.6 billion, the Plant Based Foods Association found.
The new software will allow Arla to predict how much milk will be produced by factoring in a number of variables. These include seasonal changes, how many farmers are converting to new types of milk, geographical aspects and how much milk farmers are producing each day. Consequently, the company said it is now possible to determine the amount of milk expected from farmers in different parts of the world several months in advance.
There are also transportation and environmental advantages to this AI approach. By planning ahead, companies can reduce cost and save the environment from unnecessary carbon dioxide emissions. Since consumers are increasingly looking for brands to be more sustainable, this could appeal to shoppers who are turning away from traditional milk.
Arla recently said it would move up its transition to more sustainable dairy production by focusing on farming procedures, according to Food Ingredients First. The company has set a goal of reducing greenhouse gas emissions by 30% per kilo of milk during the next 10 years and achieving net zero carbon emissions by 2050. It seems to be making better progress than some. While the global average is 2.5 kilograms of carbon dioxide per kilo of milk, Arla's farmer-owners average less than half of that — 1.15 kg of carbon dioxide per kilo of milk, the company said.
With that track record, it's likely other dairy companies will be investigating AI technology and finding ways to implement more sustainable production practices. By conveying a greener image, dairy brands might appeal to consumers who have turned their backs on milk and other dairy products and turned toward plant-based alternatives. By using AI and other high-tech tools, dairy operations may be able to boost efficiencies and stay ahead of the competition.