Bel tests digital twin for cheese recipe

Bel, a French dairy company, is testing a digital twin of its popular cheese brand La Vache qui rit at its century-old factory in Lons-le-Saunier. This virtual model aims to simulate the cheese-making process and help develop new recipes more efficiently.

Digital twin combines multiple factors

The digital twin integrates various elements such as ingredients, formulation, physico-chemical properties, sensory perception, and consumer preferences into a single virtual environment. This allows Bel to simulate thousands of recipe combinations before conducting physical tests on the most promising ones.

Predicting taste and consumer preferences

Anne Pitkowski, director of research and application at Bel, highlights the tool's ability to predict the taste and sensory perception of a recipe. It can determine whether a cheese will be creamier or saltier and whether it will appeal to different regional palates, such as American or Vietnamese.

Complementing traditional methods

Bel emphasizes that the simulation is a complement to traditional methods, which involve a series of tests, analyses, and tastings. The company aims to enhance its expertise rather than replace it with the digital tool.

Improving nutrition and sustainability

The technology also aims to continuously improve the nutritional profile and environmental footprint of Bel's products. It was developed by the company's research, innovation, and development teams and is expected to be deployed on an industrial scale by 2028.

Collaboration with Dassault Systèmes

The project is part of a collaboration with Dassault Systèmes, announced in July 2024. The agreement involves deploying the "Perfect Production" solution on Bel's 3DEXPERIENCE platform in 11 of its factories. This includes components for managing manufacturing operations and product lifecycle management.

Digital twin used in other industries

Dassault Systèmes' digital twin technologies are already used in aerospace, automotive, and life sciences. The digitization of Bel's factory, including sensors and analyzers, remains in the experimental phase and aims to reduce losses and manage milk variability.

Investment and uncertainty

Bel invests tens of millions of euros annually in data and technology, though the returns are still uncertain. Antoine Fiévet, president of Bel, acknowledges the risk but justifies the investment by noting the company's need to innovate despite its smaller size compared to competitors like Lactalis and Danone.