Carlsberg collaborates on AI research into beer flavors and aromas

07 Nov 2017 --- Carlsberg has announced a multi-million-dollar investment in a research study with the purpose of measuring and sensing flavors and aromas in beer. The research study, called “The Beer Fingerprinting Project,” is based on an idea by Jochen Förster, Director and Professor Yeast Fermentation, Carlsberg Research Laboratory.

The development of a sensor platform holds enormous potential for broader research and facilitate new start-ups, says Förster.

“No rapid assays exist today for the determination of flavor compounds in beverages, but it is crucial that we can do this to ensure that the laboratory continues to develop beer of the highest possible quality and provide a model for brewing in Denmark and the rest of the world,” Förster explains.

“We are excited to be part of a team with Aarhus University, The Technical University of Denmark and Microsoft and push the boundaries in sensor technology for flavor determination,” Förster adds. “This will enable us to select and develop novel brewer’s yeast for application in craft, specialty, core and alcohol-free beers at much higher speed and even better quality.”

A research matchmaking event at Carlsberg sparked the idea of a joint project between Carlsberg Research Laboratory and iNano at Aarhus University.

DTU Chemical Engineering brings expertise within reactor parallelization and integration, as well as data processing, to the project.

Microsoft has delivered advanced services and software within the area of artificial intelligence to support the project, and Innovation Fund Denmark is behind a financial investment, according to Carlsberg. The aim is now to validate and mature the technology and make it applicable for high-throughput screening of novel brewing organisms.

“This research study puts advanced analytics and intelligent cloud technology as a cornerstone of the project and combines expertise within several fields of research. We are excited to see the project unfold and determine how it will impact faster go-to-market processes for Carlsberg,” says Ricky Gangsted-Rasmussen, Industry Lead – Retail, Microsoft Denmark.

Today, no such rapid technology for discrimination of complex flavor mixtures exists, Carlsberg notes. Developing methods for fast and reliable assessment of flavors in complex mixtures such as in beer or other alcoholic and alcohol-free beverages are of interest for product development, quality control and safety.

Researchers from iNano Aarhus University have already developed a solution to use novel sensors and proof of principle to differentiate between four Carlsberg beers – Carlsberg Pilsner, Tuborg Pilsner, Wiibroe and Nordic – as part of the project.

Carlsberg notes that the outcome of this project will not only strengthen the Danish position in the world beer market represented by Carlsberg but is also expected to lead to new start-ups as the technology can be used for other industries than beverages, such as the environmental, pharma and food industries.