Yokogawa Electric reduces fermentation time for beer brewing by 28% using an AI algorithm for temperature control

https://bio.nikkeibp.co.jp/atcl/news/p1/25/10/14/13843/

At the beer production facility of CRAFT BANK, Fukuchiyama City, Yokogawa Electric has carried out demonstration experiments  using the company’s artificial intelligence (AI) to optimize the beer fermentation process. Typically, beer production involves brewing, filtering, boiling, adding hops, and cooling, before moving on to the fermentation and aging processes. In the demonstration experiments, Yokogawa Electric tested an autonomous AI algorithm FKDPP, developed in 2018 in collaboration with the Nara Institute of Science and Technology. The demonstration experiment proceeded as follows:

(1) CLAFT BANK carried out the traditional beer brewing process, collecting data every 30 minutes for two weeks during the fermentation process, including temperature and specific gravity, which indicates the beer’s degree of fermentation.

(2) FKDPP learned from this data and created a simulator that replicated the beer brewing process.

(3) FKDPP used the digital twin to create a temperature setting plan for the fermentation tank.

(4) At CLAFT BANK’s brewery, the temperature was manually controlled to follow the AI ​​plan.

During two experiments in 2025, the fermentation process time was reduced by 28%, from 336 hours to 240 hours, and the sensory tests of taste, aroma, etc. showed that beer of the same quality was brown as when temperature was controlled by artisans.

According to Yokogawa Electric, FKDPP is already successfully used in chemical processes during pharmaceutical manufacturing.

Most popular posts: