DaPro
Data-driven, machine-learning based process optimization in the beverage industry
Application industry: Food; process industry
Technology area: (Mathematical) optimization and planning; machine learning
DaPro is developing standards and methods for evaluating large amounts of data in the beverage industry using scenarios in the brewing industry as an example. Data analyses are supported by machine learning to optimise complex production processes.
The baseline
The beverage industry is under high competitive pressure: With prices remaining flat, sales decline while costs, especially energy costs, are on the rise. In light of this, manufacturers are being called upon to optimise production processes. The potential for this can be found in the evaluation of process data, i.e. of values obtained from sensors in technical processes. Automation and digital control have been used in beverage production for a long time, so enormous amounts of data are already available. However, the economic efficiency of evaluating this source has been very limited up to now. Progress, especially in the field of data mining, i.e. in the systematic application of statistical methods to large data sets (Big Data), now offers an opportunity to do so.
The project goal
The DaPro research project aims to develop a reference architecture that allows both producers and manufacturers of beverage machinery to collaboratively use data generated in the production process. One core undertaking is the development of a toolbox of data mining modules, i.e. building blocks for analysing big data using machine learning methods. These analysis modules will be defined and tested in practice with partners from the brewing industry. The objective here is to describe and control complex biochemical processes with the help of data mining to optimize the quality of individual stages of the brewing process. One example of such a scenario is predictions about certain processes during fermentation.
Application and practical benefit
Alongside the economic benefits expected from data-driven process optimization, the partners from the brewing industry are also pursuing the goal of making their production more sustainable from an ecological point of view, for example, to save energy. Beyond the concrete application scenarios, DaPro is intended to contribute to advancing the economic use of data in the beverage industry and for the entire process industry.
Term: 1 January 2019 to 30 June 2021
Consortium: Bitburger Braugruppe GmbH (lead member), RapidMiner GmbH, SYSKRON X GmbH, RIF Institut für Forschung und Transfer e.V., Augustiner-Bräu Wagner KG
Contact:
Dominik Polster
BITBURGER BRAUGRUPPE GmbH
Email: Dominik.Polster@bitburger-braugruppe.de
Website: www.dapro-projekt.de
- Recommend this page:
- Print view