Hybrid grey box models for the determination and prediction of quality attributes using food as an example.
Sponsored by: Federal Ministry of Education and Research (BMBF)
Project Manager: Deutsches Zentrum für Luft- und Raumfahrt e.V. (DLR)
Duration: 12/2022 – 09/2025
Consortium: benelog GmbH&Co.KG, cubert GmbH, tsenso GmbH, Universität Freiburg
The aim of the FreshTwin project is to use modern measurement methods and AI to make the actual properties of food transparent along the supply chain, from the producer to the consumer. Existing logistics and quality assurance data as well as secondary data sources, such as weather conditions at the harvest site, can be used to achieve an initial, structural transparency of quality. This data is supplemented by the use of modern, fast measurement and analysis methods such as spectroscopy and image recognition. In the FreshTwin cloud infrastructure, these two approaches, data aggregation and innovative measurement methods, are brought together, thus enabling a precise, descriptive recording of the actual state.
The research project can sustainably advance the digitalisation of the German food industry – for a more competitive German food industry, more tax revenue and more jobs.