Category Research project
  • Mikrobiologie
  • Expositionsschätzung

Future scenarios for consumer protection based on food quality and safety information, innovative measurement methods and artificial intelligence

Project status
Completed
Project start
Oct 2021
Project end
May 2025
Acronym
Zukunftslabor2030
Department
Biologische Sicherheit

Description and Objective

The aim of the project was to strengthen sustainable consumer health protection in the future by using innovative measurement methods and artificial intelligence (AI) processes to collect and evaluate data on the quality and microbial safety of food along the supply chain right up to the consumer. Within the framework of the project, scenarios are being developed as to how the consistent use of the "Internet of Things" and AI along the food supply chain from production to the consumer can achieve significant improvements in the areas of consumer protection and information, in the monitoring of food quality and safety, and with regard to a possible reduction in food waste. The project achieves these goals on the one hand through the intelligent, self-learning linking of classical quality and safety data; on the other hand through the use of innovative measurement methods: mass spectrometry, next generation gene sequencing, Raman, infrared and fluorescence spectroscopy as well as micro gas sensors. These methods will be investigated in the project with regard to their practical suitability in the meat sector. The possible added value of linking the methods within the framework of an intelligent, multivariate evaluation will be explored. Also related questions in the field of food law will be investigated. The future scenarios developed in the project will be designed together with a panel of experts from the food sector. Selected scenarios will be exemplarily implemented in practice within the framework of a FutureLab2030 and evaluated with regard to their practicality, legal implications, costs and added value. The scenarios and the results of the FutureLab2030 will be published as a white paper at a project conference.

Result

In the BfRshort forGerman Federal Institute for Risk Assessment sub-project, the focus of the work in the area of innovative measurement methods was on analysing the composition of the microbiome in the food matrix and its storage-related changes using culture-independent methods. The metagenome analysis carried out by the BfRshort forGerman Federal Institute for Risk Assessment proved to be a suitable method for detecting changes in the microbiome profile in the food matrix of minced pork. This made it possible to determine the influence of the tested storage temperatures (2 °Cshort fordegrees Celsius, 10 °Cshort fordegrees Celsius and 14 °Cshort fordegrees Celsius), the different packaging atmospheres (MAP 70 % O2 + 30 % CO2 and MAP 70 % N2 + 30 % CO2), meat maturation (7 days at 2 °Cshort fordegrees Celsius) and production conditions (in-house production by MRIshort forMax Rubner Institute, external production by industrial slaughterhouses) could be demonstrated. However, it was also found that metagenome analysis is only of limited suitability for detecting breaks in the cold chain (e.g. on the second day of storage for 6 or 12 hours at 14 °Cshort fordegrees Celsius, otherwise 2 °Cshort fordegrees Celsius), as the changes in the microbiome profile were too small in this case. Furthermore, metagenome analysis allowed the functional annotation of microbial sequences in the food matrix, with the quality of the analysis depending on the sequencing depth, i.e. the proportion of bacterial reads detected.In addition, the BfRshort forGerman Federal Institute for Risk Assessment played a key role in the design and establishment of a software architecture for the Zukunftslabor2030 platform. This comprises the following components: the openBIS data management platform, digital twins (forecast models) in FSKX format, a web application for users and the EPCIS event hub for standard-based communication between the various modules of the ZL2030 platform. The project demonstrated that the FKSX format developed by the BfRshort forGerman Federal Institute for Risk Assessment is suitable for describing a wide variety of prediction models in a standardised manner, enabling the models to be executed in a cloud-based runtime environment. These execution environments can be operated as container-based systems on different cloud infrastructures and networked via APIs, so that suitable, isolated and tailor-made environments can be used for the heterogeneous FSKX models of the digital twins. The functionality of the ZL2030 platform was demonstrated in a proof of concept. Furthermore, the BfRshort forGerman Federal Institute for Risk Assessment, in collaboration with project partner benelog GmbH & Co. KG, has published a white paper describing the FSKX cloud platform used in the project for the execution of scientific models (10.5281/zenodo.15835206). The ZL2030 project is mentioned in it as an exemplary use case.
Type of project

Third-party funded project

Research focus

Expositionsabschätzung und Bewertung biologischer Risiken / Forschung zur Sicherheit nationaler und internationaler Warenketten

Organisational units and partners

Lead specialist group: Warenkettenmodellierungen und Künstliche Intelligenz (1IZ)
Contact persons: Dr. Tasja Buschhardt, Matthias Filter
External partner: Max Rubner-Institut, Bundesforschungsinstitut für Ernährung und Lebensmittel, Bayerisches Landesamt für Gesundheit und Lebensmittelsicherheit, Universität Bayreuth, Fraunhofer-Institut für Verfahrenstechnik und Verpackung , tsenso GmbH, benelog GmbH & Co. KG, Technische Hochschule Deggendorf

Funding body and grant number

Bundesministerium für Ernährung und Landwirtschaft
28DK126F20