Stadt-Land-Fluss

Vegetable on market stall on the famous Munich Viktualienmarkt in the centre of Munich, Germany
© iStock / Nikada

The joint research project “Stadt-Land-Fluss (SLF)” developed a sustainable solution for strengthening “regional food systems” using various AI tools based on a multi-layer “SLF-ICT ecosystem” as a service and integration platform as part of the research focus “Artificial Intelligence (AI) in agriculture, the food chain, healthy nutrition and rural areas”.

The project aimed to make the regional food system more regional and sustainable through digital, AI-supported innovations. Consuming regional products strengthens the local economy and studies show that consumers want more regional products and transparent information about them. Regional producers need information about customer needs and purchasing behavior in order to plan cultivation. The installation and management of a solid database for nutritional data and the associated distributed IT infrastructure are therefore crucial for the development of a digital, regional nutrition system.

In the project, a service platform was therefore developed as an ICT ecosystem for the digital support of the regional nutrition system. The ICT ecosystem enables the communication and consolidation of relevant information and data-processing components, and connects consumers, suppliers and producers. SLF was equipped with various demonstrator components, including some based on AI technologies, for selected use cases that can be transferred to other regional food systems:

  • SLF recommender as a “content-based recommendation service”
  • SLF route planning in food logistics using constraint programming
  • SLF chatbots with speech and natural language processing (NLP) and dialog management by enriching behavior-oriented AI approaches (ML) with declarative knowledge

The underlying ICT ecosystem is based on a distributed, open reference architecture according to DIN SPEC 91357 and facilitates the integration of additional services and data sources. It promotes flexibility, scalability and sustainability. Participatory methods, training, evaluation processes and business model development complemented the technical developments.Fraunhofer FOKUS developed the ICT ecosystem, placing particular emphasis on the interoperability of its components: an AI component for optimized route planning, a component for the optimal selection of central (food) distribution warehouses, and a water consumption component as part of the SLF recommender.

For the optimzed route planning of delivery tours within supply chains, SLF has developed the requirements for the criteria to be met and optimized for low-emission planning of delivery and pick-up tours. For this purpose, a solution approach based on the “Constraint Programming” (CP) paradigm was developed and implemented using FOKUS' own CP solver firstCS, and provided as a web service for testing and integration.

In a further step, a graphical front end was developed that can be used to operate the SLF route planning interactively and intuitively via smartphones or tablets and to visually display the results.

In addition to Fraunhofer FOKUS as the coordinator, the project consortium consisted of the following research partners: German Research Center for Artificial Intelligence, Eberswalde University for Sustainable Development, Technical University of Berlin and the following SMEs: PIELERS GmbH, GHS GRUBER & HUFNAGEL Softwareentwicklung GmbH, nearbuy GmbH, Lienig Wildfruchtverarbeitung GmbH and Terra Naturkost Handels KG as well as the network associations “Association of the Software, Information and Communication Industry in Berlin and Brandenburg e. V. (SIBB), Pro agro – Association for the Promotion of Rural Areas in the Brandenburg-Berlin Region e.V.

Publications

  • Cuno, Silke et al. (2021): Datenplattformen und KI-Werkzeuge zur Stärkung der regionalen Ernährungssysteme in: Stadtforschung und Statistik: 34, 2, 2021, ISSN: 0934-5868, PID: https://nbn-resolving.org/urn:nbn:de:0168-ssoar-75080-2.
  • Cuno, Silke; Lämmel, Philipp (2024), Use of Semantic Artefacts in Agricultural Data-Driven Service Development”, in: „DATA 2024, 13th International Conference on Data Science, Technology and Applications”, https://data.scitevents.org, Dijon, July 2024 (Veröffentlichung folgt).
  • Wolf, Armin; Cuno, Silke (2023): Emission-Reducing Vehicle Routing in Food Logistics. 
in: INFORMATIK 2023 - Designing Futures: Zukünfte gestalten. DOI: 10.18420/inf2023_163.  Bonn: Gesellschaft für Informatik e.V.. PISSN: 1617-5468. ISBN: 978-3-88579-731-9,  Ökologische Nachhaltigkeit - Kolloquium Landwirtschaft der Zukunft - Ist KI ein wesentlicher Schlüssel zur nachhaltigeren Landwirtschaft?.  Berlin. 26.-29. September 2023.

Funded by

Logo Bundesministerium für Ernährung und Landwirtschaft