News_1920x250_Detail

Focus on sustainability and AI: "it's Showtime" presents pioneering innovations for SMEs

On 9 October 2024, the it's OWL innovation network presented the results of its fourth project tranche, which runs from 2022 to 2025. Nine projects with a total volume of around 40 million euros were presented - all with a view on sustainability, Artificial Intelligence (AI) and digitalisation. A particular focus: the application of "green digital twins" to optimise production processes in an environmentally friendly way and reduce greenhouse gas emissions.

The Institute Industrial IT (inIT) of the Technische Hochschule Ostwestfalen-Lippe (TH OWL) was also represented with the AI project AI4ScaDa, which presented an innovative solution for small and medium-sized companies (SMEs).

From left to right: The project staff of the AI4ScaDa project Christoph-Alexander Holst (inIT), Jan Ewerszumrode (Miele), Marvin Schöne (HSBI), Julian Bültemeier (inIT), Frauke Nyssen (GEA) and Matthias Heinrich (GEA) at the "it's Showtime" event organised by it's OWL.

Marvin Schöne (left) and Julian Bültemeier present the AI4ScaDa project demonstrator.

Knowledge transfer to SMEs: practical application of research results

Representatives of the it's OWL advisory board, the management of the competence network and the cluster board came together at the Centrum Industrial IT (CIIT) on the Innovation Campus Lemgo to review the impressive progress of the fourth project tranche, which runs between 2022 and 2025. One highlight of the event was the direct transfer of knowledge to SMEs. The projects presented illustrate the potential for sustainable innovation and practical application in the economy.

AI4ScaDa: AI solutions for sparse and incomplete data

One of the projects presented is AI4ScaDa, a cooperation project between inIT, Bielefeld University of Applied Sciences and the corporate partners GEA Westfalia Separator Group GmbH, Miele & Cie KG and Saaten-Union Biotec GmbH. Julian Bültemeier, research associate in Prof. Dr. Volker Lohweg's "Discrete Systems" working group at inIT, presented the project results together with his colleague Marvin Schöne from Bielefeld University of Applied Sciences. AI4ScaDa aims to develop AI-based solutions for analysing sparse, incomplete or inconsistent data - a challenge faced by many SMEs.

Detailed description of the AI4ScaDa demonstrator

As part of the presentation, a demonstrator was introduced that depicts the entire workflow for processing sparse data. This begins with the collection of data, followed by preparation, in which incomplete or inconsistent information is structured and harmonised. This is followed by analysis, in which AI algorithms analyse the data. A special aspect of the project is the application of active learning, which makes it possible to continuously improve the model through targeted feedback and human intervention. This enables precise and reliable results to be achieved even with small data sets.

Sustainable technologies and AI are shaping the future of SMEs

Julian Bültemeier summarises the importance of the project: "Once again, it has become clear that the topic is particularly important for SMEs. They often don't have large enough data volumes for conventional AI models - and this is exactly where we come in with our solution!"

The "it's Showtime" event impressively illustrated how sustainable technologies and AI have the potential to shape the production and logistics sector and strengthen the SME sector in Germany in the long term.

More about the AI4ScaDa project:

https://www.init-owl.de/forschung/projekte/detail/ai-for-scarce-data-maschinelles-lernen-und-informationsfusion-zur-nachhaltigen-nutzung-von-labor-und-kundendaten/

https://its-owl.de/projekte/wenig-daten-viel-erkenntnis-entwicklung-von-ki-anwendungen-fuer-small-data-ai4scada/