KI4KI project meeting:
How can internal knowledge be processed in such a way that it can be utilised by artificial intelligence? This was the question addressed by the participants at the third network meeting of the KI4KI project, which took place on 8 June 2026 at the IKV in Aachen.
Aachen, June 8, 2026 – The third networking meeting of the KI4KI project (Artificial Intelligence for Plastics Processing) focused on the structured processing of internal knowledge bases for use in AI applications. Using a local, Retrieval-Augmented Generation (RAG)-based system, the study demonstrated how existing domain knowledge can be efficiently harnessed and made available for various use cases.
© IKV
© IKV
© IKV
© IKV
© IKV
© IKV
© IKVKey topics included current developments in the field of artificial intelligence, such as agent-based AI, multimodal models, hybrid architectures, and local AI systems. In addition, the “Knowledge Management” AI demonstrator was presented, which is specifically designed for applications in the areas of maintenance and fault analysis.
Another focus was on the technical implementation of modern knowledge management systems. Participants gained insights into techniques such as chunking, embeddings, vector databases, and the local management of AI models. Specific applications were demonstrated using structured experimental data and expert knowledge.
The discussion also covered requirements for data quality, data availability, system integration, and traceability—key prerequisites for the successful use of AI in an industrial setting. It became clear that powerful models alone are not enough: What matters most is a robust data foundation and the effective integration of the systems into existing processes.
The program was completed by expert talks on the topics of requirements management and AI-driven product development. At that point, the next focus of the project was already taking shape: applying the insights gained to future applications in product development.
IKV would like to thank everyone involved for the constructive dialogue and looks forward to continuing to work together within the network.