KIProWERT: AI-supported value stream optimisation for production
The KIProWERT research project is developing an automated, data-driven value stream optimisation system for industrial production processes. The central question is how machine learning, automated process simulation and large language models can be combined in such a way that users without in-depth specialist knowledge can make well-informed decisions. Self-learning analytical methods and an intelligent human-machine interface automatically interpret possible courses of action and explain them in an accessible manner.
The approach
KIProWERT combines process mining, simulation-based methods and machine learning into an integrated system. Value streams are continuously analysed, and optimisation measures are automatically derived. The assistance system dynamically identifies bottlenecks and evaluates alternative courses of action using simulation, requiring significantly less manual effort than previous approaches. Large language models present the analysis results in a clear and understandable way and enable dialogue-based interaction to support decision-making.
Practical benefits
SimPlan AG intends to continue developing the methods after the project ends and to transform them into marketable, web-based ‘as-a-service’ solutions. The aim is to provide easy access to simulation-based value stream analysis, particularly for small and medium-sized enterprises. The results will also be incorporated into the further development of SimVSM. Landshut University of Applied Sciences utilises the project results in research, teaching and knowledge transfer, as well as a basis for further research projects.
Project partners
Funded by
The research project is funded by the Federal Ministry for Research, Technology and Space.
Funding began in September 2026 and is expected to continue until the end of August 2029.
