In December 2025, the collaborative project between Diehl Metering and the Munich Data Science Institute, “Artificial Intelligence for Predictive Quality Assurance in Manufacturing Processes,” concluded with a final presentation and an in-depth discussion of the results achieved. The goal of the project was to systematically introduce and pilot machine learning methods in quality assurance at Diehl Metering. Based on process data, models were developed to detect defective product components at an early stage, enabling well-founded real-time decisions in electronics and water meter production. A machine-learning model designed for this application not only minimizes statistical error rates but directly reduces the real costs incurred by incorrect decisions. In addition, a robust economic use case was developed that demonstrates the value of AI-supported quality assurance.

The collaboration between MDSI and Diehl Metering was characterized by close and continuous exchange, including regular steering committee and milestone meetings. As part of this partnership, MDSI Core Member Prof. Stefan Minner and doctoral researcher Nicolas Kuttruff were also invited to the production sites in Nuremberg and Apolda to gain first-hand insights into the specific use cases and manufacturing processes. The approaches developed in the project are now being further pursued within the Diehl Group and are gradually being transferred into operational use. Both partners benefit sustainably from the collaboration: MDSI through access to practical data and industrial application cases, and Diehl Metering through in-depth machine learning expertise and solid support in data analysis.

We would like to express our sincere thanks to Diehl Metering for the excellent cooperation and look forward to further expanding our partnership with the Diehl Group and driving forward future joint initiatives.