From the end of 2020, the cooperation partners KIT and Fraunhofer-Gesellschaft want to plan, test and transfer new production technologies to industry in the 'Karlsruhe Research Factory'. The project is also intended to make an important contribution to the German government's recently adopted "Artificial Intelligence Strategy": "The research factory and the machine learning methods and tools to be developed there and the targeted search for process parameters are a first manifestation of the goals formulated in the German government's current AI strategy," explains Dr. Olaf Sauer, who is coordinating the planning of the 'Karlsruhe Research Factory' on behalf of the Fraunhofer-Gesellschaft as deputy director of the institute at the IOSB.

Collaboration with industry is also important to the project partners: in order to transfer results quickly and in a targeted manner, interested companies are to be involved from the outset - through close cooperation, joint projects and workshops.

Key data on the construction of the research factory

A total budget of around 15 million euros has been earmarked for the implementation of the construction project. The cooperation partners KIT and Fraunhofer will each contribute half of this; in addition, there will be investments in the initial equipment of the production halls, laboratories and offices. Following the laying of the foundation stone in summer 2019, the L-shaped building will house around 70 scientists on two floors and an area of 4,500 square meters from the end of 2020. It will also provide 50 workspaces for cooperation partners from industry.

The project is funded by the Federal Ministry of Education and Research, the Ministry of Science, Research and the Arts Baden-Württemberg, the Ministry of Economic Affairs, Labor and Housing Baden-Württemberg and the European Regional Development Fund.

Specific objective and scientific approach

The aim of the 'Karlsruhe Research Factory' is to systematically develop and expand leads in new, challenging manufacturing processes. The scientists want to learn how to manufacture high-quality products at a very early stage - i.e. when the manufacturing processes required for a new product are not yet fully understood and mastered. Using the latest measurement, sensor and control technology, they want to develop methods that are suitable for quickly converting new production technologies into safe and profitable industrial manufacturing processes. Production can start at a very early stage because intelligent process controls ensure that the first high-quality products are manufactured despite the still immature production technologies.

Specifically, this is to be done as follows: Machine learning and artificial intelligence processes use the data collected by sensors to recognize correlations between quality-related data and process parameters. In this way, the production plant, which is already in operation, "learns" which parameters produce good results. The declared aim of the Research Factory stakeholders is not only to apply machine learning and artificial intelligence to individual production steps or directly successive processes, but also to record and improve entire process chains.

This shortened "time-to-market" should enable small and medium-sized companies in particular to be present on their target markets with new products much earlier than before.