The challenge of the joint project "ML4Pro^2" is to make machine learning (ML) sustainably available for Intelligent Technical Systems (ITS). This requires the transfer of the latest ML methods to the fields of action central to ITS in order to bring ML technologies into products and production chains, and conversely to raise companies' awareness of when and how ML can be integrated into agile business models and production chains. The joint project builds on the digitization strategies of the participating companies and the ML expertise of the participating research partners to realize the step towards efficient use of digital data through ML. Technical innovations are increasingly based on machine learning.
ML has the potential to generate added value by extracting knowledge from digital data at all stages of business processes. With the current ML research topics "hybrid learning methods", "integration of expert knowledge", "interpretability" and "learning on data streams in embedded systems", the joint project addresses central issues for ITS. The ML methods are considered across applications on the basis of three industrial use cases that are future-oriented for both production and its products.
- The joint project combines proven research facilities for ML with practical experience knowledge of the companies.
- The project partners integrate technical expertise as well as management and implementation experience in ML and its application for ITS.
- The research partners contribute their research results on the ML fundamentals and the companies their ML application knowledge as well as their problem expertise.
- The transfer of the results and the transferability to third parties are ensured by the establishment of an ML platform.
- This platform includes reference implementations of evaluated ML methods, methods for data management, data pre-processing, data visualization, and explicit application knowledge about typical procedures when using ML methods.
- The already existing ML-Labs at the research institutions in OWL demonstrate the state of the art of ML technology and will make ML tangible.