A key result of the research project is the development of an intelligent assistance system that offers product engineers and assembly workers an end-to-end process chain that enables a more efficient and flexible way of working. The system can be adapted to different production environments and continuously learns in order to provide tailor-made suggestions for process optimization. During the final presentation, the participants were able to experience the developments of the assistance system in live operation.
DiProLeA, short for "Digital Product Creation Process with Learning Assistance System", was carried out by a consortium of renowned companies and research institutions and funded by the Federal Ministry of Education and Research (BMBF). The partners included IBO GmbH, Hutchinson PFW Aerospace GmbH, Rauschert Heinersdorf-Pressig GmbH, YOUSE GmbH, SPIE Automation GmbH, Cognition Factory GmbH, S.K.M. Informatik GmbH, fortiss and Aschaffenburg University of Applied Sciences.
Digital solutions for the production of the future
"The DiProLeA project represents the next step in the digitalization of production processes. With our assistance system, we can offer industrial companies customized solutions in the field of manual assembly. A key challenge here is to take into account both the needs of the workers and the requirements of industrial companies. Both are important for securing the production industry in Germany," says Prof. Dr.-Ing. Konrad Doll, head of the research project at the TH Aschaffenburg.
fortiss made a significant research contribution to the project and developed new approaches for knowledge-based engineering of technical systems, including semantic interoperability and communication capabilities. The experts in the Platform Engineering competence field specified open system architectures for the flexible networking and digital consistency of technical systems. This made it possible to capture complex interrelationships and provide suitable neutral data formats. As a result, they supported the integration of heterogeneous data and the provision of real-time instructions for proactive error detection and process optimization. Dominik Mittel, research associate in the Platform Engineering competence field, explains: "Methods of formal knowledge representation and interpretation allow the automatic integration of heterogeneous data from the product development process and the derivation of instructions for the worker." The system thus enables automatic error detection and proactive support for assembly workers with real-time instructions, which helps to optimize work processes.
Successful application by the project partners
One successful application example is semantic process execution assistance in the assembly of race bearings at IBO GmbH. The "trustkey" assistance system enables flexible process control and monitoring, meets high quality requirements and ensures efficient task completion. The Technical University of Aschaffenburg and the state research institute fortiss have set up a workstation for demonstration purposes that integrates the semantic knowledge base and the process execution system. High-resolution cameras and modern deep learning methods record every step of the production process in the assembly cell. This data is compared with the knowledge database in real time to ensure a seamless digital representation of the entire process.
Another successful application in practice is the production of pipe components at Hutchinson PFW Aerospace. Here, the project partners SKM Informatik and Cognition Factory developed an XR-based worker assistance system that visualizes data and work instructions dynamically and intuitively. This enables rapid familiarization and precise execution of the work steps, which leads to improved product quality and job satisfaction.
