The global corona crisis has shown that classical assembly line manufacturing cannot react quickly to temporary increases in demand and to changing purchasing behavior. What the industry needs, therefore, are more flexible matrix production systems that can absorb even short-term changes in demand.
Typically, matrix production systems contain independent process modules that can be freely moved into position, and individually deployed. These process modules include specialized technology stations, assembly cells and test stations. Each process module is designed to perform a complete production step, and when connected in series, the modules of the matrix production system constitute an assembly line production system. However, the production system can be flexibly extended by further process modules or completely reassembled at any time.
The goal of the SE.MA.KI research project was to build the prototype of an application-ready matrix production system. The solutions developed by the five participating Fraunhofer institutes ranged from flexible, data-based process planning strategies and the automation of logistics and assembly processes all the way to adaptive software and plant systems.
To implement data-based process planning and control and to better master the complex process interrelationships, the research team used digital methods such as simulations and artificial intelligence (AI). To this end, the Fraunhofer IPT further developed powerful, AI-supported simulation tools and generated simulation software for process chain planning. Together they simulate the process and machine selection based on the product characteristics, process parameters and available machine and personnel capacities. Various process chains are suggested to the operator and are prepared and visualized in a user-friendly manner. The process planning tool can be flexibly adapted to any machine park or process and is available as a software package for manufacturing companies.
The Fraunhofer IPT has integrated wire-based laser metal deposition (LMD-w) into the matrix production system in a robotic cell as a demonstrator of an additive manufacturing process. To this end, the researchers integrated a specially developed LMD-w process head into the Fraunhofer IPT's robot cell and control technology and then successfully put it into operation. In addition to the pure manufacturing process, they also implemented digitized process monitoring: Force and temperature sensors in the process head recorded data and forwarded it to the software tool in near real time. The AI integrated into the simulation software was continuously trained on stable and critical process parameters by the data. In this way, the Fraunhofer IPT succeeded in significantly stabilizing the LMD-w process. The process head can be customized and is offered to the industry as a prototype.
As a second demonstrator for matrix production, the Fraunhofer IPT has integrated a large winding cell into the production system for additive manufacturing of large fiber composite components. Compared to the LMD-w process head, where only one component had to be integrated, the winding of large fiber composite components consists of several interlinked individual components. Since it is connected to process monitoring and AI, this system will be able to produce customer-specific prototypes of real components, for example hydrogen tanks, as feasibility studies and prototypes. This way, it will be possible to develop individual machine concepts as well as construct a production plant.
The SE.MA.KI project was funded by the German Federal Ministry of Education and Research (BMBF) under grant number L1FHG42421.