Robots Weblog | fischertechnik Studying Manufacturing unit 4.0 on the College of Potsdam: Optimize neural networks in manufacturing management with fischertechnik

Research into the application of artificial intelligence in production control is being simulated at the University of Potsdam using a factory simulation system from fischertechnik.

The fischertechnik learning environment is used for learning and understanding Industry 4.0 applications in vocational schools and training as well as for use in research, teaching and development at universities, in companies and IT departments. (Image © fischertechnik)

Marcus Grum is a junior professor for business informatics, especially for AI-based application systems. He leans over a model made of fischertechnik building blocks that is approximately one meter by one meter in size. He watches with fascination as a red wooden block is transported from a small high-bay warehouse via a conveyor belt to various production sites. fischertechnik’s simulation model is used at universities and in training to map the ordering process, the production process and the delivery process in digitalized and networked process steps.

At the University of Potsdam, the fischertechnik Learning Factory 4.0 was used in combination with other simulation tools and a neural twin to recreate global production control processes. In one series of research, for example, jam production was simulated. It was assumed that products were manufactured at four different manufacturing sites worldwide, but each had different IT systems.

All processes were examined, from procuring the fruit to selling it to the customer. The fischertechnik learning factory 4.0 represented one of four of the networked haptic production stations. In the system, the process was carried out from storing the fruit to introducing it into the cooking machine to make jam to filling the jars and
the delivery of the products is simulated. The haptic model was connected to a neural network (ANN – Artificial Neural Network). “We used this to carry out series of tests and experiments that deliver valid results and highlight production inefficiencies. From these results we can derive how such inefficiencies can be avoided,” explains Prof. Grum. The young scientist was honored last year by the Werner-von-Siemens-Ring Foundation for the development of a methodological approach that makes it possible to model machine and business processes with the help of artificial neural networks (ANN).

Future research will address empirical verification of neural network instructions and corresponding management interventions that emerge in real-world real-time productions. “Fischertechnik’s factory simulation system will also be used here,” says Prof. Grum.

Further information about the fischertechnik factory simulation systems: Simulation models for industry and universities – fischertechnik