The University of Wuppertal is launching the ‘GenISys’ research project together with two industrial partners. The aim of the project is to use generative AI models to make the design and configuration of bottling plants more intelligent and efficient. The aim is to contribute to the promotion of artificial intelligence in relevant sectors of the economy.
Generative AI technologies, which have so far primarily been used in the generation of texts and media content, are now also to be utilised in the industrial production sector. Dr Hasan Tercan, who heads the project at the Chair of Technologies and Management of Digital Transformation, explains:
“In many industrial sectors, the possibilities of generative AI are still untapped. Sometimes because AI methods are not yet adapted to areas of application with very specific requirements.“
Challenges and solutions
The configuration of industrial filling systems, especially for materials such as cement, is often complex and resource-intensive. This includes multiple laboratory tests and the step-by-step development and testing of prototypes. Adapting the systems to changing operating conditions and material properties is another critical and often recurring step.
“Our goal is to enable a more flexible and efficient configuration process through the use of AI,“ says Tercan.
As part of GenISys, the researchers are working with the companies Snap and Haver & Boecker to reduce the number of test cycles required. The use of generative AI methods should help to implement innovative solutions more quickly and conserve resources and the environment. The AI models developed could also be used in other industries in the future, for example through licence models.
The project's AI-based software is designed to plan new filling systems based on data on material properties and historical system configurations. Advanced image recognition methods and generative adversarial networks are used to create and continuously improve the optimum plant configuration. An important component is the integration of human feedback in order to continuously increase the accuracy of the AI models.
Funding and recognition
The project ‘GenISys - Intelligent system for resource-saving plant configuration with generative AI technology’ is funded by the Ministry of Economic Affairs, Industry, Climate Protection and Energy of the State of North Rhine-Westphalia and the European Union as part of the NEXT.IN.NRW competition. The funding totalling around 1.3 million euros, of which the University of Wuppertal will receive around 450,000 euros, was recently presented to the project partners in Düsseldorf by State Secretary Silke Krebs.
