“Conventional planning systems reach their limits when processes in industrial production—or in other areas—become too complex. In recent years, we have developed self-learning algorithms that can efficiently optimize processes with multiple sub-steps,” says Verena Wolf, a professor of computer science at Saarland University. These research findings are now to be translated into industrial practice as quickly as possible. “We are focusing on what is known as sequencing—for example, in automotive production, where certain components must be available at a defined time to be assembled into the vehicle in the correct order,” explains the researcher. In this context, artificial intelligence opens up new possibilities for incorporating the complex interdependencies of supply chains, model variants, and staff availability into process planning, thereby significantly increasing manufacturing efficiency.
Together with the German Research Center for Artificial Intelligence (DFKI) and the company abat+ GmbH, Verena Wolf plans to develop an open, scalable planning platform that can be used to optimize all types of production data with the help of AI. “We’re also creating digital twins to train AI models in a simulation environment so they’re tailored to real production systems. In addition, we want to support small and medium-sized enterprises—which still plan many processes manually—in their digital transformation,” explains the professor.
The planning system will focus, on the one hand, on production processes being built from the ground up. “But we also want to examine ongoing production lines that require last-minute rescheduling—for example, because the delivery of individual components is delayed or certain vehicles must be prioritized. These are the daily challenges in industry, which are becoming even more acute amid global competition and supply chains that are not very resilient to crises,” says Verena Wolf.
The AI-powered platform is also intended for planning scenarios beyond industrial production. For example, it could be used to optimize the complex workflows in the operating rooms of large hospitals. “For every surgery, different surgical instruments must be prepared, high-tech medical devices must be reconfigured, and the appropriate specialized staff must be made available,” explains Verena Wolf. Time and costs could be saved by using the AI-powered planning platform to schedule surgeries with similar requirements back-to-back, thereby minimizing the effort required for preparations.
In the transfer project now underway, the company abat+ GmbH in St. Ingbert will contribute its expertise in the necessary software and cloud infrastructure, as well as its experience in production planning for large industrial companies. In the coming months, the planning platform is to be expanded into an AI toolkit that can be quickly and individually adapted to the specific challenges of industrial clients. The project, titled “Preparing an AI-Based Rearranging Hub” (PreAIrranging), is led by computer science professor Verena Wolf, Timo Philipp Gros (DFKI), and Philipp Stopp (abat+ GmbH). It is funded by the European Regional Development Fund (ERDF) and the European Union.