Getting the best out of production

3,000 square meters, several floors, powered by renewable energy: More and more factories are also being built in Germany to develop cutting-edge AI models and applications for industry. The technology is intended to ensure competitive, automated production of highly complex and individual products in the long term. The benefits for economic performance are already visible today. Studies show that productivity growth has almost quadrupled since the spread of GenAI. It is estimated that an average annual increase of 0.9 percent is possible by 2030 and a further 1.2 percent by 2040. Applications such as personal AI assistants, chatbots, voicebots and AI agents will make processes across departments and systems more efficient, qualitative and future-proof.

Making complexity manageable

In most cases, artificial intelligence - specifically machines or deep learning - is used to analyze large amounts of data, identify patterns, create forecasts, plan scenarios and receive situation-specific recommendations for action. The decision remains with humans. For example, whether a product is faulty, needs to be rejected or improved. On the one hand, this form of quality control reduces workload and increases customer satisfaction. On the other hand, it also enables minimum downtimes, maximum machine running times and higher added value overall. It also contributes to sustainable production by reducing waste and saving material and energy. Other application scenarios include smart product development and adaptation, full traceability of the value chain, machine monitoring and predictive maintenance and servicing.

Successful AI implementation requires reliable access to operational and historical data via a modern enterprise data platform. To achieve this, the entire production process must be networked via sensors and algorithms must be developed to convert this data into information. Certain sub-processes can then be controlled by AI-supported automation with minimal human intervention. This ranges from data evaluation in real time to produce a new product version to independent further development of the product. The complexity of finding the right or optimal solution from numerous options cannot be directly reduced through the use of AI, but it does become manageable.

Taking employees with you

As positive as AI in production is, employees will ultimately also have to adapt to a much faster pace of change. For them, jobs and requirements for specialized practical skills are changing, and there is a new understanding of good leadership. Training alone is not enough; employees need to be inspired to adopt an open, innovative and critical mindset towards new technologies.

Many companies are still reaching their limits here. The problem: for a long time, digital transformation was seen as a purely technological project. A standard solution, a management tool or an MES system was simply introduced. In theory, everything seemed to fit. But in reality, the results were often disappointing. Why? Because the crucial point was ignored: A factory cannot be controlled from a data center. The transformation begins in production by involving those who actually operate it.