RWTH’s IoP Cluster Paves the Way for Intelligent, Connected Industrial Future Despite Labor Market Pressures

In addition to material and energy costs, labor costs are among the key costs of production. While material and energy costs fluctuate over time, there are no fluctuations in labor costs in Germany – according to Frank Possel-Dölken, the trend is pointing in one direction only: up. He is a member of the management board of the Lippe-based company Phoenix Contact, a manufacturer of electrical engineering and automation solutions for industry. At the company, which currently employs 22,000 members of staff, he is responsible for digital transformation. “In the end, we need to think in this way: Do we have any leverage that would allow us to afford these wage increases?” Although the company operates globally, 70 percent of its added value is generated in Germany. Chief Digital Officer Possel-Dölken believes that technology must be used to increase efficiency in order to strike the right balance.

He is not alone in such considerations. “There are large companies that say they are not able to offer such wage increases,” says Professor Christian Brecher, holder of the Chair of Machine Tools at RWTH’s Laboratory for Machine Tools and Production Engineering (WZL). Wage costs in German industry were about 44% above the EU average in 2022. In addition, the problem of labor shortages will become even more acute when the baby boomers retire in the coming years. Now science is asked to provide solutions. “How can we create better conditions for industry in Germany?” “How can we involve people and try to get better than the others step by step so that our wage costs can be justified?” The aim is to achieve the best quality, higher productivity, and improved sustainability.

Network for Cutting-Edge Research

Brecher is spokesperson for the Cluster of Excellence IoP. In this interdisciplinary state-of-the-art research consortium, the mechanical engineer is working together with over 35 departments and research institutes in Aachen. They are paving the way for a new industrial future. The aim is to develop a complex network of machines, software, data storage, and people that can exchange and work intelligently with data in real time. The IoP is one of 57 Clusters of Excellence across Germany funded by the federal and state governments, which receive a total of 385 million euros annually. With the start of the second funding phase in 2026, the annual funding for up to 70 Clusters of Excellence is set to increase to a total of 539 million euros.

The Aachen research alliance focuses on key technologies for the production of the future and thus on providing the core of Industry 4.0, which is often referred to as the fourth industrial revolution. Following the introduction of the steam engine, electrical energy and the computer, we are now entering a new dimension of digital transformation: Digital technologies are set to enable needs-based data analysis, for example using machine learning techniques, holistically covering all aspects of production engineering: This will cover the entire process, starting with product development, design, and material selection, through to production and assembly and even new business models.

“We have been looking at what people are doing for a long time. We have analyzed what happens at the machine tool,” says Christian Fimmers, managing director of the Cluster, explaining the innovative approach: If the controller now identifies a problem during quality control, he can look at all the information, both from machines and from employees, and check where the process deviates from the specifications. The data depicts the production process virtually and in real time; this is the so-called digital shadow of real-world production. To this end, all the entities involved in production, that is, people, workpieces, and machines, are intelligently linked with the help of information and communication technologies. Ultimately, they all should communicate and collaborate directly with each other, including with the help of artificial intelligence (AI).

Making Big Data More Manageable

The researchers are doing the groundwork for this project. Machines, sensors, machine control systems, but also people themselves, are already constantly supplying vast amounts of data, which is referred to as “big data”. To make this big data manageable, the researchers have developed an intelligent data preparation process: “We’re talking about our concept of the digital shadow,” says Brecher, illustrating the process using the example of a tie. “In the shadow, we can only see the outline and no longer the color, but we can still see that it is clearly a tie. Which means: We are streamlining this digital world as efficiently as possible so that we can achieve our goal with as little data as possible.” The researchers have now optimized this process to such an extent that the calculations are incredibly fast – they take place “in real time”. In addition, “look-ahead functions” open up completely new ways of monitoring quality.

Possel-Dölken from Phoenix Contact makes it clear how important this is in practice: “It’s about having high-quality information available at the touch of a button – how productive are we at the moment, is there a problem somewhere or is a problem arising?” In the event of a problem, time is a decisive factor: It makes a real difference whether you need four weeks to find the cause of a problem or just one day. “In many cases, it’s all about having the right information at your disposal. How long do I need to get it? This is one of the key factors in increasing production output and being able to reliably produce good quality products. This is an essential aspect, especially for Germany as a hub of industry.”

The researchers have developed a so-called conceptual reference infrastructure for the digital shadow, that is, a framework for the design and implementation of digital shadows. It can be easily customized. The reference infrastructure includes the development of models, processes, and technologies to collect, store, and process data. A key challenge is to develop a common model for very different machines. “That’s an ambitious goal. But if we can manage that, it would be a huge leap forward,” says Brecher. Two institutes are working on this project, using a milling machine and a plastic injection molding machine for testing purposes. Both have different controls, processes, and sensors. First, different models were created for each machine type. “At some point we asked ourselves: Where are the similarities so that we don’t always have to start from scratch? What can be generalized, what can be used in other contexts? Otherwise, the variety of technologies is unmanageable,” says Christian Brecher.

People Continue to Play a Key Role

And then there is the human factor. The Aachen-based researchers involve employees in the process right from the start – unlike in the 1990s, when computer-aided manufacturing was introduced. “Back then, people thought that everything could be automated and no longer paid attention to human beings in the process. That didn’t work out in the end,” says Brecher. For this reason, the technologically oriented research network now also involves experts from psychology and ergonomics, such as Professor Verena Nitsch from the Institute of Industrial Engineering and Ergonomics Nitsch argues that despite digital transformation, people continue to play a decisive role, using the problems of a car manufacturer as an example. The company had introduced automation and standardization in the production process. “Nevertheless, they were surprised to find that the cars produced at their different sites differed in quality. How can that be, since everything has been standardized and automated at both of their production sites? The differences couldn’t be due to the automated part of production.” The developments in the IoP are therefore primarily intended to support people in their decisions in production.

But then she touches upon a sensitive topic: If humans and machines are to be networked, if a virtual image of real production is to be created in real time, then you also need data from people. How far can and will we go in Germany? “China is obviously more proactive in this respect”, she says. In China, camera-based AI systems are used to collect data. But according to the researcher, this poses a dilemma: “We are right to ask ourselves: Is this what we want in Germany?” “We want more data and we need more data. But at what point are we going too far? When do the disadvantages outweigh the benefits?” No one should be monitored at every turn or have to fear negative consequences. But the IoP’s research activities are also having an impact here: Reduced, anonymized data sets should be used to enable the desired communication between companies. According to the researchers, this would solve many problems, for example within the supply chain. The IoP is working on this vision of corporate communication using reduced, anonymized data as part of a World Wide Lab. The aim is also to facilitate the symbiosis between humans and machines.

It’s about recognizing requirements and problems and finding solutions – just like the employees who monitor automated processes and need to intervene quickly if something goes wrong. The Cluster has also dealt with this issue. “We looked at how we can use physiological data, such as the heart rate variability, to predict how attentive a person is.” If attention wanes, errors may be harder to detect. This could lead to production downtime and, in the worst case, even accidents. Employee location data could also provide more information about when people need to be in certain places to ensure that production keeps running smoothly. Obviously, personal rights, co-determination rights, and ethical issues have to be taken into account here as well.

Securing Valuable Know-How

Furthermore, it is essential to secure the valuable know-how of experienced employees who are retiring. A classic example: Typically, a master craftsman is able to see to whether a component is good or bad just by scrutinizing it. Younger colleagues may ask themselves, how are they able to do that? The Cluster is exploring the question of how such expertise can be described and secured using a digital approach – possibly by using data that is analyzed and tagged by artificial intelligence. As Nitsch explains, humans are very good at recognizing patterns: “If machines are trained to recognize these patterns, they can also learn to evaluate these decisions and suggest possible solutions. And the next generation of employees can learn from the machine as well.” But in the end, people will be required to make the final decision.

As a result of automation, industry will need fewer workers in some areas, but these will have to be more qualified, says Nitsch, outlining a number of positive effects: “They earn better money in better jobs, and they pay more taxes, which benefits both the company and the country.”

In the second project phase of the Excellence Strategy from 2026 onwards, which is currently being planned, the “Internet of Sustainable Production” will focus primarily on the aspect of sustainability in production engineering in its various dimensions, seeking to optimize it in terms of economic efficiency, social acceptance, and environmental friendliness. Issues to be addressed include the following, for example: How much service life has a product or machine used up and how much potential does it still have? “We drive a car for about ten to twelve years. After that, it is less interesting for us, but it has the potential to have a much longer useful life,” says Brecher, hinting at future strategies. These, however, require a completely different product design: “It has to be repairable and upgradeable, so that I can replace some parts and say: I can continue to use many components and keep the product up to date by using software updates, for example. You can imagine such an approach with many products”

However, the question of how long the components of a production machine will last has not yet been answered. “The data from a machine does not provide sufficient information in this respect. Statistically speaking, we would need to look at a huge number of drives in order to say that our component might last another thousand or 10,000 hours under the given conditions,” says Brecher. You would have to run thousands of hours of tests on a test bench to get a result. That’s not realistic.

A Question of Risk and Benefit

On the other hand, there are many machines around the world that could provide a data basis for this – if these machines were connected. This would also help to address completely different questions: For example, when comparing processes that lead to better results in one country than in another. We could learn from each other. This is an area that has to be carefully and patiently tackled by the Aachen team. Who would voluntarily disclose their production data? For Brecher, it is a question of risk and benefit: “How do we convince companies to say: The added value of disclosing my data more in a more anonymized form is worth more than the risk.” The fact that the Aachen researchers are taking some of their developments to companies and demonstrating the benefits is certainly helpful.

In addition to a scientific advisory board, the Cluster also has an industrial advisory board with representatives from German and international companies, including BMW, MAN, Bosch, Siemens, and also Phoenix Contact. In addition, the researchers of the Cluster have close ties with industrial associations such as the German Machine Tool Builders’ Association and the German Association of the Automotive Industry. At one of the annual meetings, Brecher was asked when the Internet of Production would be ready for use. His answer illustrates the dimension of this large-scale project: “That’s too big a project for a university and for us to implement. We can’t just pretend to be Google or Amazon,” says Brecher, referring to the large US cloud computing providers.

It is more about doing basic research, providing impetus, and driving technology transfer. Due to their contacts with industry, the Cluster’s researchers have many opportunities to engage in knowledge transfer. For example, they once asked the companies: “Can we check how efficiently you’re using your machines?” Using the tools and methods developed in the Cluster, they identified significant potential for improvement – both with and without the help of AI. This creates trust. Nevertheless, it will take time to usher in the new era of production. According to Christian Brecher, what researchers do is probably 10 to 20 years ahead of their time. Verena Nitsch emphasizes that unexpected developments can also play an important role – for example, working from home arrangements were explored in the 1980s but only implemented during the COVID-19 pandemic.

And Phoenix Contact’s Chief Digital Officer makes it clear how much perseverance is needed on the path to the Internet of Production. “This is a continuous process that will take decades to complete,” explains Possel-Dölken. For the key issues, it takes around five years to see the first results and reap the rewards. And after that, there is always a next step. “There is no point at which you can say, this is now complete.”