In an interview with EFESO, Klaus Buchwald, Chief Operating Officer of Siltronic AG, explains what companies should focus on when it comes to digitalization, automation, and the use of AI. Key topics include a solid data foundation, scaling AI applications, and unlocking new potential in design, production, and cross-company value chains. 

What should companies in your industry focus on when it comes to digitalization and automation? Are there, for example, special strategic or operational potential that should be addressed more strongly?  

Klaus Buchwald: In the semiconductor industry, we have been working very intensively with machine learning technologies – for 20 years along the entire value chain, from development to operations. But I see much more potential, especially if we go deeper into the early design phases and look at classic business processes, as is the case in other companies.

We are increasingly moving from classic machine learning to generative AI. However, one key requirement remains unchanged: the ability to access “clean” data. Without solid data preparation and clear data structures, AI use cases cannot be scaled quickly enough. My experience clearly shows that if the database is right, we can roll out new applications much faster and create real added value.

The majority of INDUSTRIE 4.0 AWARD applicants improve value creation processes in a targeted manner with AI, but with significant differences in terms of focus and results.

How do you classify the current “AI impact” in manufacturing companies? Where do you see still unused fields of action?  

Klaus Buchwald:In fact, AI opens up enormous development prospects for the industry. However, the challenge at the moment is to bring AI technologies into practice step by step. I keep seeing exciting examples in the semiconductor industry in particular. This is often about individualizing designs. You have an existing chip design, and AI combines various sub-processes in the shortest possible time to generate a completely new, customer-specific solution, and in some cases even validate it.

For me, however, the semiconductor world encompasses more than just the chip itself. As soon as packaging comes into play, further variants are created – and this is precisely where AI can bring enormous speed and efficiency. Companies can develop, test and deploy new designs much faster. It also gets exciting when we think beyond company boundaries. With AI – based on anonymized data, of course – we can mitigate the classic bullwhip effect. It can probably never be completely eliminated, but AI can help to exchange information earlier, more objectively and more transparently.

This in turn has a direct impact on the resilience of value chains. After all, resilience doesn’t just mean absorbing partner failures. It also means being able to react flexibly to strong fluctuations in demand – both upwards and downwards. AI helps companies to make decisions faster, identify potential partners at an early stage and avoid having to search for alternatives in an emergency. 

“When the data foundation is right, we can roll out new applications much faster and create real added value.”

In addition to AI, do you see another trend – technological or arising from the framework conditions of industrial transformation – that will significantly influence the importance of Industry 4.0?

Klaus Buchwald: Humanoid robots are becoming increasingly relevant for industry. Performance and costs are developing so quickly that we need to take a serious look at them. In the semiconductor industry in particular, where I come from, we have made enormous progress over the years through high automation – and it remains important. Nevertheless, there are many production areas in which classic automation reaches its limits. This is often due to the buildings themselves: The structures do not allow for an overhead transport system, either because the ceiling height or the structural capacity is insufficient, or because transport processes are infrequent but involve heavy loads over long distances. This is precisely where I see great potential for humanoid robots.

In the case of semiconductor production, for example, many factory buildings are simply not designed for the installation of a modern transportation system. At the same time, there are numerous transportation tasks that have nothing to do with direct product processing. Humanoid robots are ideally suited for this, they can close gaps in workflows, speed up processes and relieve employees – especially where classic automation reaches its limits. We should therefore address this issue much more intensively.

We should also not forget that Industry 4.0 success requires more than just technological mastery. We see this in Chinese companies, which are more agile in terms of organization, make decisions faster and implement them more consistently. It is precisely in this respect that we must not fall into a trap: We cannot solve structural or cultural inertia simply by using new technologies. If we are sluggish, even the best algorithm won’t help.

That’s why we should think much more about the cultural and organizational aspects of Industry 4.0 – which is very technical in name. Technology is important, no question about it. In global competition, there is no way around AI. But how we work is just as important: how quickly we make decisions, how flexible we are, how we organize collaboration. 

The topic of “ROI of Industry 4.0 use cases / initiatives” is very present among the award winners. In your experience, what characterizes a successful prioritization of investments in the area of conflict between short-term efficiency increases and long-term innovative ability? 

Klaus Buchwald: We need to deliver efficiency gains in the short term while maintaining the staying power for strategic, long-term developments. The two belong together. Short-term results are important to show what is possible – especially to the workforce. Visible success creates acceptance and motivation.

At the same time, we must not make the mistake of focusing solely on quick results. Innovation also means keeping at it, sharpening up and constantly renewing yourself. That is why we need both: quick, concrete improvements in the present and consistent work on the issues that will shape our future. This is the only way to achieve sustainable progress. 

Learn more about the INDUSTRIE 4.0 AWARD

Visit the INDUSTRIE 4.0 AWARD website