Manufacturing companies are under pressure to optimize operations while addressing a growing shortage of skilled labor. The conversation around artificial intelligence (AI) in maintenance makes one thing clear: it’s not “humans versus machines,” but “humans plus machines.”
So what needs to be in place for this human-machine collaboration to succeed? At the SPARETECH Summit 2025, experts from Volkswagen Zwickau, Südzucker, Boehringer Ingelheim, and dankl+partner consulting explored this question. We have summarized the most important findings of the discussion here.
🎙️ The complete discussion can be found here.
AI isn’t new, but its applications in maintenance are evolving rapidly. What began with basic control systems has expanded into advanced analytics and intelligent automation. Today, AI is proving its value in processing large volumes of data and supporting real-time decision-making.
“AI isn’t just creative—it’s analytical. In analytical AI, we now have extremely powerful tools for processing massive datasets. There are already many concrete use cases and projects.”
Dr. Andreas Weber
Senior Partner | dankl+partner
However, the early dream of AI automatically making sense of unstructured data hasn’t materialized. Instead, many companies are starting by improving their data quality: a critical foundation for any successful AI initiative.
“…We decided to first bring our spare parts data from a very poor to a very good state with SPARETECH, and then we’ll definitely invest in AI.”
Andreas Veil
Business Process Owner Operations | Südzucker
One of the biggest challenges in implementing AI in maintenance is data quality. Without high-quality data, it is challenging to utilize AI applications effectively. Companies must therefore first clean and standardize their data before they can harness the actual efficiency of their AI technologies and achieve the desired business outcomes. This requires close collaboration between different departments and a clear data management strategy.
However, before we even discuss data quality, data transparency is crucial. Manufacturing enterprises need to understand the current state of their data to identify areas for improvement. Transparency creates the foundation for effective cross-team collaboration and communication within the company.
“Data transparency is the first step before even thinking about data quality. When it comes to collaboration, transparency, and communication, we often lack a clear status quo: How poor is our data really? Can we even find it? This information is not only important on the shop floor but also for senior management, and this leads us to the question of human-machine interaction.”
Samy Mohamed
Senior Manager Global Maintenance & Reliability | Boehringer Ingelheim
The integration of AI in maintenance requires a mindset shift. It's about combining the strengths of humans and machines to solve problems more efficiently. This partnership can reduce fears and promote the acceptance of new technologies among teams. The collaboration between humans and machines should be seen as an opportunity to increase productivity and compensate for the shortage of skilled workers.
"...and that's how I see AI. It's not human versus machine, but human plus machine against the problem. I think it's simply a perspective that we need to change."
Samy Mohamed
Senior Manager Global Maintenance & Reliability | Boehringer Ingelheim
In the past, maintenance was simpler: technicians would encounter a basic motor that everyone knew, and replacing it was as easy as connecting three phases. Today, the same task involves programming inverters, configuring parameters like ramps, and managing complex systems. Yet expectations around repair time haven’t changed.
“We need significantly more transparency and digitalization to work faster and deliver better information to maintenance workers in the field.”
Stefan Schädlich
Head of Maintenance Assembly | Volkswagen
This quote means making critical data easily accessible, quickly, clearly, and directly at the point of need.
The introduction and further development of AI in maintenance brings both challenges and opportunities. Manufacturing companies must adapt to the pace of change and bring their employees along on this journey. Change management plays a crucial role in promoting the adoption of new technologies and overcoming resistance.
The implementation of AI systems in maintenance is not primarily about reducing staff, but about managing a growing shortage of skilled workers.
“In the next 10 years, approximately 40% of the workforce will retire, a trend that is already evident today, with more 60th than 6th birthdays in the past year.”
Dr. Andreas Weber
Senior Partner | dankl+partner
As experienced employees retire, companies face the risk of losing valuable knowledge and expertise (know-how). Panelists emphasized the urgency of capturing this knowledge and embedding it into digital systems. AI can play a key role here, not by replacing human expertise, but by preserving and scaling it for future generations.
"We must leverage these efficiencies... we're preventing the loss of expertise, we're bundling expertise into digitalization and offering it to someone else who will follow."
Dr. Andreas Weber
Senior Partner | dankl+partner
Introducing AI tools is not just a technical challenge; it’s also a cultural one. Panelists shared that skepticism, especially among experienced staff, is common. Effective change management is about ensuring that a tool becomes part of daily operations after its introduction. This involves assigning dedicated support to teams, following up regularly, and making adjustments based on the feedback received. These efforts help reinforce usage and ensure that new technologies are not only implemented but truly adopted.
"If someone gets frustrated, they get back into their old habits relatively quickly… You can implement the fanciest solution, but if it’s not adopted, it’s worthless.”
Andreas Veil
Business Process Owner Operations | Südzucker
Samy Mohamed shared a compelling vision of the future: a fully integrated maintenance ecosystem where AI not only monitors equipment health but also autonomously triggers spare parts orders and schedules interventions. In this scenario, machines communicate their needs, “my bearing is wearing out”, and the system responds by initiating the right actions at the right time. This vision reflects a shift from reactive to predictive maintenance, with AI as the orchestrator of seamless, data-driven operations.
Stefan Schaedlich emphasized the importance of achieving transparency in inventory management and connecting sites across Europe to unlock shared potential and reduce inefficiencies. His vision is rooted in the belief that digital tools must replace outdated, person-dependent processes to reduce errors and save time. He underscored the need to modernize while ensuring that experienced employees are not left behind.
“In the past, a lot was done via people, which was a good thing because we simply didn’t have the tools to do it and now we have to take the step and we also have to take the older employees with us… switch off the old habits, establish the new tools and perhaps also try to incorporate the knowledge of the old into the new systems.”
Stefan Schaedlich
'Head of Maintenance Assembly | Volkswagen
Forward-looking applications, such as “Speak to your data,” highlighted by Andreas Veil (Südzucker), promise practical solutions, for example, in identifying alternative spare parts or optimizing delivery times. The way to achieve this is by systematically improving the database. With clean and standardized data, AI can help maintenance teams make smarter decisions by querying data directly:
These questions, once buried in complex systems like SAP, will be answered instantly if the data is clean and standardized.
The SPARETECH Summit 2025 made one thing clear: the future of maintenance lies in the synergy between people and technology. AI is already delivering value, mainly when supported by clean data, cross-functional collaboration, and a clear operational purpose. Human expertise remains key, especially as organizations face a generational shift and the risk of losing critical know-how. To succeed, companies must invest not only in technology but also in change management that promotes trust, transparency, and adoption. The journey toward AI-enabled maintenance is not about replacing people; it’s about empowering them to use their time for tasks that AI cannot yet solve.