GF Machining Solutions relies on the symmedia Hub for high-frequency data in its My rConnect portal

Our special guest today is GF Machining Solutions, part of the Georg Fischer Group and a global leader in machines and solutions for tool and mold making. With over 3,000 employees, the company is an expert in milling, EDM, laser micromachining, and additive manufacturing.

Today’s key question: How is GF Machining Solutions positioning itself in the IoT space? Which use cases are they implementing for customers, and what is the business case? What specific solutions are being used?

Andreas Rauch, Head of Digital Business at GF Machining Solutions, will explain this in detail. Also joining us is Andreas Thal, CEO of symmedia, their chosen IoT partner, providing the IoT platform for these use cases. They will share their experiences and best practices for implementation.


I’m excited—so let’s go! Enjoy the episode!



As always, you can find all the details at www.iotusecase.com. Enjoy!

Yes, hello and a warm welcome, Andreas and Andi. Andi, I’ll start with you – how are you today? Where are you joining us from?

Andi

Perfect! This is a first for me – being part of a real IoT podcast. I’m excited for the next half hour with you.

Perfect! And where are you right now?

Andi

You can probably hear it – I’m near Zurich. I’m originally from Switzerland and enjoying the Alps this weekend.

Fantastic! Sending greetings to Switzerland – and of course, to everyone listening in. Andreas, how about you? Where are you? Also in Switzerland, or somewhere else?

Andreas

I’m doing very well. Right now, I’m in Bielefeld, at our headquarters. You might notice my accent – I’m originally from Austria. I moved to Germany in 2005 and am now the Managing Director of symmedia. At the moment, I’m in Bielefeld, but I live in Bavaria.

Perfect! We’re excited to learn more about you and your work. What makes this episode particularly interesting is that we have a user from the mechanical engineering industry with us today. Andi, let’s talk about GF Machining Solutions. You are one of three divisions of the Swiss-based Georg Fischer Group and a global leader in your field. I already mentioned it briefly in the intro – your portfolio is huge.


Can you tell us more about what you do? What typical products do you offer, and who are your customers?

Andi

Of course! You’ve already covered most of it, but I can elaborate a bit. Like many companies in Switzerland, GF Machining Solutions has a long history. We bring together multiple brands: In the EDM sector, that’s AgieCharmilles; in milling, we have Mikron MILL, Liechti, and Step-Tec; and in laser technology, we have Lasertec with Microlution.


We are a classic mechanical engineering company, offering technology-independent solutions for our customers. That’s why our large System 3R division is also part of our portfolio, providing tooling and automation solutions, along with our new digital services, which help us develop innovative business models in technology management and customer service.

Very exciting! And those digital services are exactly what we’ll be discussing today. Before we dive into them, a quick follow-up: Who are your typical customers? Do you serve multiple industries, or do you focus on specific sectors?

Andi

Great question! Since we cover a broad range of technologies—from milling and EDM to 3D printing—it’s essential that our digital solutions work across all products and markets.


Our customers come from a wide variety of industries, including medical technology, aerospace, automotive, energy, packaging, tool and mold making, and ICT. The key is that all our machines speak the same language and offer standardized interfaces—this forms the foundation for our digital services and solutions.

Let’s build on that. You already mentioned digital services—can you walk us through that a bit? Many listeners know you well; you are highly recognized and have a strong market presence with your machines. What is your vision for digitalization and IoT data? What is happening in your market right now, and what future do you see for your company and customers?

Andi

Our market has traditionally been defined by precision, quality, durability, and machine availability—values that our name stands for. With digital services, we are expanding this offering to include fleet management—machine administration—to provide top-tier customer service. Through static or high-frequency live data, we provide our customers with additional insights into our equipment and production facilities. We differentiate between Customer Digital Services, Augmented Machine Capabilities, which provide deeper insights into processes through live data, and process integration and automation, ensuring seamless integration into our customers’ production workflows. These are the three non-hardware solutions we offer.

Very exciting! Andreas, I’ll come back to you shortly to discuss symmedia’s role in this. But before that, I have one more question because I’m particularly interested in the practical side: Andi, you mentioned high-frequency data—can you give us an example? When I think of milling, spindles and various machine data come to mind. What kind of data are we talking about exactly? Can you walk us through your processes?

Andi

In EDM—wire cutting using electrical sparks—we can precisely detect where a spark or discharge cuts the metal, down to the millisecond. We’re talking about gigabytes of data per day that need to be processed.

Incredible! That’s exactly why I asked—working with real-time data brings a whole new set of challenges. We’ll dive deeper into that in a moment. Andreas, what role does symmedia play in this, and what joint project are you presenting on the podcast today?

Andreas

In short, we have undergone a paradigm shift. In the past, every machine manufacturer developed its own service portal, often based on our legacy product. But suddenly, every machine manufacturer had a portal, and customers ended up with multiple portals for each machine type.

Our new product, the symmedia Hub, enables a multi-tenant solution where operators can independently organize their machines and processes and decide at which level they want to collaborate—both horizontally and vertically integrated. The vision is a centralized portal for all machines, regardless of whether you are an operator or a manufacturer.

Exactly, that’s an important point to highlight. symmedia provides an IoT platform as a Software-as-a-Service solution that manufacturers can use. GF Machining Solutions has adopted exactly this solution—as a white-label product for their customers.

Andi, how exactly does that work? So, you purchase the symmedia IoT platform and have developed your own product for your customers from it. What is it called?

Andi

Exactly, our product is called My rConnect. The name comes from our history, as we originally started with remote support and remote control.

What makes it special is that we use the core of the IoT platform—including connectivity, the edge device, secure data transmission, and data storage—from symmedia.

Andreas already mentioned it earlier—as a machine manufacturer, we do not have core competencies in these areas, nor in regulatory topics like the EU Data Act or the Cyber Resilience Act. That’s not our core business.

That’s why we rely on this solution and focus on our strengths in technology management, top-tier service, and remote support.

Perfect! For everyone listening—I’ll include the links to the symmedia Hub and the My rConnect platform in the show notes. Feel free to check them out! I have the platform open right now, and it’s really exciting.

Now, the big question: What was your goal with this solution, and what do your customers actually want to achieve with it?

Andi, you already mentioned some use cases like remote support and fleet management. Can you walk us through this in more detail? Or maybe you and Andreas can complement each other—what’s the ultimate goal for you and your customers?

Andi

The key point is that it’s not about the technology itself—it’s about supporting our customers in the best possible way and solving problems together with them on a daily basis using our equipment.

That’s exactly what the symmedia Hub allows us to do. The platform doesn’t just cover classic remote support and ticketing, but also entire workflows. At the same time, our technology apps help us add value to our machines.
Andreas can certainly elaborate on the technical details, but for us as a machine manufacturer, the most crucial aspect is that we were able to bring our own requirements—and those of our customers—directly into the platform. symmedia has developed a state-of-the-art solution that is open to all OEMs—including competitors—while maintaining a strong focus on our end customers.

Andreas

For us, customer value comes first. The goal is for our customers to maximize efficiency by managing their machines optimally and structuring their service processes effectively.

This leads to cost savings—for example, remote connections reduce the need for on-site service visits. Especially for international customers, these visits can be extremely expensive, and with the right solution, they can often be avoided. At the same time, new business models emerge. When machine sales stagnate or decline, the focus often shifts to the service sector. This is where we can provide real added value, both in terms of cost reduction and efficiency gains. At the same time, the customer remains independent—they have their own look and feel, full control over their data, and they decide how they want to collaborate.

[12:12] Challenges, potentials and status quo – This is what the use case looks like in practice

You mentioned being closer to the customer and creating real value. That was still quite high-level—maybe we can get more specific. What tangible benefits do customers gain? At the end of the day, there’s a two-sided business case: for GF Machining Solutions, transitioning from pure machine sales to digital service business, and for the end customer, gaining real benefits from the collected data. You are true experts, you know your processes inside out, you understand wear limits, and you can optimize upstream and downstream workflows. These insights can be leveraged through data to save time and money, which in turn justifies the investment. Can we talk about specific business cases? Why should a customer invest in connecting their machines? Do you have examples?

Andi

Yes, absolutely—we have concrete examples. As Andreas mentioned, it’s a paradigm shift and a mindset change. Many customers, and the entire industry, are still cautious when it comes to connecting machines. That’s why we start early—at assembly. As soon as a machine passes certain quality gates or processes, we capture the data and can show customers in real-time how their machine is built—while it’s still in production at our factory. Machine acceptance at our facility is now fully digital. The customer receives a complete digital machine validation and immediately has access to all relevant process data. This means they can check at any time whether their machine is still within tolerances—at the start of production, after a year, or after a maintenance cycle. Our goal is to demonstrate the digital added value from the very beginning, build trust in data security, and support customers digitally every day in using their machines.

Andreas

Once the machine is initially connected, the customer can get it up and running faster. The key benefits include shorter setup times, optimized processes, and more efficient calibration. If this process can be guided remotely, it saves a massive amount of time. It’s not just about cost savings—it’s also about the shortage of skilled workers. Experts are in short supply, and remote access allows them to be deployed more efficiently. Additionally, travel times are eliminated, which reduces the CO₂ footprint and lowers process costs. For our customers, this is essential because a fast time-to-market is more important today than ever.

Andi, you already mentioned some data earlier. I’m not sure how deep we can go into this, but I’m really interested in the practical application on your machines. Maybe you can make this more tangible – you mentioned high-frequency data, spindles, vibration, and noise data. Is there a real use case that demonstrates this? 

Andi

Yes, of course! I prefer to stick with milling machines and spindles because a typical milling machine consists of many components. There’s the machine’s 3D workspace, the spindle, the tools, and various auxiliary systems like cooling or consumables. Until now, each component manufacturer has monitored and recorded their own data separately, but our approach is to create a digital twin of the entire machine, where all components are optimally coordinated.
For the customer, this enables predictive maintenance, meaning maintenance intervals can be predicted to avoid production downtimes. A concrete example is final finishing, where the machine must be calibrated to the last micrometer—about a third of the thickness of a human hair. By capturing all relevant machine data in real-time—such as a production temperature of 21.5 degrees or a coolant temperature of 15 degrees—we can perfectly calibrate the machine for the next 10 to 15 minutes and ensure the required quality. Our customers are increasingly recognizing the value of these digital representations, which must always be available live. They become part of the customer’s continuous knowledge base, which evolves over time.

Andreas

From a technical perspective, IoT connectivity is achieved through modules that can read specific data formats. While MQTT is still common, it’s somewhat outdated—OPC UA is currently the primary standard. In this format, the machine can securely transmit the collected data. The data is encrypted during transmission over the last mile and stored in the customer’s tenant, ensuring data ownership remains fully with the customer. They retain full control over their machines and the associated data.

[18:32] Solutions, offerings and services – A look at the technologies used

So, every machine has a connectivity option for collecting data, depending on the use case and the customer’s needs. You have the ability to securely process this data on the symmedia side and selectively pass it on where it’s needed.

Andreas

Exactly. We not only have native support, but also our own OPC UA adapters and the option of connecting third-party modules to the machine. If a manufacturer requires a specific data format, we can retrieve the data via a module while ensuring secure transmission to the data storage at all times. This is essential. Additionally, we collaborate with companies specializing in Brownfield integration and older data formats. Through our application management on the IoT device, we can integrate almost any data adapter into the machine and securely capture the data.

Andi

Sorry to jump in for a moment. Madeleine, you mentioned that symmedia has access to the data—but no, symmedia does not have access to the data. That’s an important distinction. You can compare it to permissions on a smartphone: an app like WhatsApp can request access to the microphone or camera, but the user decides whether to grant those permissions. That’s exactly how the symmedia Hub works. The end customer determines which data they want to share and whether they allow remote support access. symmedia itself has no knowledge of what is being produced or which data is shared with whom. This is essential, and our customers need to be able to trust that they are always in control of what they share and what they don’t.

Thank you for clarifying that. My wording was a bit misleading. The key point is that the customer makes the decision, and the technical architecture ensures that there is a clear distinction in how and where data is stored or shared.

Andi

Exactly. A great example is the aerospace industry. In this sector, customers must certify and qualify the entire process. Licenses and intellectual property are often involved as well. If it’s not clearly regulated who has access to which data, and if there’s no seamless traceability, companies in this industry can quickly find themselves out of business.

Yes, absolutely! You already mentioned at the beginning that you deliberately decided not to develop this solution in-house, but instead chose to source it externally. Andi, can you explain again why you made this decision as a machine manufacturer? Many companies in our network develop these solutions internally because they have their own software developers. Why did you decide to go with symmedia?

Andi

We can be very open about this—it’s simply not our core competency. In the past, we consisted of various brands that pursued different approaches, and we learned from that. The complexity of such developments is enormous, and it requires true experts. Our strength lies in the industrialization environment—in other words, how to connect an existing machine to an IoT platform without physical modifications. This is exactly why we pursued a co-development approach with symmedia. However, technological expertise and software development are not our core business.

Many listeners may already have their own IoT platform or are considering building something similar. Can we summarize your solution once again?


From what I’ve understood so far, the symmedia Hub is not just an IoT platform, but also a full ecosystem that enables various service use cases. Additionally, you have the option to offer the solution as a white-label product, allowing manufacturers to use it as their own portal for their customers. Is that the correct high-level summary?

Andreas

Exactly, that sums it up well. Security is the foundation—cyber resilience, compliance, and IEC certifications ensure trust and protection. We have two key areas.

First, there’s the ability to offer service solutions even without an IoT connection to the machine. Customers can bulk-upload all previously sold machines into the system and immediately access services like ticket management, asset management, or document management. Additionally, external systems such as Salesforce, INOSOFT, Jira, or SAP can be integrated, even if the machines themselves are not yet networked.
However, once a direct machine connection is established, entirely new possibilities open up. With WebVNC and Datalink, we can deeply integrate at the IP-broadcast level, interact with machine control in real-time, and even program it using a high-security mechanism. Additionally, Raw Data APIs and OPC UA interfaces allow us to extract data, while the Application Management system enables the deployment of third-party and proprietary applications directly on the machine. This is the ecosystem framework we operate within.

Andi

Maybe two additional points. The ability to provide custom applications via Docker containers – for us, this is the technology of the future. Our goal is not to market pure connectivity as a competitive advantage, but rather to focus on delivering applications and real value for the customer. Even within an open platform, we can leverage our core competencies through our infrastructure and provide them directly to the customer.

Yes, that’s an important point—so essentially, like an app store, where you can offer your applications while optimizing the infrastructure.

Andi

Exactly, because the USP is no longer just in the hardware, but rather in the know-how and integration at the end customer’s site. That’s one key aspect. The other, which Andreas almost casually mentioned earlier, is ERP integration—for example, with Oracle or SAP. One of the most important insights for us was that internal organizations need to be properly addressed and interconnected.
Internally, we use SAP for our ticket system, asset management, field service management, and spare parts logistics. At the same time, our portal serves as a single point of truth for the customer. So, integration is not just about the customer side, but also about optimizing our own internal processes. Only by doing this can we effectively use these technologies, demonstrate them to customers, and ultimately market them with new knowledge models.

That’s an important point you just highlighted. On one hand, there are the live IoT data you mentioned earlier, and on the other hand, ERP and IT systems that need to be connected to execute the use cases. After all, you need to know which customer is using which machine, where it is located, its history, and other key details. This really shows how crucial integration is and the role that symmedia plays in this process.
Great! Maybe one last question about the business case. Many companies are currently dealing with this transformation—some for a few years now—but it remains a new era that is still evolving.

Andi, you are establishing a completely new business model. How do you, as a traditional machine manufacturer, handle this shift, especially considering that your business has historically been CapEx-driven? What challenges does this transformation create within your organization? It’s a huge shift, after all.

Andi

Exactly, the shift from CapEx to OpEx is a real mindset change. While CapEx involves a one-time sale, contract closure, and often a completed transaction, OpEx requires a completely different approach—both internally and on the customer side. Different stakeholders are often involved, making seamless integration a challenge. The key is to clearly demonstrate the value to the customer.

Yes, and you have described this added value very well. We already had an example of this, and that’s exactly what helps your customers move forward. It’s also interesting to see your strategy of becoming even more software-driven in the future, not just developing applications but also providing them through the ecosystem.

[28:14] Transferability, scaling and next steps – Here’s how you can use this use case

Many companies are heading in this direction and are looking for best practices or common pitfalls. Based on your experience, are there things you would say companies should be aware of? Were there moments where you lost time due to additional iterations? Do you have any insights to share?

Andreas

One crucial factor is the network infrastructure at the customer’s site. When we talk about IoT connectivity, it is often very heterogeneous and varies from customer to customer. That’s why it’s important to establish the initial connection as early as possible together with the customer. Once this foundation is in place, it’s just a matter of further developing the solution. The biggest effort usually occurs at the beginning due to different infrastructures and network conditions. Addressing this early can save a lot of time and prevent unnecessary iterations.

Andi, at the beginning, you mentioned that your machines speak the same language—we’ve already discussed that. Another key challenge is working with data. Are there any lessons learned from your data-driven approach? What insights have you gained?

Andi

We are heavily engineering-driven and initially realized that we didn’t always know when certain data would become valuable. At first, we tried to build a static data model, defining in advance which data points were important. However, with today’s capabilities for storing raw data, we’ve learned that value often emerges over time—for example, when AI later detects patterns or discovers new correlations.
We underestimated this dynamic. The future clearly lies in technologies that enable such insights. Everyone is talking about artificial intelligence, but the real question is how to generate tangible value from it for ourselves and our customers. How can we derive know-how from data? Why does Machine A perform differently than Machine B, even though they are identical? These insights often become visible only in hindsight. The key is to continuously learn, share this knowledge with customers and suppliers, and develop the best possible solutions over time.

Would you say that applications like spindle noise analysis are the future AI applications you will market?

Andi

Absolutely! This is not a trade secret—the development is clearly moving in this direction. Prediction and closed-loop systems will play an increasingly important role. In the future, machines will be able to self-correct, whereas manual calibrations are still required today. This is exactly where the next AI-driven applications will emerge.

And Andreas, is this something you actively drive and distribute as part of the symmedia ecosystem, or do you have dedicated modules for this? How does that work exactly?

Andreas

We are integrating a native AI module to optimize the Hub’s usability. For example, when a document or manual is needed during a service case, it will work similarly to ChatGPT, providing fast and targeted information retrieval, even for root cause analyses. This will be a built-in feature.


At the same time, we support external developments. We are collaborating with Ruhr University Bochum on a federated learning project, where machine models are trained directly on the machine. Instead of exchanging raw data, only the optimized models are sent back. This is still relatively new in the market, but it is definitely a field we are investing in. AI is a major focus for us.

Yes, absolutely—this is currently a huge topic in the industry. It’s particularly exciting when these developments find practical applications. It’s great to see that you’re not just relying on existing solutions but also developing your own native features for the symmedia Hub.


For now, thank you both for the insightful discussion about the project, its challenges, and its implementation! If someone is listening and wants to implement something similar, would it be okay if I include your LinkedIn profiles in the show notes? Or what’s the best way to reach you?

Andi

Absolutely, that would be great!

Andreas

Yes, LinkedIn is a great channel.

Perfect! Are there any trade fairs this year where people can meet you in person?

Andreas

Yes, we will be at all about automation in Hamburg at the beginning of June. For those who can’t make it there, you can also visit us in Düsseldorf, where we have a full setup with references, technologies, and everything you’d want to see. We welcome everyone who wants to stop by!

Sounds great!

Andi

And of course, at EMO – that’s our next big trade fair.

That’s right, EMO is also a key event! We’ll definitely stop by. For anyone listening, feel free to visit us at your favorite trade show and experience everything live.


At this point, a huge thank you to both of you! I think it’s now very clear what kind of data we’re talking about, which use cases are possible, and how the business case works both for your customers and for GF Machining Solutions. And thank you, Andreas, for sharing insights into how everything works with symmedia. A truly exciting project—I’m really curious to see how it evolves over the next few years. Maybe we’ll have another conversation next year for an update. Thanks again for joining us and for this great discussion!

Andreas

Thank you, Madeleine, for the opportunity to present our project here. A big thank you to GFMS as well—through close co-creation, we’ve reached where we are today. Something like this only works with a strong team and great collaboration.

Andi

A big thank you from my side as well, this was a lot of fun! I hope we can do a follow-up soon and share the next insights with you.

Thank you both! Wishing you a great rest of the week. Take care, bye!

Andi

Thanks, bye!

Andreas

Thanks, bye!