Food Production Saves Energy with autinityE3

Hello, dear friends of IoT. Today we’re talking about a use case many of you will be familiar with: energy monitoring. Rising costs and a lack of transparency often make it difficult to make the right decisions. Today, we’re looking at a practical example from the food industry. We want to find out how Kohrener Landmolkerei approaches this topic. Together with their IoT partner, Schaeffler Digital Solutions, they explain how they built a hands-on energy management system using the autinityE3 software.
For context: Schaeffler Digital Solutions is the digitalization division of the Schaeffler Group, operating independently on the market with headquarters in Chemnitz and specializing in software solutions for the manufacturing industry. Since 2025, the company has also expanded internationally, with a location in India among others.
Joining me today are Raphael Elm from Kohrener Landmolkerei, as well as Maximilian Adam and Adrian Büttner from Schaeffler Digital Solutions. They provide insights into the project, explain how it was implemented, and share best practices. We’ll also cover community questions: How do you get data from legacy machines? Which data is actually relevant? And how do you identify the biggest consumers?
As always, you’ll find all information on implementing this and similar projects at www.iotusecase.com and in the show notes. So with that, let’s get started – enjoy the episode!

Hello everyone, hello Raphael, Max, and Adrian. Let me start with you, Raphael. How are you today, and where are you right now?

Raphael

Hello, I’m doing well. I just came back from vacation and had my first week back at work, which has been quite eventful with everything piling up after two weeks. Today, I’m a guest at Schaeffler and we’re recording the podcast together here.

So you’re all sitting in the same building today.

Raphael

Exactly, it’s a good chance to get to know each other.

And a quick question about the dairy itself – where is your headquarters located?

Raphael

The dairy is based in Penig, a small town near Chemnitz.

Very nice. I just pulled it up on Google Maps – when you zoom in, you can see the site from above.
Really interesting. And normally, you’re based there on site with your colleagues, right?

Raphael

Exactly, I’m there every day, managing the office IT and also handling smaller projects in production. That’s how the project with Schaeffler came about, so we could take a closer look at our energy management.

We’ll get to that in more detail in a moment – perfect segue. Let’s go around the table again. Adrian, how are you doing and where are you right now? You’re probably not all in the same room.

Adrian

I’m joining remotely today, but I’m often in Chemnitz, where the product development teams are based and where development takes place. Since we work closely with Schaeffler, I’m also frequently in Herzogenaurach at headquarters. That’s where we gather customer requirements, shape the product together, and get direct feedback.

Perfect. And then last but not least, Max, how are you today?

Maximilian

I’m doing great. I just got back from a business trip and I’m back in the office catching up on the past few days. Together with Raphael, I already looked at what’s coming up next for us. We had lunch together and are now starting this recording relaxed and in good spirits.

Perfect. We’ll get to what exactly autinityE3 is and how it’s used in just a moment. But first, Raphael, a quick question about you personally. On LinkedIn you’re still listed as ERP Key User, but you said you’re now in IT. I imagine that means you’re in something like an IT project lead role at Kohrener Landmolkerei, where you also initiated and implemented the autinity project on the user side. Is that correct?

Raphael

Exactly.

Okay. So I assume you also have a background in IT?

Raphael

Yes, I studied computer science and got into PCs early on through my family’s business. I learned what a network is, how to manage it, and what possibilities there are. After my studies, I found my way into Kohrener Landmolkerei in the ERP systems area. There I was able to support projects in addition to IT administration. Among them was energy management, which was a natural fit because of its technical focus. In that role, I act as project manager, helping to implement and build such projects and to generate insights.

Very interesting. Maybe a quick look at your company itself. I would categorize you in the food production sector. Your focus is raw milk processing. You probably have equipment for cooling, steam boilers, and so on. Can you tell us a little about what production looks like for you and also about your industry? Who are your customers?

Raphael

Exactly, the name already says it. We are a dairy. We process raw milk and refine it, producing regular UHT milk you can buy in the supermarket. We also produce flavored milk drinks, meal replacement drinks, and high-protein pudding – basically everything you can make from milk. Cheese is not part of our portfolio, but beverages and pudding are our core products. In recent years, the vegan segment has grown strongly, and we’ve found and developed a niche there as well. We now produce plant-based milk alternatives.

Okay, so you have equipment for raw milk processing and additional lines for plant-based products, such as oat milk.

Raphael

Exactly. In milk processing, many machines are involved that consume electricity and also require steam. The product is heated, cooled down again, adjusted to a specific fat content, and then processed in the filling machine. There too, steam is used to ensure the product is sterile and aseptic. That requires a lot of energy, but at the same time it holds great potential for optimizing processes.

When you talk about energy and say there’s a lot of potential, can you share how you even got started with energy management in the first place? Since when have you been doing this and how did the collaboration with Schaeffler Digital Solutions come about?

Raphael

Energy, of course, affects every company, and costs are constantly rising. So we thought about where savings could be achieved. That was the basic idea – to take a closer look at energy consumption in the first place. Before that, all we had was the electricity bill from the provider, a single number at the end of the month. That wasn’t enough for us. Through the Chamber of Industry and Commerce, we came across Schaeffler, who offer a system that allows you to break down that number and identify savings potential.

Okay, maybe let’s talk briefly about Schaeffler, or more specifically Schaeffler Digital Solutions, before we dive into those potentials. Adrian, you are Product Owner at Schaeffler Digital Solutions. Can you explain which IT product you are responsible for here and in which area it fits?

Adrian

That’s right, I’m responsible for the product autinityE3. At Schaeffler Digital Solutions, this is the energy management system. We also have other products, which have already been mentioned in other podcast episodes, such as autinityDAP. With autinityE3, we build on our core expertise of data acquisition, but we specifically picked the use case of energy management and developed it into a standalone product.

Okay. And Max, you’re a Sales Consultant at your company. So you look after this project, and probably many others too. Is that a good way to describe your role?

Maximilian

Yes, exactly. It’s the classic mix of sales and consulting. Getting to know companies, understanding their requirements, and then identifying the pain points. Where can we step in, where can we use the system to address that exact pain – like Raphael mentioned, the high costs – but also to first of all create transparency.

What would you say is special about this joint project, for you personally or in terms of supporting this customer? Is it a typical energy monitoring project, or something particularly exciting? How would you categorize it?

Maximilian

Both, actually. It’s classic in the sense that many companies in Germany, especially medium-sized ones, don’t really know their energy consumption and lack transparency on where the levers are in their operations. That’s exactly where we want to start with our system. What makes this project particularly exciting for me is that Kohrener Landmolkerei was our very first external customer for this product. Previously, it had only been used internally at various Schaeffler plants, where we already knew what to expect. Together with Adrian, we decided that the product had reached a certain level of market maturity, and we were ready to launch it externally. That’s how the contact came about. This makes the project especially interesting because we couldn’t rely on existing internal data but really started on a greenfield.

[09:18] Challenges, potentials and status quo – This is what the use case looks like in practice

What I’ve understood so far is that this is about production, specifically energy management. You have networked various systems. I heard steam mentioned as a data point, so probably a steam boiler or other equipment that’s integrated. Is that correct, or can you tell us more about the scope of the project where you investigated the energy-saving potential?

Raphael

Exactly, using the plural was a bit generous. We have a central steam system that generates steam to heat the milk and then cool it down again. But that’s only one aspect. We have many different machines that require electricity, water, steam, or compressed air. This combination of different machines and media was especially interesting. Later, we gained many specific insights at the level of individual machines.

Very interesting. Let’s then talk about the business challenge, the business case. At the end of the day, this is an investment in technology, not just for its own sake. As you mentioned, the focus is on identifying where the potentials are. Did you look from the beginning at what data was already available and what potentials could be derived from it? Was there some kind of business case calculation, or how did you approach it?
Which challenge were you addressing?

Maximilian

We approach this in a structured way and look not only at energy but also at other resources such as steam, heat, or compressed air. Energy is often the entry point because it’s tangible for everyone, even at home. Together with the customer, we look at total consumption and identify potential savings. Every company has levers, sometimes small, but always there. Then we examine which are the main consumers in the plant. Which systems, for example at the dairy, are running around the clock. Those are the ones we make transparent in order to derive measures for savings. In the project with the dairy, we followed the classic project sequence. We looked at the equipment and first analyzed the energy consumption, since energy meters are often already built into the control systems. That way, we could tap into data directly. Once that worked, we moved to the next step with the customer and asked what was most important to them. Is it a process with high steam or water consumption, or is it about compressed air, where you might detect leaks early on and achieve savings.

Okay, and Raphael, would you confirm from practice what Max just said? Or would you add something?

Raphael

By and large, I can only confirm it. People always think first of electricity, but the issue is much more complex. For us, steam and water are very big topics. With the system we are now using, it was relatively easy to integrate these resources. That was important for us because we had the data, but no one had actually used it before.

To unlock the potential, you first need access to the data. Did you face specific technical challenges during implementation? In our community, we often hear about very old machines without Siemens S7 controllers, or equipment where people would like to measure consumption but no data collection exists yet. What were the technical challenges in your project?

Raphael

One surprise was that there were already measurement devices in place, namely Siemens SENTRON PAC. These devices can measure electricity and energy and were already installed in some machines. With other machines, we had to start from scratch because they had no network connection at all. A machine was sitting in the corner doing its job, but with no connectivity. Another challenge was meters that existed but were so old they were no longer compatible with current systems. These had to be replaced because they could no longer be integrated into a controller.

And Max, Adrian, you see many such projects. Would you say these are typical challenges, or are there other areas where companies lose a lot of time today because they would otherwise have to connect data manually? How do you see it?

Adrian

These are typical hurdles and requirements we see in many projects. Even at Schaeffler, the issue of heterogeneous connections was always present. This is exactly where our strength at Schaeffler Digital Solutions lies. We support various query protocols to access the data. Energy management is just one area, but even here there are many meters and protocols that can be integrated. As Raphael mentioned, for example the Siemens PAC meters. These are standard query protocols for us, easy to integrate. We also provide standardization with device templates so that users basically only need to add necessary variables such as the IP address. Everything else, such as where the power or energy consumption is stored, is already preconfigured, so the data can be captured quickly and easily.

Before we get into the details of the implementation, one more follow-up. Raphael, you mentioned ERP is a topic for you. Were IT system data also relevant for you, or did you focus purely on OT data such as from steam boilers and equipment?

Raphael

That hasn’t been a focus so far, but it will come. The question is: when I manufacture a product, what does it actually cost me, and where can I find savings? With the new version of E3, we can map that, so we can see exactly what a product costs on a production line. Up until now, that wasn’t a priority for us, but that will be the next step.

Okay. I can imagine you’ve seen other projects where this is already being done. You also have a huge amount of experience yourselves, since you’ve connected your own plants and solved the challenges there. Every project is a bit different, and in the end, it depends on the specific use case which data is needed.

Maximilian

Yes, definitely. There are always two challenges. First: do I already have the data, and where does it come from? That’s where the whole metering infrastructure comes in, everything we’ve just described. Second: what do I actually do with the data once I have it? In classic energy monitoring, the first step is creating transparency. The next step is moving into management. As Raphael already hinted, we can also include production data—seeing whether a machine was running or idle, which product was being produced, and how much energy was consumed per piece. That’s where it gets really exciting, not just for energy management but for every level of the company. You can identify optimization potential, achieve savings, and at the same time improve processes and process chains to maximize value creation.

[17:07] Solutions, offerings and services – A look at the technologies used

Let’s start with data acquisition. You said the data is collected from various devices. How did you practically go about connecting legacy machines or the steam boiler to the system? How does it work, and which of your products are used in the process?

Maximilian

At the dairy, we examined the setup closely and, through our partner network, brought in a partner who went onsite to inspect the equipment. They checked what was already in place, what the control system could do, and where we could intervene. Often we find energy meters already installed but not being used— these can be integrated directly. At the same time, the partner works with us to analyze which additional measurement points make sense and how the implementation should look. That’s exactly how we approached it with you at the dairy, Raphael.

Raphael

Exactly. For example, we had a machine with a meter installed that was also capturing a downstream machine. In that case, it made sense to add a second meter to differentiate between the two. Both machines were running at the same time—but do they really need to? In our case, it was a compressor generating compressed air and the upstream machine that consumed that air. Both were connected to a single meter. By splitting them and looking at them separately, you can detect potential compressed air leaks. Specifically, the compressor was running 24 hours while the upstream machine was only running 23. Such differences help uncover problems.

Were there any other “aha moments” where you realized it made sense to look more closely because certain systems could be controlled more precisely or energy consumption measured differently?

Raphael

Not directly as a savings potential, but we noticed differences between processing raw milk and plant-based products. The plant-based products are usually water-based with oats or soy as the base. We found that heating and cooling these products requires less energy than raw milk. At first, we didn’t realize this—we assumed that since the machine runs anyway, the energy use would be the same. But we learned that the differences between the products themselves do have an impact on energy consumption.

That’s really interesting, being able to analyze such data for the first time and uncover new potential. The next step was evaluation. You’ve already mentioned the E3 software. Max, could you describe from your perspective how autinityE3 works, what kind of software it is, what it can do, and how it solves these kinds of use cases?

Maximilian

autinityE3 comes in two software packages: autinityE3 Starter and autinityE3 Enterprise. At the dairy, we began with the Starter pacckage. This package was developed for companies that have never worked with their energy data before. It allows you to record pure energy data, with up to 50 measurement points. That way, you can set up initial energy monitoring, add a simple measures management function, and then start working with the data. The next step is autinityE3 Enterprise, designed for larger companies and corporations like Schaeffler. The focus there is on linking multiple sites, creating comparability, and integrating everything into one system.

So internally, Schaeffler is like an internal customer for you, is that right?

Maximilian

Exactly, Schaeffler is an internal customer, but we also provide the solution to external companies. In those cases, we often go deeper into production analysis. That means integrating production data, for example through a BDE or MES system. This way we can see if a machine was idle or running, what job data was recorded, and what material was being processed. At the dairy, it’s mainly milk and water. In industries like automotive or steel, things are much more complex, with different materials and material thicknesses. That has a direct impact on energy consumption and how it can be controlled. We had already discussed this with Raphael—when you process water, the energy data looks completely different from milk. Measures that work for one material can’t be transferred one-to-one to another. With the software, we can validate and test that.

Very nice. If you’re listening right now and think you have a similar use case or are working on a comparable project, I’ll include the contact details of the three guests in the show notes. If you’d like, you can reach out for an exchange of experiences.
I think that’s valuable, because every project is a little different. You’ll also find the links to autinityE3 in the show notes, so feel free to check them out. One more question about the evaluation of the data: How do you handle this from an architectural perspective? There’s autinityE3 as the software on the application level, and then there’s autinityDAP, which already had one or two podcast episodes. Did you also use it in this project, or was that not relevant here?

Adrian

It is used. That’s essentially the foundation. When we deploy autinityE3 Starter at the dairy, it runs on an IPC with autinityDAP to enable data collection from the energy meters. In parallel, the autinityE3 software handles the visualization. Both are delivered as a standalone solution so that the customer can immediately begin with data collection and visualization.

So autinityDAP acts as the data collection layer from OT to IT, and on the IT level your E3 software runs, where the actual evaluation happens. Very nice. Maybe one last question, Raphael: What has been the result for you so far? Where do you stand right now, and what’s next? Can you share how you currently use the product in your daily operations?

Raphael

We’re still in the setup phase. The Starter is installed, and we’ve already configured some measurement points. But you quickly realize you want more measurement points to get a closer look. That’s why we’re currently adding more measurement points and integrating additional energy carriers. For us, wastewater is also important. Here, we want to measure the pH value and temperature, and that will be added. For us, it’s important for Controlling, to keep an eye on the numbers and potential savings. But our QM department also uses the data, for example to monitor wastewater values and make adjustments if necessary. So the tool is not just relevant for Controlling but also for different departments, giving them visibility into what’s happening in the system at any given time.

So that means every user has different roles and therefore needs different data. That depends heavily on the use case. One area might be wastewater, like you said, another could be something entirely different. Would you say these are like small subprojects for you, each with its own views for the users?

Raphael

Exactly. Each department needs its own graph and its own dashboard, depending on what’s important for them. Some want to see specific numbers, while others are fine with just a temperature trend or another curve. With E3, that can be mapped very well, even for different departments at the same time.

It’s also interesting to see that you started small and are now moving step by step. With the data, you’re gaining insights that prompt you to look more closely. I think that’s a classic project trajectory.

Raphael

Exactly. Other departments also have their own requirements, such as monitoring wastewater pH values. That came up later, when the QM/QS department said this would be important as well. I don’t think this will be the last measurement point we add. The project will continue for quite a while and will definitely grow larger.

That sounds like a good point for an update. Maybe we’ll talk again in a year and see how things have developed. Adrian, a follow-up question about the product itself: Do you use specific KPIs to prioritize the data, drawing on your experience or product knowledge? Or is it more of a case-by-case approach, where you check each time which processes require which data? How exactly does your data analysis work?

Adrian

Based on our experience at Schaeffler, we always start with the energy data for the different media and build from there. That was the case internally at Schaeffler, and it’s now also what Raphael and the dairy have recognized. When you dig deeper, you realize that consumption is heavily dependent on production. From our experience across Schaeffler plants, it became clear that for a solid plausibility check or analysis, you need more than just energy and consumption data. That’s how the idea came up to include production data as well, to enable more meaningful evaluation and analysis. That’s the foundation of our approach. With autinityE3 Enterprise, we can connect additional systems with production data, which significantly improves the analysis. This forms the basis on which future functions will also build.

[27:23] Transferability, scaling and next steps – Here’s how you can use this use case

The project is evolving, and so are your products at Schaeffler Digital Solutions. Can you give us an outlook on how things will develop over the next few years? Raphael, maybe you can start, and then of course Max and Adrian. If you look into the crystal ball, what’s your forecast for the next three to four years?

Adrian

I’ll start, and Max can add to it. As I mentioned earlier, we’ve now integrated multiple types of data into the energy management system. The feedback we quickly got from technologists was: you already have all the data in the system, can’t you just tell us directly where higher consumption or waste occurs. That’s exactly what we’re addressing now with an initial AI feature. The system automatically analyzes the energy and production data and gives users specific hints, for example when there’s higher consumption during a weekend production run. This way, users get concrete starting points for implementing measures to save energy. Developing this kind of autonomous analysis is what we’re currently focusing on intensively.

Maximilian

And we take these steps together with our customers. We listen to how the dairy, but also other customers like Schaeffler, envision the long-term implementation. Which features are needed, where the biggest pain points lie. One topic always comes up: peak loads. Every peak load causes enormous costs, depending on the size of the company between €25,000 and €70,000, without creating any value. Das tut jedem Unternehmen weh. That’s why we want to focus more on energy storage and its control in the future. How can storage be used efficiently, how can we actively control cooling units or other systems. In the coming years, development will move toward active peak load management, where the system supports the users. With AI features like anomaly detection, we are laying the foundation. Once that’s established, many further opportunities will open up.

That’s a great direction and a huge topic. Maybe we can do another episode in the future and see how the product develops, because there’s definitely more to come. From my side, thank you very much for your time. I’ll give you the final word. You can just close the session with a goodbye or add something if you like. I wish you every success for the next steps and I’m excited to see how things evolve.

Adrian

Thank you, we’re also looking forward to continuing our collaboration with the dairy and with many other customers. That really helps us develop the product in the right direction.

Raphael

I can only agree. Thank you very much for the invitation. I don’t think this will be the last collaboration with Schaeffler. I’m sure there will be one or two more projects we can pursue together.

Maximilian

That’s always nice to hear. From my side as well, thank you for the time and the pleasant interview. Maybe one more note: on September 30 and October 1 we’ll be at the all about automation in Chemnitz. You can meet us there in person, have a chat, and gain further insights.

Perfect, then I’ll include the link in the show notes. And if someone listens after that date, no problem—just get in touch directly. Thank you all, have a great rest of the week, and see you next time. Bye.

Raphael

Thank you, bye.

Adrian

Bye.

Maximilian

Bye.