New LoRaWAN Sensor System – Uncovering Energy Waste in Existing Plants

Hello Uli, hello Johannes. Nice to have you with me today. Welcome to the IoT Use Case Podcast. Uli, how are you doing today, and where are you right now? Where are you off to?

Uli

Hello Madeleine, thank you for the invitation. I’m working from home at the moment. It’s a wonderful Friday. I’m looking forward to it and thank you for having us.

Very nice. Where are you working from your home office, in which city?

This is in beautiful Lower Bavaria in Schönberg. This is in the border triangle on the Czech border and Austria.

Very nice. Great. Greetings to everyone from Bavaria and the surrounding area. Many listeners are from that region. Glad to have you with us today. Johannes, how are you doing? Where are you at the moment? Where are you off to?

Johannes

Hello Madeleine, thank you for the invitation. You can also reach me at my home office, not too far from Uli actually. I’m in Grafenau, in the Bavarian Forest, specifically in the border triangle area near the Czech border, enjoying the beautiful weather. I’m looking forward to the podcast.

Very nice. I am also very happy that you are with us today. Uli, briefly to you, to Thomas-Krenn. You are one of the largest manufacturers of individual servers, storage systems and IT solutions and are based in Germany. You produce high-quality hardware in the field, but you are also a solution provider for individual customer projects. Accordingly, you are also involved in development and contract manufacturing. What “solution provider” means and which use cases you implement here, we will find out today directly from practice. Uli, you are a business developer at Thomas-Krenn. What are you doing there exactly? What does your department do and what clients do you work with there?

Uli

Yes, so business developer is a really fancy term. We focus on the production of individual servers and we are known on the market for it. We are now seeing more and more how important the whole issue is, especially for industrial companies. It’s a very big need when it comes to digitalization, which we are of course very happy to provide support for. And this has now become a core issue for us, where we want to take a focused approach. For me as a business developer, we deal with all kinds of companies these days, whether it’s a doctor’s office that you’re equipping, with traditional IT hardware or, where we’re going into now, a real industrial production line at the automotive company. Today, it is a server that is located in the control cabinet or in the server rack of the IT specialist. And tomorrow it will be the IoT gateway or the IoT server. We are realizing more and more that we want to get the machines connected to the network, and that requires hardware. That is why we mainly deal with medium-sized companies in the DACH region. That’s where we look at where we really want to digitalize and where we can move forward quickly. These are our core projects that we are currently tackling.

Okay, cool. That means it’s also really cross-industry with very different customers. And that’s about using more and more data from the hardware that you’ve probably already installed anyway. And of course also new hardware that you will now also build in the future, which will then be IoT-capable, so to speak. Can you put it that way?

Uli

That’s right, you could say that. But it’s also completely independent of whether that’s on-prem, at the customer’s site, or in cloud or hybrid approaches. Many customers, especially machine builders, prefer to have it locally. It is currently in double-digit growth ranges compared with previous years. You can see that hardware is becoming more and more important.

Yes, very nice. Uli, if I have understood correctly, you are a solution provider. I.e., you have partnerships with software experts to leverage that data. Johannes, you are one of the founders of IFOX Systems, you are partners of Thomas-Krenn. You are a software provider and IT service expert. With your software, I can optimize processes, extract data, and measure value creation and corresponding optimization potentials in euros, all at the push of a button. If I understand correctly, you offer the software-as-a-service solution in addition to the hardware. Your software is then the building block, so to speak, that I need to then analyze and evaluate this data. Did I say that correctly or are you going to add to it?

Johannes

Exactly, you’ve summed that up well. With our software, we show optimization potential at the push of a button. Today we do it in hours and in euros, but now together with Thomas-Krenn we have extended the solution to be able to do it in terms of energy as well. We supply the software components for the hardware that our colleagues at Thomas-Krenn offer.

Yes, very nice. I would also get right to the use cases that you’re implementing in that area. Maybe one more question. Is there a personal story there? Uli, for you, this is also a new area where you are now saying, okay, let’s establish partnerships with software companies. How did you get to know each other? Have you really been actively searching or did you meet at a trade show? How did you get to know each other?

Uli

As Johannes said at the introduction, our offices are not that far apart. We areBayer-Waidler’s, so one of us knows the other. It was actually a spontaneous visit from Johannes and his co-founder. Together we saw where there was potential for a customer to get everything from a single source. And then we said: let’s look at it together, where do we have common approaches? Especially during a crisis, it becomes evident where solutions are needed. That’s why we said this would be a common use case that we now want to address together.

[07:47] Challenges, potentials and status quo – This is what the use case looks like in practice

Very nice. I always try to talk about use cases to get away from the products and solutions to the added value of IoT. You are here as a solution provider, hardware plus the software. Can you give us a bit of an idea of what use cases Thomas-Krenn generally addresses in this area as a solution provider and which use cases we are looking at here together with Johannes?

Uli

Thomas-Krenn has its own division for this, called Custom Hardware, where you can really completely cater to individual wishes. If you don’t have a really secure Internet connection or power supply at the site, the technology will look familiar to some, LoRaWAN as an example. That’s one of those use cases where they said, we’ve equipped local businesses that always need to know water level readings in a timely manner so they can pass them on to the environmental agencies. This at a location with no internet connection, no decent wifi and the like.Company equipped to always know water level readings in time to pass on to environmental authorities. There, they simply built it with solar panels, with a self-sufficient supply. Accordingly, you have to store data in a secure environment in order to visualize it.

Also an early fire detection system. If you imagine recycling yards, there are accumulators stored and batteries. There are LoRaWAN sensors that you have connected in a common infrastructure network and then you have to transfer the collected data securely somewhere and store it. The use case we are looking at now specifically is the one with IFOX Systems. If I want to capture energy data from my machines immediately and find out who the main consumers are and where waste is occurring, then we put two and two together and go straight to the machines. Then we equip them with plug-and-play hardware. This allows us to securely capture the data and address the overall solution of storage, data security and visualization together with Johannes and IFOX.

Custom Hardware is a separate brand from you. If you type in custom-hardware.com, you will get to the page where you are the solution provider for this use case. Is that right?

Uli

That’s right. That’s also one of the core competencies that distinguishes us from them. Fine for SMEs, if I then have a special request, how far do I get, where can you do that? We are very flexible in that regard and always welcome it to learn and improve, of course.

Right. I had researched that online, including the use cases there. Could you perhaps elaborate a bit more on your shared vision? Johannes, what is the vision with the data here for machine builders? What are you doing there exactly?

Johannes

The vision we have in general is to help our customers turn waste into profit. We want to support them in this. Energy is becoming more expensive, so this is an important factor that is becoming more important and will remain important, in our opinion. This is the issue we now want to tackle together, we want to help optimize energy consumption and above all prevent energy waste.

Is this new for you, Uli, or is it something you’ve been thinking about for a while?

Uli

It’s not entirely new; it aligns well with the current topic. The problem affects us all. The energy crisis affected us ourselves because we previously heated with gas. We came up with our own waste cases and said, it’s absolutely awesome, let’s see it live right now. Then we said, if it already works for us and we can uncover waste, then we should really tackle this problem in its entirety.

Yes, makes sense. You are also manufacturers yourselves, so to speak, which means you also have your own works and you know the challenges yourselves. You also have a store where customers can order your hardware. That’s also about leveraging data and ensuring availability of the hardware you’ve already sold, right?

Uli

Exactly. That’s what we want to combine now, also together with the IFOX solution. If I now produce a system A, of course you have to know, is it available for the long term? That is our core competence and we try to combine it in a sensible way by including licenses to create both on-prem solutions together with IFOX and a real cloud connection. Everything from a single source in the store.

I will come back to the shop topic later as well. Let us take a look at your customer case. You have a project that is currently under NDA, what kind of machine manufacturer is it? What is it about, what is the process there?

Uli

So it’s a machine builder that’s in the special machine building business for inspection systems, food and food industry, for plastics areas. If you have 350 to 500 employees and a sheet metal fabrication shop, for example, these companies have air compressors in use. Generating compressed air is the most expensive medium there actually is. Those are classic cases where you realize that you receive the billing for the main meter twice a month, indicating the overall energy consumption, but you cannot see which specific areas have the highest waste. The machine builder wanted to address the main waste and measure it for a test period of three weeks. Even there, there was a savings potential of 2000 euros.

Let’s delve a bit deeper into the challenges faced by the customer. There’s a waste somewhere, that ends up being money that could be saved. Can you go into more detail on that? What are the challenges and potentials of your customers?

Johannes

Exactly, now, for example, with the topic of energy, we have the challenge that, depending on how production planning is, how production capacity is utilized, certain things can be switched off in the infrastructure. For example, a company has several compressors that do not always need to run. When you see how the production utilization is or was, you can infer that the devices need to be controlled differently. So right from the start, determining when certain processes need to run, but also adjusting the control system itself to optimize energy consumption. It is also possible to change production planning in general, for example, to minimize heating processes on machines that operate at high temperatures. These are the kind of topics we aim to highlight and then assist customers in improving upon them.

After all, these are precisely the exciting approaches that you have to come up with in the first place. Looking at the compressors and questioning whether they need to be turned on and off in the same manner as they currently are. What data is relevant to this use case, in terms of data collection? You were talking about a compressor and the controller. What is the data that is collected there or relevant to your customer?

Johannes

In the energy use case, the power consumption, how many kilowatts are being drawn and when? Then, of course, it is also interesting to have data on what is actually happening in production, which machines are running, which products are currently being manufactured, and how much energy is being consumed in the process. If you bring transparency into it, if you see how the energy consumption is, then you see that very high energy consumption and peaks occur, although perhaps nothing is produced. The compressor could also ramp up in a way that is not necessary because the production is starting slowly or running at a low level. These are then findings through which one can definitely initiate measures. It is very important to create transparency and to simply see that.

Okay, Uli, do you have any additions from the hardware side as to what data you need? Or is that already what Johannes mentioned?

Uli

It depends on how you get the data. WLAN and as you know, a WLAN infrastructure in an industrial plant, where I might have steel structures and the like, is of course always the best means of choice, purely in terms of connectivity. In terms of data, I can only add that it’s similar to what every household is familiar with. At the end of the month, you would be looking at where your actual consumed kilowatt-hours or kilowatt (kW) peak values are. On average, this has increased to around 100€ net per kW peak. This is such this absolute indicator that has also affected us ourselves. Therefore, we want to capture and visualize this specific value, including the peak, the activation times, and the utilization levels, both for employees and for consolidating the information. Only by achieving this transparency can one truly derive insights and extract concrete measures.

Yes, perfect. Before I get to the solution, one more question. You from Thomas-Krenn also have very different hardware and server structures at the customer. Was it the same here? Did they use your hardware or was it a different hardware? What was installed at the machine builder?

Uli

We have now done this use case twice. In the use case, they provided their own server infrastructures, and then wanted to run this instance of IFOX. That was not a challenge for us. Thomas-Krenn and IFOX collaborated together when engaging with the customer. That is, Johannes ultimately managed this local instance with the customer. And we only have this smart meter connected to the customer’s site, so to speak. And that was the interaction.

[18:04] Solutions, offerings and services – A look at the technologies used

Perfect. Now you’ve just mentioned it, keyword smart meter. Can you briefly explain what exactly the solution can do? After that, I would dig a little deeper into how exactly it all works. Uli, what can the solution that you built together do?

Uli

The joint solution is a complete solution consisting of a hardware and a software platform, where we specifically address the existing systems in the machine park and completely retrofit them. This means that we really want to get to grips with machines without having to know complex values about how the system communicates, but rather just retrofit them. I take the sensor, clip it onto the machine and get the finished visualization directly in my platform. The platform has to be open, of course, because there are a large number of different subsystems at customers. It needs to be an open platform where I can integrate my data once more.

Okay, Johannes, do you have any additions from the solution side?

Johannes

As Uli has already said, the sensors are only clipped on. I think that it is very important that the solution is easy to install. That means the hardware, the sensors, need to be easy to install and set up, as well as the configuration process of integrating the sensors with the software. And that’s where our focus is.

Then I would like to ask for some more details now. That is, you said it’s kilowatt hours somewhere. This data is recorded via a sensor and not sent out via the WLAN infrastructure. How do you extract the data from the hardware? How do you record this data?

Uli

This is the comprehensive package that is part of our solution. This is this smart meter, which is already equipped with summation current transformers. This is known from the principle of an RCD circuit breaker in the circuit. You simply clip this sensor around the power cable of a machine’s phase. In the best case, this is always three phase in industrial operations, because we are talking about large-scale power consumers. They are simply attached, and the current flowing through the line is indirectly calculated, either through the electrical current flow or the magnetic flux, to determine the amount of current passing through the cable. With this, you can calculate the consumption per machine down to the second. That’s the beauty of the solution because you don’t need to physically access the machine and retrieve machine data. You don’t need the machine manufacturer; you can simply plug-and-play with the existing machine, install the sensor, and immediately obtain the data through a hard-wired connection to our smart meter.

Is it always that easy to access this data? I’m asking a bit critically now, because you also have old infrastructure in some cases, you also mentioned the topic of retrofitting. Is it really always the case that you can access the data, or is it sometimes a bit challenging?

Uli

At the beginning of the solution, we focused on electricity because it was easier to visualize compared to gas and flow measurements. That’s why we started with power, with a normal cable that goes to a system, whether that’s going to a PC or whatever. I can attach this immediately regardless of the system. That’s the simplicity and that’s what we want to start with as well. A solution must be simple, quickly integrable, exactly what Johannes also said earlier.

Now you’ve just mentioned infrastructure. How then does this data get to the next level? You have recorded this data. You were just talking about the summing current switch. That is, these data are recorded via the sensor. How do they get to the next level now and then into the software at the end? Do you guys do that over LoRaWAN as well or how does that processing of the data work at that point?

Uli

We retrieve the data from the system by directly reading it from the sensor onto a small mini-PC, the smart meter. Now there are several ways and the ideal case would be simply to get the data into the network via a cable connection, Ethernet so to speak. I can also access it via Wi-Fi, utilizing these two conventional methods. In an industrial environment, using standard Ethernet would be recommended due to its wired connectivity. There is a local database, a time series database, where we can also cache data. This ensures that in case of a system or network failure, there is a time frame during which the data can be temporarily stored in the local database as a backup. And from then on, this integrated solution with Thomas-Krenn and IFOX takes effect, Johannes can take over at this point, because now we could access the cloud or access the network at the customer.

Johannes, how does it work with the data that needs to be stored somewhere? Does the customer have their own systems for that, or do they have their own systems? How does this data hosting work at the site?

Johannes

It depends on how the customer wants it. We offer two options. On one hand, we provide our solutions in the cloud and take care of all the storage, data handling, and filing. However, the customer can still access it if they want to. The other case is an on-premises installation where we use the infrastructure at the customer’s site. If the customer has a central database server, we can create a database and store the data there. You can also set up a separate database for IFOX. That simply depends on how the customer would like it to be.

Not every customer has a WLAN connection. You do have systems out in the field somewhere. How do you handle this with other customers? How do you handle this?

Uli

This is absolutely dependent on the customer case, because it depends on the existing network infrastructure. Security needs to be ensured. The IT manager has the main responsibility for this. That’s why it is crucial to engage with this target group from the start, and an additional challenge is the communication between IT and OT. If there are limitations where we notice that a connection via WLAN is not sufficient. That’s where you can fall back on the familiar LoRaWAN technology from our other use cases. With this, I can transmit small data packets over long-range distances, currently reaching 6 to 10 kilometers, and buffer them with miniature batteries and other components. If you take a quarry, for example, where I can’t measure the level at every point because I don’t have electricity and an Internet connection everywhere, I can do that very, very easily with a LoRaWAN connection and push this data back to the database and from there back into the IFOX system.

Thank you very much for the remarks, because I know that also quite a lot of listeners are not working in the manufacturing sector. You already said it’s about kilowatt peaks., This means it’s about an analysis, because in the end you want to analyze this time series data that is sent from the hardware. How does that work? The goal is to save time and money. That means one needs to perform this analysis and take a closer look. How does that work? Can you tell us more about that?

Uli

In simple terms, I specify in the time series basis at what time x I consume what current power consumption or kilowatt hours. This can be tracked very easily and also visualized in dashboards. It is much more important to look at it over a longer period of time. Johannes had also mentioned at the beginning that it is always important to see the utilization, where to get the utilization data from. That means I need a secondary system, an ERP or something similar.

How many employees are in the plant today, how many are working from home? This also affects the power consumption throughout the plant. That’s why it’s important to think in advance about what belongs in the complete overall structure, in the overall view. You measure that for two weeks. This already gives me a meaningful statistic to see where to start now. Very important is the topic of analysis and breaking it down. This is not part of the complete solution, because a process specialist is usually needed for the analysis. The manufacturing operator knows in their production what, when, where, and how things are running, making it their core competency. That’s a topic where we don’t really want to get involved, we’re just there to immediately provide the data from the machine via retrofit and visualize it together with Johannes. What one makes of it and the derivations for improvement measures are always an integral part of the continuous improvement process.

There the customer is your expert and at the end I can see that there are two compressors running and not one, there is a peak. I can then analyze such things, of course it depends on what the customer then wants to see. Right?

Johannes

At IFOX, we see ourselves as a bit of a data clamp. We break down data silos, we link data together. Exactly the same thing is happening here now. We have several software modules and one module is energy, energy analysis and there the same thing happens. We take the data that comes from the sensors, that is, this time series data with the energy consumption, and we then combine that with other information that we already have in our platform. This can now be, for example, the price information, i.e. the cost data for electricity, that we link the data together and then visualize it accordingly. So we also have other data from other modules in our the software. For example, when it comes to what happened in production, which we can then also link back to the energy values and thus make further evaluations that you can’t do with the energy data per se, because we already have data from various, let’s call it silos, in our platform that we can then use to generate more value.

Exactly, and then the actual IoT case also comes into play, if you can make this connection, then you can theoretically also connect other systems, like ERP. Uli, you said at the beginning that you also have a store. If you implement various cases in the future, you could also connect such data with a store of yours and also use data from customers for you in turn to offer a better service, right?

Uli

There are several ways to do this. After all, we are not a provider of pure hardware solutions, but also have a certain cloud offering. In other words, at the end of the day, the customer’s only concern is where they would like the data to be stored, and from where conclusions can be drawn later. This is individual per customer and we simply talk this through in advance in the customer meeting. That’s the procedure.

You also addressed environmental authorities and so on, in the future it will be the case that more and more stakeholders will want to have data. With the data clamp there is the possibility to use this data also and to link it with other trades. And then, of course, such an entire business case comes into play at the end when I have the possibility of integrating a wide variety of data.

Johannes

Exactly, it’s always up to the customer to decide what they want to use their data for and how they want to link it. But definitely the openness is there for the customer to then use that accordingly and use it for other things.

[29:41] Results, Business Models and Best Practices – How Success is Measured

After all, the business case is crucial for the customer. In other words, you mentioned the topic of energy as a major buzzword at the very beginning. So what is the business case summarized here for your customer? Uli, can you say a little bit more about that?

Uli

We all have the issue of climate and CO2 neutrality, which is actually already legally mandated. Of course, we want to improve our environmental thinking on this and continuously get better and not just waste any energy and really precious resources. That is already the first basic goal. Additionally, one can also see that more and more companies are engaging with the topic of ISO 50001, which is the energy management certification. These are the typical use cases we want to focus on. We want to simplify this in an IoT case. With a comprehensive bundle where I don’t have to shuffle from A to B, we want to address the entire medium-sized business sector.

This is not only the topic of electricity, as you mentioned earlier, but also other data that is then added, gas data or other data that is relevant there. And that’s sort of what you can enable.

Uli

That’s right, and today we’re starting with electricity because it’s the easiest to implement. In this compressed air case, we have already taken the extension into account, so now we can say that we want to focus on flow rates because that’s exactly where we know where waste occurs. We want to get much further into sensor technology and position ourselves more broadly in this area.

Do you have any experiences from the first projects that you might want to share? I think a lot of listeners are in the process of implementing projects themselves. Do you have any experience where you say you actually have to pay attention to it, or are these bottlenecks that really cost time and money? Do you guys have some advice?

Uli

Johannes and I have already been active together, including lectures at start-up centers. We also sought feedback directly with customers and saw that out of 40 participants, only two had really already dealt with the topic and had approached these values somewhere with solutions they had put together themselves. The challenge is quite clear. At some point, more data silos are created, which is what you originally wanted to avoid. You collect the data, retrofit sensors, and don’t know where the data is anymore? Who still has the upper hand? How can I visualize them? By now, both of these companies had to assign a dedicated employee for this purpose. He takes care of the whole thing, so that the data is also secure. At some point, you get in over your head. This is such a principle, a “lesson learned.” It is important to discuss this beforehand and to address this discrepancy between IT and OT right from the beginning in order to eliminate it. Because if one doesn’t talk to the other, you already have a completely wrong objective and that’s already the first principle to start with. If I coordinate beforehand, then I know where the data has to be later and don’t build additional silos with it. These are the two “lessons learned” that I have found for my part.

Johannes, do you have any additions from your projects?

Johannes

Yes, just for this topic with the energy analysis. I don’t think you should underestimate the technical complexity of getting the data consistently and reliably and handling it and then making it really easy for the user to use. You can always put a lot of things together, but it should be really easy for the user to use, otherwise it won’t be done. It’s not enough just to measure data from one device or measure one area, it must be a solution that can easily be applied to other areas. other areas. This is an issue that should not be underestimated.

Last question for today. What’s to come in the future? Wha kind of topics? Uli, what else is there to come in the area of your solutions business? Johannes, what other trends do you see in the software environment? What else can we look forward to in the future from your areas?

Uli

We can definitely be excited about it because we can see that hardware is constantly evolving, and the solutions and software offerings are also expanding. We need to address the right issues completely and together, and really approach them with open platforms. We will jointly build a larger platform where one can truly use various data types, data silos, and similar aspects. From the hardware itself, there’s also a lot going on, whether that’s GPU systems, 4-core, e-learning, with just this collected data, where you then also need special hardware to be able to process that as well. That’s where the current trend is heading.

Johannes

First, two points about energy analysis. One is that we will push the bundles with the production data. If we identify waste in productions, we can transfer it more easily and effectively to energy analysis. Then you can really show how much energy was used to create value and how much was wasted. The other is the analysis and the derivation of measures, that we support the user more in this and simply also make suggestions. How can production then possibly be converted with the aim of minimizing energy consumption?

That was indeed a nice concluding remark, highlighting the many possibilities and developments that lie ahead in the future. You can see that the market continues to grow. There are many approaches that are slowly evolving and hold promising potential for the future. First of all, thanks for the podcast as well and for your insights today. It was really exciting. I think we had a lot of practical use cases. Thank you very much for talking about them. I believe it is not taken for granted to openly discuss and share these insights. Therefore, many thanks to you for joining the podcast today. First of all, from my side, thank you very much. Then I would give the last word to you.

Uli

Thank you very much for the invitation and see you next time!

Johannes

Thank you for the opportunity.

Exactly, so thank you very much, take care and have a nice rest of the week. Ciao!

Johannes

Ciao.

Uli

Okay, ciao!