Hannover Messe Week - Day 3 - Rainer Brehm

Good afternoon, good morning, and good evening. Welcome. This is the Industry 4.0 Club with IoT World, and we're doing our— one of our broadcasts this week from Hannover Messe. We're going to be speaking today with Rainer Brehm from Siemens. But before we do that, I'd like to give some of our other hosts an opportunity to introduce themselves. So first of all, I'm Mike Ungar. I am an Industry 4.0 Club co-founder. When I'm not spending time here on the Industry 4.0 Club, I am a business coach and executive coach. Mike, you want to quickly introduce yourself, please? Sure.

My name is Mike Yost. I'm also one of the founders of the Industry 4.0 Club and also founder of a company named Bennett. So Lucian, would you like to introduce yourself, please?

Yeah, Lucian Fogores. I'm the co-founder of IoT World, and we're on a mission to basically democratize the industrial market insights.

Excellent. So let me introduce Rainer Brehm. Rainer has worked for Siemens for almost 2 decades. So during this time, he's held many leadership roles both in Brazil and Germany and worldwide as a project manager, management consultant in the general motion control area, process industries and drives, and automation systems. And he's now the CEO of factory automation. I really appreciated what he had on his LinkedIn page, so I'd like to just read it because I think it's very telling about who Rainer is. He says, our customers in the manufacturing industry can, can rely on our products and solutions to accelerate their business. Digitalization will leverage new potentials and ensure competitiveness. I'm enthusiastic about collaboration, co-creation, and the future of automation, shaping and developing new products and solutions for our customers are his key priorities. And he also shares one more point about— I thought was interesting— his personal credo is lifelong learning and a passion for the things he does. So, Rainer, thanks for taking the time to talk with us today about Hannover Messe and about Siemens. So could you share maybe one or two key things that you've seen so far from from the Hannover Messe show?

Sure, I can. But first of all, also, thanks for having me on your call. And I was a little bit jealous because I'm not the founder of Siemens. Siemens is already 175 years old. But, you know, I'm really jealous that I have 3 founders in this call. Yeah, but a good question. I mean, we are streaming live from Hannover Fair and it's the 3rd day. And it already felt like we are here since 2 weeks because there's so many interactions. And you really feel after all this COVID virtual meetings, people are really looking forward to meet in person. You know, at the first day, at the beginning, a lot of internal people even met the first time since 3 years. But now a lot of customers came in, politicians came in. So it was really, really exciting. And everybody kind of was saying, wow, I mean, the COVID situation, the pandemic changed a lot of things. Digitalization really became— became real and became accelerator. And I heard from a discussion on the opening of the NOVA fair that somebody said, you know, it's kind of now a ketchup bottle. You know, you always try to put on that and that something comes out and suddenly everything comes out. And it seems to be now digitalization after that time really became reality. And you see so many real use cases, so many real use cases which went away from from proof of concept really into scaling. And so we really see that transformation speeding up.

That's fantastic. Is there one thing in particular that has struck you that maybe you weren't expecting?

Well, you know, everybody talked about digital twin aspects, but I know there are, even on our booth, I was surprised of, you know, there's not this one digital twin, but, you know, how many digital twins you can have and how you can combine them And also now, and this was also something, how this digital twin can be used in an execution mode. So while you run a machine, while you run a plant, you have a digital shadow or digital twin running in parallel. And that's something which really I saw in a lot of cases.

Well, that's excellent. I want to just for one second back up just a little bit for my sake. I know you're not the founder of Siemens, appreciate that, but maybe talk a little bit about the mission of your division, the factory automation division. vision of Siemens? What are you guys focused on?

Well, first of all, I'm responsible for the SIMATIC business, which is the Siemens PLC, HMI, IPC business. The SIMATIC business is on the market since 64 years. So basically it was kind of defining Industry 4.0 when it was founded. That was kind of the initiation of having that kind of product. automating structures. And I think that our role is shaping the future, driving it forward, and driving Industry 4.0 forward. That means providing best-of-class products that, you know, the challenges of today, which are different than the challenges maybe 46 years ago, can be solved. And those challenges are more going into decarbonization, sustainability, which is a huge topic on the fair. This goes into lack of skilled workers. You need to automate things which might have not been automated in the past. You need to go into the aspect of smaller lot sizes because you have a deglobalization in some areas or you have individualized products, how you automate small lot sizes. Because in the past, when Ford was introducing the assembly line, everything was the same and you could automate it. And there's automation good because you automate something which is repetitive and which is known. The question is how you're going to automate in the future something which is not repetitive, might be a lot-sized one, and might be not known. And I think our goal must be to drive that forward to solve, have answers for the challenges of today.

I think that's fantastic. And one of the things that we see a good bit on is that, to your point about automation, to your point about the realities that people are facing today and now potentially automating things that we didn't automate in the past, or maybe we were afraid to have conversations about automation, right? Because it's displacing workers and now we have labor shortages and we have to, you know, allow people to do more work. And so we're leveraging automation. A lot of those principles and practices have been around for a long time, like you just talked about. And that, if that's more of an Industry 3.0 approach, right? absolutely nothing wrong with getting people there. But the transformational aspects of Industry 4.0, right, the different mindset, right, the way people are looking at business opportunities, looking at big challenges like globalization and carbonization like you were talking about, to me seems like the point of really getting people grounded that there's really good stuff happening there that we've been doing for a whole long time. But also there's a whole bunch of new things from a mindset and from a business model and a transformational perspective. Do you guys get into that? Do you see the transformational aspects and people getting excited and understanding just what's potentially possible on the 4.0 side?

Absolutely. I mean, just what you said, I think without automation, digitalization, there will be no sustainable future. That is one of the main enablers to driving this forward, number one. Number two, You know, the technologies you need to do that are not the classical technology. So you need to leverage much more IT. And, you know, we have Industry 4.0, IoT. So you need to leverage IT technology in the OT world. And that convergence, leveraging those technologies, really, really going to help. And that helps of, you know, making things like automate the unpredictable. leveraging AI that makes that possible, number one. And number two, but if we want to make this possible, we also need to move the people into this direction that they can leverage and use this technology without having a PhD in AI or how to use TensorFlow or programming in Python. So we need to have that skill, that people being able to do that, number two. And number 3, out of that, for sure, new business model, as you said, will evolve. Because if you talk as a service, if you talk about, you know, when you talk IT, you talk about DevOps. It's a different thing how you now update frequently, maybe your plant, your machine, and the value streams are also differently and the profit pool might be differently. So we see that as well coming with these new technologies.

Thank you for that. That's real interesting to think through how things have evolved. I know you talked about the IT really and the OT coming together. Is there anything else or is there another area that maybe you think manufacturers need to discover about Industry 4.0 to really help them move forward?

Well, yeah, I mean, the one topic is, as you said, it's a lot of technology. And I think there is a big dominance of IT technologies. And then you can tell all the buzzwords and probably most of them are relevant. The other thing, the aspect which we really cannot or must not underestimate is the skillset of the people. And how many customers doesn't have any automation department anymore? And how many operators are capable of leveraging those tools? And, you know, we always talk about, you know, POCs are fine if you do it in a lab. You want to scale it. And if you want to scale something, you need to make this complex technology as easy as possible. and make it available. And that upskilling, reskilling, getting OT attracted to really new talented people, that's another dimension which is so important.

Yeah, I think it's really interesting that you talk about that because it's, you know, having worked in manufacturing for decades now, that story about upskilling people and involving the people hasn't changed. It's really now different technology and different things that we need to learn about. It's very interesting that you say that. Lucian?

I'm just going to expand a little bit on that if you can, Rainer, is what are some of the things that Siemens is doing to democratize some of these tools like you've talked about digital twin earlier in the conversation and also the sustainable tools that you've introduced?

Yeah, well, you know, if you look back, I don't know whether it's Industry 3.0, but, you know, when it started, I mean, the PLCs, you normally had a PLC and you had central I/O, and then suddenly the field buses came up, and then everybody, there was a field bus warrior. So, you know, Rockwell had DeviceNet, Siemens had PROFIBUS, and then there was EtherNet/IP, and then was PROFINET, and so on. I think the future will use standard technology because, you know, this is over that, you know, you do very proprietary, even when it was— it's still open, yeah, it was accessible. But I think it will much more use standard technologies. And one example is when we talk about IoT or about edge computing, Siemens was developing an own edge device, which was really, really good, really good. And we decided to stop the development. I was acquiring a company which was using Docker technology because I said, hey, We cannot use proprietary technologies. Even I had it already 5 years ago, I could deploy from the cloud functionality to the OT world, but it was proprietary and that is not state-of-the-art anymore. So we stopped that development, acquired a company which used standard technology to make it easy and accessible for everybody, to make it easy and accessible for IT people to write functionality to the OT side and using that technology on the AI side, on the IoT side, on the edge side. I think is so important. And that is for me a democratization of technologies and making data available from the shop floor, which might come from proprietary or from legacy systems.

Thank you for that. I think Mike and I were in a conversation just a few weeks ago in Cleveland for Manufacturing Works, one of the organizations that represents a lot of the small to mid-sized companies, 20 people or less sometimes. And the challenge is how do you get them engaged in a conversation?

Yeah, that's where I was gonna go also. I was just gonna ask, Ryan, your thoughts on the channels, right? So the channels to serve the masses, right? 'Cause the technologies, and we talk with a lot of technology providers that say their solutions can scale, but their business models can't, right? And so, you know, traditional automation type of solutions were sold from a technology provider to a systems integrator or a machine builder, and then deployed to the manufacturer. And that if every— if all the solution providers are all trying to funnel through that same channel, right, it gets really hard. It becomes a bottleneck. And obviously we can't serve the masses. So your thought at all on not only the technology, as you mentioned, but where do you see things going from the channel perspective?

That's a good question. No, that's good because, you know, what is the value of system integrators and solution partners? Normally they have a certain closeness to the customers from a location perspective, but they also have domain knowledge. They speak the language of the customer. And if you look at standard IT integrators, they are more abstract. They don't understand the process. They don't speak the language. So from my point, also the channel partners, they need to re-upskill. They cannot stay in the world they are. They need to gain that IT knowledge, but combine it with their domain knowledge on the processes. And then basically, I think they need to grow as well there. I think this is number one. And number two, when we talk about maybe complex technology. My question is, how can we make it easy for the people already using it? Because the one is the integrators, you know, to catch something, but then you have the plant operators, you have the maintenance persons. How you can make it accessible there? Because they need to run the plant. And what Siemens— this was one of the reasons why Siemens was also doing another acquisition, which was called Mendix. Mendix is a low-code platform. And the idea was, well, there are a lot of OT people in the past. You know what they can do? They can do Excel and they were maybe writing something in VBScript. But couldn't we make the person kind of the supervisor in the operation? They have a problem and they want to solve it, but they cannot program something in, as I said, Python or C#, but maybe they can do low-programming. And if they can use Mendix in this case as a local platform, describe their problem, and then deploy it on the shop floor, not doing it in Excel or whatever, I think that will also enable them from a go-to-market perspective leveraging these new technologies.

Yeah, to me, that's the thinking that's different. And we have challenges. We talked about the channels and the way we serve the markets. There's friction and there's issues all the way across, and that's in the manufacturer side and that's in the training and education side, it's in the economic development side, in addition to the solution provider side. So I think that thinking of how can we make it low-code, no-code, because a lot of times people in our space are just afraid to even have the conversation, right? That can't happen here in manufacturing. We're, you know, we're different. And if we think that way, and if we don't offer solutions to give people the opportunity to solve problems with these kinds of solutions, we're just going to continue to hold ourselves back and hold the industry back.

Absolutely. And I hope that there's always kind of 30% who are very enthusiastic and doing new things. You don't need that for that. But then you have the other 30% who at least need some, you know, getting over the inertia, having some inertia, but then are running. If the other 30% then see, okay, wow, now they do it, so we also are forced to do it, then we have over 90%, and the other 10%, okay, Yeah, I cannot help anymore.

So that's fantastic. Yeah, I want to return to something that I heard you— we heard you talking about it at Hannover Messe, and that's this idea of IoT burger. So can you talk— what is an IoT burger? I mean, that's a— I know some folks may have listened to you already, but for those who haven't, can you explain what that means, what that's all about?

That analogy was coming up because everybody talks about platform and data and so on. At the end, that is not tasty. And you want to eat something. You want to eat and you want to eat a burger. But the burger, for me, the analogy, you have a real problem and you want to solve the problem. And whether you solve it with edge or with cloud or with AI or whatever, I don't care. I want to eat a burger and I want to give you a burger which solves the problem. I don't want to talk to you, oh, Siemens has a very good industrial edge, take this, or we have very nice cloud, take this. I want to offer the customer something solving their problem, and that's for me the burger. For sure, when you want to do a burger, you need a kitchen and need ingredients in order to do the burger. So you need the ketchup and so on. But, you know, how we can provide this, but all that kind of ingredients, which might be interesting for the cook. We also need to serve the cook. This was a channel partner, which is maybe a system integrator. So, he's doing the burgers, yeah. But, we also need to address the one who wants to eat the burger. And, for them, it's not good enough to give ingredients in the kitchen. You need to give them a burger, which they already can eat and then easily digest.

Oh, that's a great analogy. And, I really appreciate the The thought process. I worry sometimes a lot of solution providers tend to sell the, if you will, a solution looking for a problem. But you're exactly right. And in fact, you need— a manufacturer needs to have a problem and find solutions out in the marketplace that can help them solve that problem. Yeah, it's a great— that's a great analogy. I really appreciate that a lot. We talk about one of the technologies I know you're very well versed in is AI. Can you talk for a moment about, if I'm a manufacturer, what are some of the things I need to do to be ready to really leverage AI in my environment?

Yeah, a very good question. And, you know, it's a fancy technology, so everybody likes to talk about AI. And I think Siemens has alone more than 200 people working in pre-development or kind of technology research on AI. But now we also built up a lot of people who are really implementing it. And what we found out, scalability of AI is really hard. It's really hard. I don't know many use cases really scaled up, but at least know how to approach it, how to scale it up. First of all, on AI, you need to know what is the problem and what is the KPI you want to measure AI against. Because else you cannot judge, is it better or not better? Yeah, number one. Number 2, you need to get the data and you need to get the connectivity. So connectivity is key to get the data and also get labeled data. Maybe you add, because sometimes in a lot of cases you have a lot of good data, because if you look at a manufacturing process, normally you don't have kind of 50% scrap rate where you say, oh, you have bad data and good data. Normally, the scrap rate is low and you want to improve it even further. So, you know, you need to have good data and also not so good data. So if you have not so good data, you don't have so many, so then maybe generate synthetic data. So use data but add with synthetic data, but the data quality is very important. So you have this data and then leverage it. And I give you maybe one example. We work with a lot of automotive companies on the topic of welding spot. I mean, this is a thing which is scaling. So you have on a car, I think you have 5,000 welding spots. And you need to have a quality inspection on the welding spot. And if you look at a chassis, 5,000 welding spots, you don't do that manually anymore. So currently you somehow do this welding spot analysis on very selected areas, but you don't do that at 100% welding spot analysis. But you can maybe think, okay, a welding spot is at least something which is kind of repeatable. Which can be scaled to different plants. And so what we did with automotive customers is say, let's do this welding spot analysis. Let's have good welding spot, bad welding spot, get this connectivity done, number one. Then afterwards, we train the model. I think that's important. But then after training the model, you need to validate the model. And to validate the model that it really fits not only at that one application you look at, but also validate the model that it works different applications. They need to deploy the model, and then the work is not done. AI, you always need to supervise. It's not like you install it once and it's done because somehow maybe something changes, a sensor is changing, you get wrong data in. So you need to also supervise the data and see if something changes in the process, and then you have to retrain the model. At the end, you get really good results, but it's not a You buy it once, deploy it, done. You really need to kind of look at it, maintain it, and you need to make it industrial grade. And so I think the big step from AI moving into the manufacturing world, you need to make it industrial grade. And welding spot is one example. Maybe another, we had in China a very big success with ingot growing for silicon. So we improved the quality of the ingot significantly. We also improved the speed of the growth process significantly applying AI. But again, it was collecting the data, training the model, and then supervising it while it's implemented is a very, very important step.

If we could stick with that for just one second, I want to come back to something you had mentioned earlier when we talked about IT and OT and the automation integrators and the IT integrators. and that focus on domain expertise, right? 'Cause I've had experience with a manufacturer that started on an AI initiative and they got a lot of data and they got good results on a first wave when they did the initial data collection and analysis and, you know, found some insights and they went and took action. And after they did that first round, both sides sort of stand there, stood there looking at each other because they didn't really know what to do next. The people in the plant really didn't have the understanding of how to take those insights and now go to step 2, step 3. And the AI solution provider did not understand the domain. And so we eventually found that company, a person who understood the process, understood, in this case it was process instrumentation and things like that, and they worked together To say, you know, this decision we've just made or this insight we've just uncovered leads us to want to know, you know, where, you know, what's happening at this point. And so they actually started down a path of doing more process instrumentation to get more insights and drive their process. So I guess it just sort of reiterates your point that it's not just— I guess the domain expertise is a big part of this as well.

Absolutely. And you need to know what you want to improve. What is your problem? What do you need to improve? And what are the input data which you believe you can improve it? So this is— you need domain expertise for that. Absolutely.

Yeah, I agree with that. Excellent. Thank you.

Lucian, did you have— did you want to ask the next question? I think you're on mute. You're on mute, Lucian.

Apologize for that. So we're getting close to the end of this. I wanted to kind of ask Rainer one question around the last day. What's on the agenda? What looks interesting to you at last day at Hannover Messe?

You're right, tomorrow is the last day. First of all, tonight we have a booth party, so I'm looking forward to that. But it's only internal now, forget this. For our visitors, yeah, well, last day, it's still very exciting what we show. We have tomorrow, I think, a very great panel discussion about something which we, I think, could do a complete livestream on LinkedIn as well. It's about how we can make the product carbon footprint transparent over the complete supply chain, leveraging IoT, leveraging blockchain. And there's something called SeaGreen, which we have a panel discussion tomorrow, which I really, really found very interesting. So that's on the agenda tomorrow. And on the other side, also for you in the industry which cannot visit, there's also a virtual booth. So we have this year a real booth and we had a virtual booth. In the virtual booth, you can also basically find everything you find in the real booth. Well, you don't talk— can talk to people face to face, but still there's assistance in the virtual world and getting excited in the virtual world.

Thanks for that. I know you're the booth captain, so the party won't start without you. So I want to get you right on time, but I'm going to I'll take over to Mike.

Yeah, thank you so much. So, Rainer, we've really appreciated spending the last half hour with you, and I know there's so much more we could talk about with respect to Industry 4.0. It's just, it's so much to learn and there's so much going on. I hope our manufacturers in our audience have picked up a few things today from you and from really from Hannover Messe in general. So thank you for taking the time to spend with us. If you've liked this conversation, we've been hosting conversations from Hannah Remes this week. The Industry 4.0 Club has. And if you like this confirmation, this sort of conversation, please follow us on our industry4.0club.com and also on LinkedIn. So we've had the pleasure today to have our co-hosts Mike Yost and Lucian Figueres speaking together with Rainer Brehm. And, uh, we have one last opportunity tomorrow where Industry 4.0 Club will be hosting sort of the last conversation about how Hannover Messe went. That'll be happening at 11:30 AM Eastern time or 5:30 PM Central European time, um, with our guest Kevin O'Donovan. Lucian will be joining us again, and tomorrow it'll be Anne Wyatt, Jason Bean, and Ryan Treece. Again, thank you, Rainer. Um, have a beer on us tonight. And we appreciate you taking the time to join us. Thank you so much.

Thanks for having me, and I will do—

Thank you.

Thank you, Ryan.