Episode 41: Industry 4.0 in Action

In this episode of The Sound of Automation podcast, we sit down with Gary Krus, the Vice President of Business Development at HIROTEC. Bryan and Gary discuss the practical side of Industry 4.0 for manufacturers, including the importance of data collection and ERPs. Listen in to learn how to navigate the evolving Industry 4.0 technologies and prepare your organization for change.

For more information on Wipfli, please visit our website at:
https://www.wipfli.com/manufacturing. You can also contact us here.

Transcript

Gary Krus 00:00

Setting up an ERP system is not an easy thing to do. But if you think through it and go through the steps and get prepared, that ERP system is going to help you through that digital transformation, which industry 4 .0 is, right? It's really trying to get to that digital twin. 

Intro/Outro Narrator 00:18

Welcome to the Sound of Automation brought to you by Wipfli, a top 20 advisory and accounting firm. 

Bryan Powrozek 00:37

Hello and welcome to the Sound of Automation. I am Bryan Powrozek with Wipfli. Today we'll be diving a little more into Industry 4 .0. Our last episode kind of covered some of the more theoretical aspects of it. So we figured it'd be good to kind of give the other side of the argument what it's like to go through a practical application of some of these technologies actually out on the shop floor. So joining me today is Gary Krus of Hirotec. Gary, how are you doing this morning? 

Gary Krus 01:09

I'm doing well, Bryan. Thank you for having me on. 

Bryan Powrozek 01:12

No, thanks for thanks for agreeing to come on. So I guess before we dive into the kind of the meat of the episode, you want to just give everybody a little bit of your background and kind of why we would have tapped you to talk about Industry 4 .0?

Gary Krus 01:27

Sure. My background is an electrical engineer. I graduated from Lawrence Tech. I started with Hirotec just over 30 years ago, and I've been doing controls and automation for the last 30 years in the automotive business. I work for a company that is a global company. We have operations in eight countries, 30 locations, and we're a tier one supplier to the automotive. So we supply closure systems that steer doors in your vehicles and exhaust systems to the automotive. At the same time, we design and build all the equipment that we run manufacturing on. So my background is really automation. 

Bryan Powrozek 02:07

Yep. So you obviously very familiar and have bumped into a lot of these technologies that, you know, I often like to say, I could probably talk to 10 different people about industry 4 .0 and get 10 different answers of what exactly it is and what they're trying to do with it. And so from my position as a Reformed Engineer and how to turn to Accountant, I see there's this gap, right, between the theoretical of, you know, additive manufacturing and machine learning and digital twins and all those things that on principle sound great. But then when it comes down to actually implementing that on a shop floor, there's a pretty big gap in my mind. And I think you would agree with that. So just curious, based on your experience, both within Hirotec, and I know you've done some work with some other smaller manufacturers and things like that, where do you see the common issues that the small to mid -sized manufacturer runs into when trying to adopt some of these technologies? 

Gary Krus 03:11

Um, I'm going to start with a preface. Being the automotive, automotive and automation has been pretty well advanced. So we've been doing this for 30 years. We're designing parts in product design in order to manufacture them. So the goal has always been from the OEM. How can I take their tooling costs down? How can I take the running costs down? How can we lower the costs of this part so I can keep my vehicle at a lower cost? Right? So the automotive has really been doing this heavily because we're automated. We're highly, we're making a lot of parts. So what we're doing up front is we're designing the part so that we can manufacture it, design for manufacturability. And if you're running a manual process and you're, you're doing things the manual way where you have a hammer and hand and you're tacking it together with, with a torch or a manual spot-welding system, you don't think about what it takes to do it automatically. So the big, big thing I see when I go to a shop that's making stuff and they have a whole bunch of welders that are doing it manually, they can take a hammer, move it over a little bit, slide it into place, tack it down and make it, make the assembly go together. Yeah. And if you think back to the fit and finish for the last 30 years, at one point, the Japanese were better than the than everywhere in the world for fit and finish. And 30 years later, everybody's about the same for fit and finish. And what to do there? We designed the product for manufacturability. Yep. That's the biggest key starting there. 

Bryan Powrozek 04:51

Yeah, and I think you hear, or at least I've heard, you know, some horror stories of, oh, we tried to automate a step in our process, and it just went horribly. And so we, you know, we're, we're a little gun shy to try and do something else. But oftentimes it's I think it comes back to exactly what you're saying there is not having a good understanding at the outset, or just saying, okay, I'm going to replace my existing manual process with a robot. When, as you mentioned, if the, if the design's not capable of meeting the criteria for that robot to weld, you're just going to, you're going to be throwing away money and you're going to be upset and miserable with the process as you go through it. 

Gary Krus 05:33

Right, right. So if you're doing arc welding and you've got to do a fillet weld between two pieces of metal and that line keeps changing Because you didn't put some geometry gd &t to the parts so that when I set them together that they may Repeatable right within a certain intolerance so that my process my automated process can happen automatically Yeah Technology is getting there you could use vision you can use some other things to get there, but it's not 100% there yet Not like a refined process where everything's perfectly in place And now I can have a robot and program it to a point and do its work um And technology is changing. I mean It's getting closer and closer to where we can put vision on a robot and do it almost everywhere but it's not quite there yet the last 10 years have been huge in how it's changed Technology keeps changing every day and it's advancing quickly.

Bryan Powrozek 06:30

Yeah. Yeah, and then you find yourself in kind of in this loop, right, of the technology advances. So you need to update your designs to accommodate that newer technology, which then drives changes in your processes. And it's kind of that continual loop of, you know, evolving the design of the part and the process for manufacturing it. 

Gary Krus 06:54

Right. I mean, even for tooling design or the designs we're doing today, in the past, we had all manual, um, million centers and bridge ports and, and lays. And you did everything with a, with manually and with a, with a, um, protractor, right? Or, or Caliper to measure your, your parts as you're making them today with CNC automation, the whole manufacturing process has changed. Right. It's all digital. And now if I do a digital and I do it all according to math and my cat, my cat drawings, it should be darn, darn good. I don't need as much slip planes or adjustment because my tolerance seems going to be a lot closer. So that's changed the way we design things also. So the manufacturing process, laser cutting out of a flat sheet metal, where before when you took a torch and you blew it apart and where you put it on a big band saw and let it cut forever. But even the new processes help drive ways to, uh, automate better. 

Bryan Powrozek 07:57

Yeah, and I know my engineering background was in the world of stamping, right? And so you just always had that idea that there's going to be some level of scrap, right? The runners that were holding the parts together as they went through the progressive die or whatever, it's like, well, that's just going to be scrap. But now, as you mentioned, are there other ways to get those parts fabricated that are then going to go into the next step of the process that don't have that same level? Okay, maybe you're paying a little more on the front end to process the part, but in overall, the process is going to be more efficient because you've got less scrap or you're able to maintain tighter tolerances or whatever it might be that the benefits are as you go through it. 

Gary Krus 08:40

So, you know, Hirotec, and I'm going to put a little punch out here. Industry 4 .0, big data, advanced robotics, simulation, integration, IoT, AI, cybersecurity, cloud computing, additive manufacturing, augmented reality, right? Big data. We're just starting on that because that is fairly new. We're collecting data from everything out there. Advanced robotics, we've been simulating full lines for the last 15 years. What we haven't done is simulate it 100% where we put in all the safeties, all the controls, all the paths, everything where we can get to a point where you can do a virtual tryout. 100% virtual tryout. Simulation is huge, right? And not only for automation, but you just mentioned dies. How can I simulate my die process to eliminate scrap? How can I simulate my final part so that I can do the assembly and it matches perfect what I want for final dimensional? All that's right on the verge of hitting the next level. There are companies out there that are using this technology to do virtual simulation, right, to help us get to that point where we have virtual tryout and we can take less of it, less of the tryout to the floor and have it all done in a room, right? Yeah. Yep. So the technologies are advancing pretty quick. Integration, I mean, that's your digital ERP system. How do I tie each department together? That ERP system ties every department together. You can't miss it, right? From purchase order to final production and shipping, it's all in that ERP system. You have a digital chain, right? We're talking about digital twins. The digital twin is what we get from the simulation as well. 

Bryan Powrozek 10:33

So just out of curiosity, because you ran through a pretty big list of technologies that most folks associate with industry 4 .0. And just curious to hear your approach or Hirotec's approach of how do you prioritize and select these things? Because it's kind of the proverbial, how do you eat an elephant? There's just so many places to go and to start. I'm curious how you've seen successful companies prioritize where they're going to spend their time and their energies first and then work down the list. 

Gary Krus 11:07

You know, it's a little bit of everything. And where I was going is we've actually done a little bit of every one of these topics, right? It's a step -by -step process. You've got to start, where can we start? How do we get into this, right? So the big data, we started collecting data on our presses. If our press goes down, it shuts operations down. So we're starting to collect data so that we can figure out when that press might predict it to go down or when we should be doing maintenance. It's not a big project. I'm just, I got some sensors on it and I'm collecting data. At some point, I'm going to figure out how to do the analytics to that data. Yeah. In time, I'm collecting some data so that I have something to work with. It's not, I don't have a full -blown solution, but if I don't start collecting data, I won't have it to actually figure out how to use the next step in the solution. The advanced robotics, we've been doing robotic assembly. Actually, Hirotec, we brought a lot of different practices to the US in the 90s, where we are doing full robotic systems and most of the OEMs are, we're doing full hard automatic tooling. So flexibility of robotics is there, but we've also started using the AMRs, right? So we added AMRs to one of our processes. We added AGVs. Now we've added AMRs. We're testing it out. We're running production. We're seeing how we can be more flexible with it. Simulation, that really came along with robotics in the late 90s. Simulation was mandated by our customers. So now that we have that simulation, we're trying to take it to the next level. The integration, the ERP system got rid of all the manual Excel sheets and manual paperwork and all that. Everything goes into a database. It's easy to report, right? I think every company that's running production should have an ERP system. And if they're setting it up properly, they're going to be using it to help them run the operations with bad data today that helps me solve my problems for tomorrow, right? Instead of waiting for a month, hey, they got some bad parts. Well, I've got a month report before I got it because somebody put it into an Excel sheet and then figured out how to put it into a PowerPoint to show it to management. Instead of I got that data today because the ERP system can give you that data today. That's to me is the integration side. It's really a, we're touching each area and I'm saying just touch it a little. Look at how you can put it into your operations because setting up on the ERP system is not an easy thing to do. But if you think through it and go through the steps and get prepared, that ERP system is going to help you through that digital transformation, which industry 4 .0 is, right? It's really trying to get to that digital twin. 

Bryan Powrozek 13:51

Yeah, and I like what you said about the project you're undertaking with the data acquisition and, hey, we're going to start collecting data from a couple machines. We're going to see what that data tells us and then we can figure out. I feel like it's kind of borrowing from some of those, maybe not exactly, but the agile methodologies, some of the methodologies that the technology and software companies have really focused on, failing fast, just do something, learn from it, then do the next thing and the next thing and the next thing versus kind of that waiting till you have the perfect mousetrap before you implement because you're never going to know all the potential challenges and pitfalls. So taking that step -by -step iterative approach gives you a chance to, because you might start going after these two machines and then realize, yeah, we're not getting much there, but if I take this exact same thing and put it over here, now I'm getting some actionable insights. 

Gary Krus 14:50

So yeah, the first the first project I did was I set up our CNC machines to collect data about the cutter cutting. One was the spindle spinning and one was it not spinning. And if I had an operator working at the machine and it spent more time open and not spinning, it helped me have insights, right? Oh, I got to figure out how we can change this stuff out faster. Yeah, how can I get that cutter cutting because that's where I'm making money when I'm cutting, not when I'm setting it up. So that was the first project we did. And then we took that same concept and we brought it into our automation. And we put it on one of our assembly lines in production. And we looked at every slice that was stopping the line from running. We collected enough data that we could look at every step of the way and say, hey, we're getting a lot of slices here. If we can get rid of those, that robot will be on time. It'll be less, less of a bottle. And we applied it little by little. I mean, we've been doing this for six years, seven years, actually, applying it to our automation, step by step. 

Bryan Powrozek 15:54

That's, I love that example of the spindle spinning, because I've heard so many examples like that where it's something just that simple, right? Is it spinning or not? And then when you realize, okay, we're spending 50% of our time with this spindle not spinning. Now let's start figuring out what the cause of that are. Oh, hey, half the time we're just waiting for parts. You know, we don't have good material flow. So if we can fix that, you know, it's not that it's a maintenance issue. It's not that it's, you know, the workers aren't there. It's literally, we just need to get parts to the cell to keep it going. Right. Quick fixes and it's all coming from one very simple KPI spinning or not. That's all we need to know. And a lot of this speaks to, I think having the culture that kind of embraces that willingness to try new things and fail fast and know that not every idea is going to be a home run. But if we, you know, if we step to the plate enough times we're going to get a few that are and come back. So is that, what's your feelings on how culture impacts the ability to have successful implementations? 

Gary Krus 17:08

Fault your impacts just because people don't like change. So if you're running a manual system now and all of a sudden you want to automate it, they're going, ah, you can't automate it. It can't happen. You can never do that until you actually show them the process it takes. So don't try to automate everything all at once. Show them one good area that you can automate, and then let the rest of the team figure out where's the next best area to automate. You're trying to get engagement. Management wants to automate. I'll give you money to automate, but if you don't have the team on the floor engaged, it's pretty hard to automate. 

Bryan Powrozek 17:45

Yeah. There was a book I read a few years ago called Switch, and it's all it gives some basic ideas for how to manage, you know, transformations and things like that. And that was one of their examples is, you know, go for those couple small wins, where you can get people going, get motivation behind it. And then you can build to other things but find one. And I can't remember, this wasn't from the book, this was another article I read somewhere, but they were talking about the thing was the British Cycling Team. They set a goal of a 1% improvement in all facets of their program with the idea that, yeah, 1% is not much, you know, get the bike 1% lighter, go 1% faster. But as you now start adding all those 1% together, you pick up momentum and it really makes some pretty significant change. So I think that's a great piece of advice to not think you have to do this huge project and this huge change. Focus on little things that then get people excited about doing this. 

Gary Krus 18:51

Right. And that fits along with the, the Toyota way, right? The Toyota way. I got 30 ,000 employees if each one come up every day and figure out how to save a penny. That's a lot of money. That's a little bit of change, but it's a lot of money in the end. It's a lot of change in the end. If you get everybody to come up with that one penny of savings, maybe today it's got to be a dime. But back when you stated that a penny was a lot. Yeah. No, it is the small, but the other one is having a champion from the floor. You do have to have a champion from the floor or from the area that you're there. You've got to, you've got to get somebody on board that wants to take it home because the guy upstairs that's dreaming it up and doesn't live it is not going to make it happen. 

Bryan Powrozek 19:36

I guess how of how have you approached that thing in trying to bind and get that champion on board because I could I could see in a lot of environments where it's like, okay, now this is just another thing and this is another flavor of the month that I'm going to have to deal with. And so I guess any anything you've done special in that area to try and. Simply people. 

Gary Krus 19:57

I go out and listen to their problem. And then I hopefully can help them change to eliminate that problem or get, maybe it's an upstream change that has to happen. So I work on helping them get that change, but I also work at them saying, now how can you help me get to this change, this next change? This is what I see, this is my dream. Start thinking about how do we get there, right? So we have to plant the dream or the seed to your employees. This is where we want to get to start thinking about what would, how can we get there? It's actually planning that challenge. Excellent. 

Bryan Powrozek 20:34

Well, Gary, I appreciate you coming on. This has been helped. I know it's as interesting as I listen back to the last episode and it's very high on the conceptual and all the great things, which I do think industry 4 .0 creates a lot of potential for manufacturers of all sizes. It's just how do I take this nebulous concept and draw it down to something that will be impactful for my business, which I think you did a great job at giving some practical examples of how folks can do that. Just out of curiosity, if anybody is listening to this and they want to reach out, understand a little bit more about yourself and Hirotec, what's the best way for someone to get in touch with you? 

Gary Krus 21:23

Email GaryKrus@HirotecAmerica.com or you got my cell phone number 248 -640 -3886. Send me a text, we can talk. 

Bryan Powrozek 21:34

Excellent. Well, Gary, thank you very much for the time today. It was a great conversation, and I appreciate you coming on. 

Gary Krus 21:41

Thanks for having me and I love talking about this. I love change and I love technology and I like to figure out how to use it. 

Intro/Outro Narrator 21:47

Thank you for tuning in. Don't forget to like us, subscribe, and share on social. To learn more about Wipfli, visit us at Wipfli .com. That's W -I -P -F -L -I dot com. Perspective changes everything.