Looking For More?
Watch this episode on YouTube
Connect with the Manufacturing Industry here www.amtonline.org/events
Discover the past, present, and future of American manufacturing with www.mmsonline.com/madeintheusapodcast
Explore, watch, read, learn, join, and connect at www.imts.com/
Tune in to the AM Radio podcast www.additivemanufacturing.media/zc/am-rad…o-podcast
For the latest in Manufacturing Technology news www.amtonline.org/home
Ramia Lloyd:
Welcome to the Tech Trends podcast where we discuss the latest manufacturing, technology, research, and news. Today's episode is sponsored by Hexagon. I am Ramia Lloyd.
Benjamin Moses:
Hi, Ramia. I'm Benjamin.
Ramia Lloyd:
Hi, Ben.
Benjamin Moses:
We're back to the two-person podcast.
Ramia Lloyd:
Just chillin. I love it.
Benjamin Moses:
Just chillin. Two different people this time.
Ramia Lloyd:
I love it for us.
Benjamin Moses:
We should mention Steve and Elissa are not here.
Ramia Lloyd:
Yeah.
Benjamin Moses:
Why is Elissa not here?
Ramia Lloyd:
Elissa is at Meltwater Conference in New York. She seems to be thriving, and I'm so happy for her.
Benjamin Moses:
They're covering everything that she loves.
Ramia Lloyd:
She... That's literally a conference tailored for her. There's going to be Reese Witherspoon, there's NASA.
Benjamin Moses:
Yeah.
Ramia Lloyd:
She's sent to Slack today saying she was like at the first keynote, and it's a woman from NASA speaking, and an astronaut, and I was like, "They literally pluck that from your brain."
Benjamin Moses:
And they're talking to someone on the space shuttle, or the space station. So they're literally going through NASA to talk to someone on the space shuttle.
Ramia Lloyd:
I just know she's so happy.
Benjamin Moses:
It's so cool.
Ramia Lloyd:
Yeah.
Benjamin Moses:
And then Steve is starting season three for Road Trippin' with Steve.
Ramia Lloyd:
Season three. I know he's also very excited to continue going back to doing what he loves, which is terrorizing people in their own shops for the sake of manufacturing.
Benjamin Moses:
The whirlwind of Steve just coming through.
Ramia Lloyd:
Just coming. He is like a Tasmanian devil, but who asks the most important questions and is just full of energy.
Benjamin Moses:
Exactly.
Ramia Lloyd:
Yeah.
Benjamin Moses:
So one thing that's come up is, with the weather getting a little bit nicer in the DC area, which in summer it's humid, but it's pleasant, it's fine.
Ramia Lloyd:
Yeah.
Benjamin Moses:
We're starting to do stuff outdoors now.
Ramia Lloyd:
Yes.
Benjamin Moses:
And unfortunately, my daughter has friends.
Ramia Lloyd:
God.
Benjamin Moses:
Ah.
Ramia Lloyd:
Dude, that sucks.
Benjamin Moses:
But-
Ramia Lloyd:
Hate that she's so well-rounded.
Benjamin Moses:
... her friend has a birthday party. We're going to Kings Dominion, which is a theme park nearby. So this is kind of breaking in, I guess, our summer, other than yard work we did like a past week or so ago. Do you have any fun plans going on besides-
Ramia Lloyd:
My friend's moving.
Benjamin Moses:
Oh.
Ramia Lloyd:
So I-
Benjamin Moses:
So you're going to get pizza and sweat.
Ramia Lloyd:
And beer, yeah. The obligatory, like, "Hey, come help me move," as you're like struggling. We're not young anymore.
Benjamin Moses:
No. No.
Ramia Lloyd:
So we're moving boxes and all this manual labor, and at the end of the night, she's like, "Hey, you want pizza?" And I'm like, "No, I want a chiropractor, but this is also very kind." I think being able to help your friends move is a different kind of level that... It's kind of underestimated. It's so good to be able to do that camaraderie, and that's what friendship is about. But also, pay somebody to do that, sister. I can't. I can't do it anymore.
Benjamin Moses:
There's a certain age threshold where you're like, "You have to offer pizza, beer, Icy Hot and Advil, or you pay someone."
Ramia Lloyd:
Yeah. Yeah, please. I'm going to need a back brace.
Benjamin Moses:
Yeah.
Ramia Lloyd:
This is going to take my gym experience for the next year.
Benjamin Moses:
That's going to be fun.
Ramia Lloyd:
Yeah.
Benjamin Moses:
So Ramia, can you tell us about today's sponsor?
Ramia Lloyd:
I can. This episode is brought to you by Hexagon's Manufacturing Intelligence, leading the charge in manufacturing innovation. Join influencers, Chris Luki, Nikki Gonzales, and Jim Mayer as they explore tech trends reshaping the industry and catch them this summer at Hexagon Live Global 2025 in Las Vegas where they'll host exclusive sessions featuring inspired innovators, production trailblazers, and enterprise visionaries. Go to hexagonlive.com to register. You can also find the Hexagon Radio podcast wherever you listen, and follow Hexagon on LinkedIn to learn more.
Benjamin Moses:
Awesome. And as part of this effort with Hexagon, we did an interview with Nikki and Travis.
Ramia Lloyd:
Absolutely.
Benjamin Moses:
Can you give us a little background on Travis? Because he hasn't had much context on the podcast.
Ramia Lloyd:
Yeah. Travis is going to be... They created the series called The Architect, and it's being produced by MTS Plus, and he's kind of the architect of manufacturing. That's kind of what the goal is. He'll be the main person for this manufacturing track. They have... I want to make sure that they're calling it tracks. Yeah. He'll be the main person of the manufacturing track at Hexagon Live. And there'll be a couple other different tracks that they'll experience, but Travis is the face.
Benjamin Moses:
Yeah. He is the face of manufacturing.
Ramia Lloyd:
The face of manufacturing.
Benjamin Moses:
And Nikki Gonzales is representing the Inspired Innovators track underneath the manufacturing line, and we're happy to interview Nikki, along with Travis, to talk about some of the technologies relevant to their tracks. So Nikki is the director of business development at Weintek HMI and the co-host of Automation Ladies, a popular podcast in the industrial automation and manufacturing niche. The podcast explores the future of work and the role of automation, robotics and artificial intelligence in transforming our world. So I want to kick it to the interview.
Ramia Lloyd:
Yeah. I look forward to it.
Benjamin Moses:
Hi, Nikki. Hey, Travis. It's great to have you on the podcast. I'm really excited to talk about some really interesting topics. I'm going to jump right into it. The first thing I think we've been discussing is the pace of change of technology is gone significantly higher, a lot more agility is required, a lot more technology is being developed quickly. I'm wondering how that cascades into the leadership for innovation. We talk about technology, but there's a leadership model that's required to support this fast-paced change. Nikki, do you have any thoughts on the leadership model required for this pace of innovation?
Nikki Gonzales:
I have thoughts, yes. Do I have answers? No, not exactly. The thing is, like you said, we're seeing a pace of change that we've never seen before. So we have no mental models or leadership models or management models to really prepare us for exactly what's needed in the future, because I think it's also just going to continue to evolve at an even more rapid pace as we go forward.
So really, leadership right now needs to focus a whole lot on change. People change, types of work change, technology changes, and I think balancing being on top of those changes, while at the same time not letting yourself get too distracted by shiny objects and forcing constant change on people, which can become daunting. But I think right now is the time where organizations and leaders really need to look at, how do they become an agile organization? How does their leadership enable continuous learning and continuous change for their organization that helps them keep up, while at the same time, not just constantly overwhelming people with new initiatives and new things? I mean, we're still working on digital transformation initiatives from 10 years ago, so it can be a little bit disjointed to say, "Oh, we've been working on this ERP implementation for the last 10 years, and it didn't quite go so well. And now, we're going to jump into the next thing," which is the AI, or whatever the newest thing coming out of startups or even making its way down to more mainstream industry.
So it's really... There's a lot of plates to be juggled if you're a leader today. And I think a big part of it is also continuing to educate yourself. No leader is going to be proficient in this day and age or the next 10 years if they're not also continuously learning and changing and educating themselves and becoming somebody that can support change really as part of the organizational fiber, rather than the way that we were able to do things in the past was to take a five-year plan to implement something with kind of a known structure of how we look at ROI, how we look at our expectations based on that change. That's all becoming a lot faster and harder to predict. So being a leader in an uncertain time definitely requires some new skills.
Benjamin Moses:
Definitely. And I-
Steve LaMarca:
Being distracted by shiny objects, I feel personally attacked, but-
Nikki Gonzales:
Don't we all?
Steve LaMarca:
I'm just kidding. I have to ask, have you noticed any patterns in these changes when it comes to generational differences among workers?
Nikki Gonzales:
Yes and no. So I think that there is definitely a generational difference in that the older workers are used to being able to take a skill, learn it early on in their career, and then kind of hone that skill and just get better at it through time. They haven't had to have a wholesale change in how they do things. So they're going to be more reluctant and/or struggle a little bit more with that.
But I think there's a set of people, and we have a bigger concentration of them in the engineering space maybe and manufacturing... Well, I can't speak to the shop floor side of things, but on the engineering and technology side, you just tend to have a higher concentration of what I would call lifelong learners, which are people that have embraced the mindset that they're continuously learning. And we have a lot of those people regardless of generation, I feel like, in the network and in the industries that I'm in quite a bit. So I've seen quite a few of the older leaders that don't seem to be defined by that generational difference or their age in that they are the type of people that are continuously learning, so they're just as gung-ho and/or more effective at implementing change and driving these outcomes than a younger engineer or person.
I would also say I feel like, and this is, I don't work with a ton of Gen Z people, but some of some... I'm in the millennial band, and I know quite a few people that manage Gen Z's, and they also are a bit risk averse and not necessarily as gung-ho about constantly doing a new thing, learning a new thing. They want some stability. I think they've looked at us and go, "What the heck? I don't want that." I don't know. So yes, definitely, there is a bit of a trend there, but I wouldn't say wholesale across the board.
Travis Egan:
Nikki, representing the older generation on the screen here, I mean, I think the tenants of leadership are the same.
Nikki Gonzales:
Yeah. It's about people.
Travis Egan:
Those practical tenants are the same. But to your point, the speed is the change. And it comes from technology and innovation, and that just makes things go that much faster. I think a really important thing to keep in mind is sticking to the practical. So whatever level of innovation or development you're talking about, keeping leadership's mind around sticking with the practical, don't go all the way to the end game, realize that there are steps and stages to get to that point, and really have some successes with more practical things before you really set your sights on the long-term goal. I think it's really an important thing from a leadership standpoint. And we all sort of get caught up in the speed of everything and what technology can do for us, what's possible, versus what does it take to get there.
Benjamin Moses:
Yeah. Or what we need it to do.
Nikki Gonzales:
Well, and the building blocks are so important, because by the time you get to the end stage, it probably has changed.
Travis Egan:
Fair.
Nikki Gonzales:
So take one step, a small step, at a time that can prepare you for the next step, even if you may need to pivot somewhere along the line, because technology outpaced your ability to integrate it into your process. That's quite possible.
Travis Egan:
And you will pivot several times.
Nikki Gonzales:
Yeah. And that's not just about startups anymore. I mean, that's pretty much any business.
Benjamin Moses:
Yeah.
Travis Egan:
Absolutely.
Benjamin Moses:
So one series of technology that's helping increase this agility in the marketplace are simulations and prototyping. I think you've got some interesting perspective on the need for reducing the timeframe for prototyping and the number of prototypes and simulations might be able to support that.
Nikki Gonzales:
Oh, absolutely. They're... The cool thing about simulation is it's not new, but it just keeps getting more and more and more powerful. So Moore's :aw is what has enabled all of this huge leap forwards in AI and all kinds of other things. It's compute, right? We have little bitty computer chips that can run a whole lot more power than they used to. And they've just gotten exponentially better, they've gotten cheaper, they've gotten more powerful. And that's been a huge boon to simulation technologies, which in the past, it was very hard to take something as large as an entire airplane, and to be even be able to model and mesh the CAD model on a single computer, it was just impossible.
Now, you can fit a whole lot more information, you can fit bigger, more complex models onto these computers. You don't have to do as much threading on different machines and kind of stitching these models together. It's become a lot easier. So I think that's probably brought down the cost and the made simulation, advanced simulation a lot more accessible, even to smaller firms. But it's a huge advantage to be able to know how your new product is going to work. Is it going to pass regulatory compliance? Is it going to pass whatever stress tests and things that you need in order to get it to your customer? Can you design it a little bit better using optimizations? If you can simulate how it's going to work, you can have technically an AI, or just there's some various different methods of doing this, run hundreds of thousands, if not millions, of scenarios to see what exactly is the optimal position of a component or the type of component or material that goes in there. From the standpoint of lowering cost as well as increasing manufacturability, knowing what your constraints are.
It's definitely a critical component, I think, today in most manufacturers that are putting new products to market, technically complex products or engineering heavy products. I don't know how anybody at this point could be competitive and put out new models, new products with the speed that the market requires while having to design, manufacture, prototype, test and do it all over again probably multiple times in the physical world. That type of time doesn't really exist anymore.
Travis Egan:
Yeah. I mean, it does kind of bring everything together theoretically, at least from, like you said, there's the design aspect, then you stage to the make or manufacturer, and then you get to sort of the inspection, which then brings you back. But typically, these were siloed and very separate. And to some degree, they still are. But we're much closer to bringing that all together and doing it at a speed in which the effectiveness of your opportunity to get that product to market or get that change to a design or a part or a process can be done much faster. But one of the challenges is the human brain only works as fast as it works and the computer can go that much faster. So you almost need more of sort of an army of human brains to take the little pieces and harbor those along and connect all those stages. More expertise, I guess, per individual and per component of that process.
Benjamin Moses:
And that's a good point, because I was thinking back to my early days. One of first projects I worked on in 2000, I had to work on a routing system for a bleeder system in an airplane. I had to work through the engine into the leading edge of the wing. So on-site to check out that little section in the digital mock-up unit, I said, "This is the frame that I want to check out. It's going to take 30 minutes to just check that out so I can run the routing for that system." Where we are today, it's such a different place where you can iterate and control that space. It's refreshing to be able to look back and say, "I don't need to take a coffee break just trying to check out a model anymore." I'll have that model instantly, because it can start iterating pretty quickly.
Travis Egan:
Does it make you wish you were back there, being able to do it in today's environment?
Benjamin Moses:
Sometimes. But then I realized I probably lost some skillsets from back then that I would probably get blown away. So it's a little humbling when I do tinker a little bit in modeling stuff from my computer and stuff like that and realize how long it's taking me just to get a simple bracket done versus when I did it back then.
Travis Egan:
Sure.
Benjamin Moses:
But that's a different interview we'll talk about.
Travis Egan:
That's why you need all those human brains focusing.
Benjamin Moses:
Exactly. Contextualizing all the different data sets between each stage.
Nikki Gonzales:
Yeah.
Benjamin Moses:
One thing that has come up quite a bit, when you talk about simulations or prototypes, is a concept of the digital twin. Even in our prep session, we're talking about the realistic approach or the myths behind digital twin. Can you guys elaborate what you consider a realistic digital twin in today's space or something more achievable than the idealization of having everything modeled in the perfect space and then running simulations on those?
Nikki Gonzales:
Yeah, I think it depends on a little bit too what you're referring to exactly. A digital twin of a product that's being manufactured, or today, we also have huge leaps in terms of when people are designing new factories and being able to have a digital twin of that factory. And I think that is kind of more of the holy grail of what digital twin as a technology or as a terminology has always promised.
You kind of have everything modeled, a complete mirror of what the physical is going to be. You have all of your inputs and outputs mapped. You may be looking at a virtual version of the control system for the factory, the control system for the robot, and then you have a physical... You have a model of the robot and everything. You can test out how everything is going to run before you put it together and you build it. And then as you run your facility, those models that the physical production actually brings your information back into the model, and everything is kind of a mirror. That's what I would consider, and I think most colleagues that I talk to would consider, an actual digital twin. And it's connected to all of your sources of information, your production system, your scheduling system, all of the raw materials. It's a complete twin of your operation.
On a product side, I'd say we're probably less, and that goes for a brand new factory that's being built today using the latest technology. Again, we kind of have this dichotomy of the front of the line, the early, not even early adopters, but just those that are able to take advantage of the complete cutting edge technology. And then we have the 98% of manufacturers that don't have a brand new build. They've got their factory already, or if they've got their product that they've been developing, and now they're just doing iterations. I think we're very close to, those companies can have a lot of the advantages of a digital twin without it maybe being the complete thing with all inputs and outputs mapped and all data contextualized. That's more where we're talking about the simulation.
Different departments have a twin of the domain that they reign over. And that, I think is very practical and being adopted much more across industry. So whether we want to call those kind of digital threads or some other flavor of digital twin, it doesn't have to be the whole kit and caboodle for it to be extremely useful and for it to be adding a ton of value. So that's where we talk about somebody simulating the product itself, and we're getting closer to having a lot of those inputs and outputs and material information and all of that embedded in the model, and then those softwares being able to talk to the next step.
So we talk about simulation, and then virtual prototyping, and then optimization, and then you manufacture the product, then you inspect the product, and that information goes back into the design cycle. That is, I think, more of a practical application of digital twins of where we are in the majority of manufacturing companies in the US. And that's extremely valuable in and of itself, even if you're not connected to absolutely everything. The more those departments and those softwares can also talk to each other through some basic connectors like APIs, being able to export a model that has context to another software that maybe a different department is using. It's not quite that very sexy digital twin idea that we all had 20 years ago, but it's practical and it's adding a ton of value.
Steve LaMarca:
Nikki, have you noticed anything in... Has there been a spike or more demand for higher system hardware requirements in controls and HMIs with respect to this more capable software and this demand for more detailed simulations? I can't help, as you were talking, to think, Ben and I have had a lot of conversations about when it comes to gaming and ray tracing in our gaming PCs at home. One thing, one reason why NVIDIA has gotten, I think, so popular in the industry is because of ray tracing. And all gamers know what I'm talking about when it comes to ray tracing, but it's the reflection and refraction of light as it passes through various materials and off of different materials that makes graphics in games so much more realistic and lifelike to what we have in the universe. And I think it's really cool how that's being applied to simulation and modeling. So have you seen or noticed a spike in system requirements on a lot of industrial hardware?
Nikki Gonzales:
A spike, not necessarily.
Steve LaMarca:
Okay.
Nikki Gonzales:
But I think the industry has been preparing for this for a few years now. A lot more simulation, a lot more flexibility in system architecture to where you have traditional PLCs and then also soft PLCs that are able to run a lot more of this. I'd say most factories are not wholesale changing from one to the other, but there definitely is an evolution of those systems, not abandoning the traditional PLC architecture, the hardware that makes the machine go, say, "Yes, go," say, "No, stop," "turn this on, turn this off," there's also a lot of safety aspects to why we want to use some of the "older" technology for those sorts of things.
But from an HMI and other sort of additional systems' perspective, tying that in, absolutely. There's been that trend now for a few years. And NVIDIA has been an interesting player in this space, because they make a lot of the chips and software for this,.and they've created open source technologies and really pushed an ecosystem to say, "Hey, we know this part really well. We can do graphics, we can make the chips that can run these simulation models and do all these graphics." Open USD is, I think, the standard that came out of Pixar actually that was developed for movies to make things look hyperrealistic. They've now taken that and applied that to the industrial world. They've made that available through their omniverse simulation software suite, and that's a really cool thing.
Benjamin Moses:
Yeah.
Nikki Gonzales:
I think, honestly, you develop something for movies, let's use it in the real world. Because if we can make stuff look that realistic, why not do that in our side of the world?
Benjamin Moses:
Yeah.
Nikki Gonzales:
And I think it's cool how they've done that. They've partnered with all the major manufacturers in the automation space to make that technology available, not to try to come in and say, "Hey, we know what you should do," but rather, "Here's building blocks. And go ahead and do it."
I still see a lot of companies buying the same old stuff though. So again, I feel like there's a big difference between what the big Fortune 500 companies are doing and what the majority of small manufacturers are doing. They're more worried about keeping their production up and not having unplanned downtime with some of their equipment that's 10, 20 years old. And they don't have these budget cycles of changing things every five years, like a big IT department does. They've got stuff that's been out there for 20 years, and they plan on running it for another 10, even if the hardware is getting to be obsolete.
But that's where you can kind of... The small steps matter. Because you can start to implement some of this stuff, let's say on your HMI, even if you're not changing wholesale, the computing power of the systems that actually run your machines, then you can do that a layer up or two layers up to start to get more into that sort of stuff. The hardware has been just increasing in power though as the cost of the processing has gone down.
Benjamin Moses:
Sure. Sure.
Nikki Gonzales:
So I think that serves everybody. That serves the entire sort of market, as well as the capabilities of being able to do this.
Travis Egan:
Yeah, I mean, it speaks to the importance of digital twin. You spoke of omniverse, and we at AMT were part of that AOUSD, so sort of that alliance to bring it towards the manufacturing space, which as you described started with Pixar and more in the visual entertainment space. The issue is like they're looking at such a big picture with the omniverse and from the entire factory or entire operational perspective. Then you got to bring it down to the process, and then you got to bring it down to the specific machine, and then you got to bring it down to where the cutting tool hits the material. And all of that has to be digitally developed and interplay to get to the end game that that omniverse wants, which they're putting a lot of time and effort money, and to your point, they're doing it in a practical way where they are aligning with leaders in each of those space to get there. It'd be exciting to ultimately get there, but again, I think practical is the way to get there, starting with the smaller applications and growing up from there.
Benjamin Moses:
And I think we hit on a key element, because the past couple of topics that we've been talking about is fairly high level. Now, when we talk about HMIs and human interaction with the machine itself, obviously, we've been talking about some advancements towards there, but you've got some thoughts on the flexibility and recent changes that HMIs have seen for automation and the equipment in general on the shop floor.
Nikki Gonzales:
Yeah. So that's one of the things that I've seen, and I use the term digital twin loosely there as well, but the thing with manufacturing floors are a lot of factories. There's a difference also between obviously machine shops that run machines to make custom parts and factories that run machines that make a high production volume of something, consumer goods or food. There's a whole lot of industries that do a lot of automated manufacturing, or some levels of automation. A lot of these factories are still small businesses that have a lot of things manually going on. But they typically tend to buy equipment, either the machinery to do machining and those sorts of processes, or a custom engineered machine to make the type of product that they make, or a line that's kind of put together between, there's a cell that does this and then the conveyor brings the product to the next stage. And we've seen some really cool ways that these standard machines can work together using robots to handle the material in between.
But usually, those machines are manufactured by different companies, because they have expertise in a particular type of process. There's a packaging machine. There are all types of different processes that go into this. And typically, they are developed by different companies, and they have different control systems. What I've seen is some of those companies are starting to develop a twin of some sort of their machine and then making that available to the operator on the HMI. So the HMI is typically... Most people think of HMIs as kind of a standard replacement for push buttons. It's now a touchscreen. Great. The operator can start and stop the machine. They can see some alarms, or they can see the state of the process on the touchscreen.
What I think is an overlooked feature of HMIs these days is to have an actual model of the machine, what it does, how it works as an explanation to the operator, as well as to any new engineers or technicians that need to come in and work on that machine. Having the bill of materials of items that are on that machine on the HMI as well, so they know exactly what components are in there, if any of them need to be replaced, having schematics, having drawings of the cabinet. You can basically make that HMI, the instruction manual, as well as everything else that you might need to know about that machine, including how to operate it, what it does, all those sorts of things. And that requires the graphics power to be able to represent that machine on its little interface.
And then the other thing that is, I think, cool about HMIs that a lot of people don't know, and that I know at least the company that I work for, Weintek, we've had an HMI, like a smart HMI, and I'll relate it to like a smart TV. There's a TV that you can just watch, and then there's TVs that let you download from the internet and do all kinds of stuff. And HMI can also be almost a gateway to connect that machine up to the plant network and up to any kind of software where you may want to simulate it as a whole.
So if we talk about Brownfield factories, how do we get all of this existing equipment somehow into a simulation like an omniverse of the whole floor? Well, you need information from each and every one of those machines. And some of that is IP owned by the OEM, the original equipment manufacturer. And if they're not providing that right off the bat, it can be hard to recreate that as a manufacturer on your floor. But as these machines are coming out with smarter and smarter interfaces, that information can all be available to bring up to the plant network, up to have that information contextualized, even be able to run some AI simulation, optimizations on top of that, but it requires getting the information out of the machine itself first. And a manufacturer may have 10, 20 different machines from all different manufacturers that they need to access.
So having something that acts as a bridge or a gateway from all the devices that are on that machine, the motors, the VFDs, the cameras, all those different things, how do we get them up to a larger system so that we can start to develop a complete picture of the factory floor? And the HMI actually plays that role. So it doesn't just connect the operator to the machine, it connects the operator to the machine, and the machine to the larger system.
Steve LaMarca:
Got you. Have you seen the physical location of the HMI shift from the onboard controller of a machine to a browser-based UI on an operator's laptop that they use to operate the machine that's not actually physically attached to it other than through a network? Has that been the case?
Nikki Gonzales:
Yes.
Steve LaMarca:
Okay.
Nikki Gonzales:
Yes and no.
Steve LaMarca:
Okay.
Nikki Gonzales:
So again, not wholesale, are we moving away from the operator having access to the HMI directly on the machine-
Steve LaMarca:
Sure. Sure.
Nikki Gonzales:
... I think it's-
Steve LaMarca:
Still need a big ready stop button.
Nikki Gonzales:
Yeah, that's really important. And in a lot of cases, you really want that local control of the machine. You don't want the lag that may exist in the machine, in the network. So certain things need to be absolutely instantaneous. You press that button on the HMI, and you can't have a few seconds lag.
What we're seeing though, is more interoperability and more flexibility between having a traditional HMI that can also connect to the web, that can also be accessible remotely. It can also be accessible on multiple other screens. It can also provide data up to a dashboard that's viewable in the control room. Maybe a shift supervisor now has access to 10 HMIs, of 10 different machines on their line, all on one screen, either on their laptop or on their phone or whatever device, a tablet.
There are some changes to where we're seeing... So we make web panels now that do nothing but serve up a web-based HMI that is part of our larger system. But I think it kind of depends on what type of factory you are and what you're making, what would be the best fit for that particular process. But HMIs nowadays, the smart ones, they contain the ability to serve up a webpage on a traditional HMI as well. So you can kind of have the best of both worlds.
Steve LaMarca:
Wow.
Nikki Gonzales:
It doesn't have to be one or the other.
Benjamin Moses:
And I do think you hit on a key element. So when we talk about a broad spectrum of devices that are trying to communicate, the importance of standards, and we kind of hit on that digital twin, and one of the things I was thinking about is artificial intelligence is the thing. We're kind of embracing it.
Nikki Gonzales:
Yeah.
Benjamin Moses:
But I think where manufacturers are today is kind of... The biggest thing that they can do is kind of prepare their data sets, prepare their business model for more stable AI applications. What are your thoughts on where we are today versus preparing for the future on artificial intelligence?
Nikki Gonzales:
I think we're getting a lot closer. I've been talking to companies in this space probably over the last two years. I've had a podcast in the industrial automation industry called the Automation Ladies for about three and a half years now. And I've stayed away from talking about AI, up until we have an AI panel scheduled next week finally. Because for me, I've actually been in the artificial, well, it's not an industry, but I also worked in the supply chain about a decade ago using machine learning and AI before ChatGPT and LLMs and everything came along.
And we had real high hopes back then that AI was going to wholesale, make a huge difference in the manufacturing industry. And 10 years later, it hasn't really. It looks really shiny when you start with it, and then really when it comes down to practicality, it's like most other technology, it requires very careful implementation. It requires very good inputs, it requires constant tuning, it requires ongoing maintenance and control and change and supervision. So it's not as much of a magic bullet as I think people thought. And even ChatGPT, right?
Benjamin Moses:
Yeah.
Nikki Gonzales:
It's a large language model. It's built to sound good. So again, to a novice, it could seem like it could solve all your problems, and it'll probably tell you that it can. But in a manufacturing setting, you absolutely can't have it do the wrong thing, or you can kill someone or you can cause a plant fire. I mean, there's a whole lot of things that in our industry are critical that in consumer industries or entertainment or services wouldn't matter. In our industry, it's critical.
But what we're seeing right now is there are companies getting value out of either projects they've piloted for the last five years. We've kind of talked about, for many, many years, this whole pilot purgatory of AI applications. Because once you get down to it and test it, they're often not as revolutionary as you thought. But I think we're getting... Right now, we're to a point where we're a lot more realistic about this. We have now some framework for implementation, because we've done these pilots that may or may not have given enough ROI to make sense to implement fully. But I think what's really come out of that is also the realization that you cannot do it without proper data, data from everything, data contextualized, data that's cleaned up, data that's in the same place. Because if you have an AI working on a siloed data set that doesn't see the other side of the operation, again, you're just going to get garbage output.
So I think it's really brought the conversation to a better place, a more practical place of, "Okay, let's take these steps, these building block steps, and then we need to upskill the entire industry to know how to handle AI." Not just the end users, it's the systems integrators as well that deal with all this equipment. AI is not going to wholesale replace the need for engineers and systems integrators to put everything together and babysit it as we go forward. In fact, I think we just have a bigger need for that.
So right now, I know a lot of the companies, even in the data and AI layer of this, are looking at, "How can we help upskill our partners to make them experts in this so that they can support customers going forward to getting real value out of AI?" And there's a lot of investment being done in that, but it's a team and a leadership thing as well. We got to support from the top of these big companies developing very, very cool technology, the Fortune 500 and large companies that are pilot this and that can invest in that. And then how do we, once we find success, build that out into let it trickle down to the long tail of the industry, which is the smaller manufacturers and the systems integrators that support them. And that, I think, is going on right now.
And a big part of that conversation, and I think one of the biggest pieces to its success going forward is this model of, how do we get the data, and how do we contextualize it properly? So a lot of conversations about UNS, or a unified namespace, or a way to actually create a single source of truth that is accessible and useful to AI. And then, I mean, even there in the standard side of things, when it comes to that, there's still debate about, what does UNS mean? How does it actually work? How do we contextualize the data properly so that we can do all of that without being tied to one specific vendor? And that's kind of all being worked on right now. I know Sesame has a new kind of API standard that they're asking people to take a look at. Can we as an industry come up with a standard way that we pass this data back and forth, and what does it mean? So that as we start to work at the higher levels of AI, we're not just reading garbage in and garbage out.
Travis Egan:
Yeah. AI is definitely going to have a tremendous impact on manufacturing. I mean, all industries for that matter. It already is, and it will continue. But it does create a host of new challenges at the same time. You talk a lot about data, data security in particular, you're putting information out there, intellectual property. There's a host of things that need to be considered in that process, and in manufacturing in particular with that.
But there's also the, "Will AI take my job" question. And I think the answer to that is similar to what we've seen with automation. There's a big fear that automation, robots, they take everybody jobs. And it really creates more opportunities. And I think AI will be a very similar output in the end, is that it will create more higher level opportunities for people in new and interesting jobs.
Benjamin Moses:
That's awesome. And the big reason we're here is because you guys, Nikki and Travis, are doing a lot of support for Hexagon Live that's coming up pretty soon. So Nikki, can you give us a little sample or tease us about what you have planned on the Inspired Innovators topic?
Nikki Gonzales:
Yeah. So I'm going to be moderating or introducing a whole host of sessions related to the simulation piece, which has always been kind of a passion of mine. And I think it's going to be really cool to see what all these innovators in the space, Hexagon customers and their applications areas really where they're sitting at today.
And then I am leading a very cool session about leadership really in this area. It's called Built to Lead, stories of innovation, resilience, and the future of manufacturing. And I'm going to be interviewing a number of female leaders. We're going to have a panel. And one of the things, so we want to attract more people into the manufacturing industry of all kinds, women, underrepresented people. I think the number... The technology that is coming into this industry and has been maturing should make it a more attractive place to work for everyone.
I am an eternal optimist, so my view of this has always been, being involved in automation for the last 20 years is, like Travis said, there is this opportunity for this to be better for all of us. There's an opportunity for it to bring more people into the industry. Part of that is they need to see the opportunity, they need to see the potential, and they need to see their representation, like, "Hey, there's somebody here that maybe looks like me or comes from a background similar to me, and I can do this." So I wanted to put together a panel of impactful women leaders, because that's part of what the representation I see that's made me feel comfortable growing in my career in this industry.
I think we also need more diversity in leadership, because it's a people thing. Leadership, at this point, yes, you have to support the technology, but how you handle it is going to be based on your style as a leader. Are you going to look at an ROI of an automation project and say, "I can replace three operators," or are you going to look at the ROI of, "How much more output can I get, and how can I upskill these operators?" And that in and of itself is just a leadership question. How do you see it as a leader, and where do you want to go with it? Because you can go one of two ways. You can do it to empower your organization and your people, or you can do it to try to replace them with automation or AI or whatever the technology de jour, of the day is. That's always been the question, "Do we sack the secretary now that I got a laptop, or do we put them somewhere else?"
And I have this growth mindset. I think that our economy, our world has the potential for everybody to grow and for the pie to get bigger. So I want to talk to leaders about how they see it, what they've done, what stories of resilience and of opportunity can they tell so that other leaders can start to see this as an opportunity and not treat it in the, "There's got to be a loser somewhere, so I'm going to have to lay off some people in order to justify this investment." And I try not to make women sit on stage with a panel saying they're women, because I think they have a lot more to speak to just about their experience than the, "Hey, I'm a woman in the industry," kind of a thing. But that's really what I'm focusing on with my session, and it really ties into the people of the technology and the representation.
Benjamin Moses:
Awesome. Thanks, Nikki. And Travis, you're doing quite a bit for Hexagon too. Can you give us a little sneak peek of what you got planned?
Travis Egan:
Yeah, this is a pretty exciting, Hexagon Live is a biannual event, and this one's global. So they're pulling a global audience in to Las Vegas for a several-day event. And my role with it will be to be somewhat of a host or an emcee for really the manufacturing track. Hexagon's a technology company. They're going to be really diving into a lot of the things we touched on today. How is AI affecting all of the technology outlets of Hexagon, whether it's in the safety area, manufacturing, lifestyles area, construction, building? They have all these different tracks. And specifically for manufacturing, how is automation impacting the growth and opportunities for Hexagon technologies, like data development, digital twin, robotics, things of that nature?
So we're going to be diving in, and as an adjunct to this, in helping to sort of develop out the concept, AMT is working with Hexagon on a new promotional thing. It's called The Architect. And it really is kind of, "What is the blueprint for success or innovation in manufacturing?" So that's kind of the thread that we're going to be pulling on and sort of the journey that we're going to be taking people on throughout this several day event.
And in leading up to it, I've had the opportunity just recently to go to Hendrick Motorsports. So this architect approach gave me opportunity to go to several of their customers and take a look at how they're really applying these technologies and create a video to sort of promote these concepts. And then ultimately, some of these folks will be there at Hexagon Live being interviewed, and we'll have an in-depth discussion about what was learned.
Benjamin Moses:
Thanks. Thanks, guys. Steve, let's hand it back to the podcast team, and then it was a great interview. Thanks, guys.
Steve LaMarca:
Yeah, thank you.
Travis Egan:
Thank you.
Steve LaMarca:
This was awesome.
Nikki Gonzales:
Thank you.
Benjamin Moses:
Yeah.
Ramia, that was a fun interview. It was really cool to have Nikki and Travis on the podcast talking about a bunch of different topics. We kicked off with leadership and innovation, particularly how the face of that changes in this rapidly changing technology. Then we talked about simulations and prototypes, digital twins, human and machine interfacing and controls. And that kind of all prepared us to, how do you prepare for AI in the future? So it was a really fun conversation. Ramia, can you tell us more? Can you tell us where listeners can find more info about us?
Ramia Lloyd:
I can. First, I would like to say thanks to this episode sponsor, Hexagon. Everyone go sign up for Hexagon Live. But you can find more information about us on techtrends.ATMonline.org. Like, share, subscribe.
Benjamin Moses:
Bye, everyone.
Ramia Lloyd:
Bing bong for Stephen.
