[00:00:00] Eric Cook: and welcome to Beyond Built. I'm Eric Cook, Tech Solutions Strategist at Accruent.
[00:00:05] Richard Leurig: And I am Richard Leurig, Chief Product and Technology Officer at Accurent.
[00:00:09] Eric Cook: On today's episode, we'll be talking with Todd Beckerdite about maintenance and reliability. He's the Senior Manager of Maintenance Foundation at Ajinomoto Foods and has been an expert in maintenance and operations for 35 years, so this is sure to be an excellent conversation. Todd, it's great to have you here.
[00:00:26] Todd Beckerdite: Thank you very much. It's great to be here.
[00:00:28] Eric Cook: Alright, listeners and viewers, we like to start every episode with a segment we call in the news, Richard, in our, in the news segment today, it's all about a topic that's been making a lot of headlines lately, it, AI and its impact on facilities management.
[00:00:41] I know this is an area that you follow closely. What developments have caught your eye recently?
[00:00:45] Richard Leurig: Yeah, I think the most interesting thing about AI and specifically what people are talking about now is the evolution of generative AI or what you hear referred to as Gen AI. And I think what's really emerging is how do you take this, this powerful set of capabilities and embed it seamlessly into a workflow, into a process, into an application.
[00:01:09] Such that the users are not even really aware that they're interacting with a, a generative AI model or generative AI technology. So it, it basically augments and improves what they're doing without them actually knowing or having to know anything special about interacting with it.
[00:01:28] Eric Cook: Do you think that that means that we're going to have applications and services and products out there that have Gen AI and we're not going to actually know that it's Gen AI or is everyone going to know and it's just going to be commonplace to interact with?
[00:01:46] Richard Leurig: I think it's gonna become so commonplace that you won't know you're really interacting with a gen AI model, um, or with something that's giving you gen AI or AI type responses. It'll just become part of the day to day. And in fact, I don't even believe that the, what you would refer to as the killer app, or the things that are really going to be evolving over the next two to three years, have actually even been developed yet.
[00:02:10] So, in a sense, the AI technology has outpaced our ability to incorporate it and integrate it directly in the day to day lives of facilities managers, of reliability managers, of, of people doing work, um, across the board.
[00:02:26] Eric Cook: Well, that's interesting. So one of the things, you know, cause I've played with like chat GPT and Google Gemini and, you know, things like that, Bing search. Um, and one thing that I've noticed is that it will often give me answers that aren't quite accurate. So what is it that we're going to need to do? Cause I think most of the people out there who work in maintenance or work in, uh, reliability and operations management, they're not necessarily experts in this AI technology and they're going to rely on the tools to be accurate. How do we keep them accurate?
[00:03:00] Richard Leurig: Well, what you're seeing in those chat GPT or the, you know, the GPT models is an evolution to towards accuracy that sits on top of a large amount of data and how it learns and understands the data. Absolutely. Today, you see what are known as hallucinations, the desire for chat GPT or the gen AI technology, to give you an answer even when it doesn't have one. It tries to generate one. So, what's very important when you're talking about facilities management, or let's say repairing an asset, is that you get it right. And so, I think at first, some of the things that you're going to see are more like your interactive sessions with ChatGPT. They're going to be suggestions. They're going to be helpers that help you use a product or an application in an easier way. They're going to be summaries of information or suggestions on what a repair technician. I think that will evolve over time, though, into really a more perfect science as these models and as the technology evolves.
[00:04:02] Eric Cook: Yeah, I can't imagine that anyone isn't going to want to adopt that sort of thing, because I can imagine being on a factory floor, having to go out and repair a machine, and then having all the details of the machine's maintenance manual be already fed into the AI so it can tell me how to fix the problem.
[00:04:20] If I say, you know, this belt has snapped, and here's a picture of it, I think people have tried it with things like the Microsoft, um, uh, HoloLens and things like that. But this is going to be a much more consumable product, right? I mean, it's going to be something that everyone can use.
[00:04:37] Richard Leurig: And then, and the nice thing about generative AI is it can take specifications or information from multiple places that are put into a single location and actually summarize the information in pieces and parts. So if you know what the symptoms are of a problem, it's likely it will find the information in various manuals and specifications that will help you guide you through what is the most likely cause of the problem. Obviously, what we want to get to eventually is
[00:05:05] more than that.
[00:05:07] Eric Cook: So I think this is maybe a great time to bring Todd into the conversation as well. So when we talk about AI and how that's going to affect not only maintenance, but ultimately reliability and the operations of organizations, how are you guys currently trying to adopt that or investigating it today?
[00:05:26] Todd Beckerdite: thought about that quite a bit. So the branch of the company that I work for, their AI is being used, but in very limited places and for very limited reasons. From a maintenance standpoint, you know, Richard was talking about a few things that are absolutely the thing that I would like to see from, from a, just general maintenance overall.
[00:05:52] So a couple of things is one, a decision tree, right? So a lot of maintenance manuals have things that, that both of you were talking about. There's, it's troubleshooting. If this is broken, look at this. If this is broken, look at this. AI ideally would help build a huge decision tree that says, okay, this piece of equipment isn't working.
[00:06:16] Is the power on? Yes, no. And then you go from there, right? And I think that's entirely possible and, and, and very probable. I'm hoping within the next five to 10 years, that becomes a very realistic thing on a production floor. I wouldn't be surprised if the Amazons of the world are using it, oil and gas, things like that.
[00:06:41] I'm kind of crossing my fingers that that comes because if you have a comprehensive tool like that, your downtime likely would reduce, right? So it takes a lot of the guesswork out. And from, from a lot of maintenance and engineering standpoints, guesswork is kind of the thing that takes the most amount of time, even if a part is difficult to change out. And if you can reduce that overall, then obviously you're going to reduce the amount of downtime. And I've also seen, I think it was even some, some years ago, is, is VR on a, on a platform. And I've seen that in oil and gas. Uh, I've also actually, so I see the military actually was starting to. Utilize that as well.
[00:07:32] I don't know how that would interface with AI as a reasoning voice that you could interact with, to Richard's point. I think that would be really powerful. The most work, I think will occur in bringing this to a realistic standpoint from maintenance is how do you pull old equipment into a new world, right?
[00:08:02] So, it all starts from the start, which is the manufacture of the equipment. So, if you have a piece of equipment that is being built right now, you know, are we going to reasonably expect that they are building a piece of equipment with in CAD or they're doing it in a 3D model where they can explode it in a 3D model as opposed to a, you know, black and white exploded view in a manual, which isn't bad, by the way, which is a good start.
[00:08:33] And then taking that, loading that into an AI database and building it out and then having that available for technicians. SoHowever it works, whether it's on a mobile device or, if we ever get to goggles or something wild like that, they have it and they start to learn how to use it and it becomes a powerful tool. The one thing that I, that I see a lot is technicians saying, AI can't turn a wrench. Yeah, AI can't turn a wrench. But I don't see AI taking people out of the equation when it comes to the physical maintenance and troubleshooting of a piece of equipment, um, at least anywhere in our near future. But I do think that it has some great merits and I think that the possibilities are there. Um, so. I don't know, I want to adopt it, and I think that Maintenance Connection in and of itself is a database, has the right setup to help start building that internally.
[00:09:43] Richard Leurig: that's very interesting, Todd. Um, and a very interesting take on AI and the other technologies and how they interplay. And when you broaden that out to the Internet of Things and the industrial Internet of Things and other automation, um, and how these may interplay with each other, not only to help once something is broken, but before it is broken. And how it may help to take elements of these different technologies and predict what's going to happen. How are you seeing that evolve over time from your perspective?
[00:10:14] Todd Beckerdite: That is, so that's a really good place that most maintenance departments want to get to, is as close as you can to true predictive maintenance. There are systems out there that can get you fairly close already. Uh, there are SCADA systems. Uh, I know that, uh, Accruance is actually working on some bolt on software for maintenance connections that I've been working with their developers on and saying, Hey, I think you could apply this more in some other places, uh, and, and broaden your scope.
[00:10:47] There are predictive technologies to your point, there's oil analysis, which still has to be done by, uh, hand is sent off to a lab, but there's vibration analysis. Uh, there's thermography, uh, there's even, uh, uh, what's the term I'm looking for? It is airborne, uh, vibration analysis, which is essentially like, uh, listening for air leaks in a facility.
[00:11:10] I feel like that's something that could happen in relatively short time. Uh, what I would love to see AI do is take the information. And build its own, you know, upper and lower limit. If you go above the upper limit, that triggers an alarm to say, hey, you need to go look at this, or below the lower limit, same thing. To where you're not setting them yourself, all you say is, build your baseline over this period of time, and then, give me what you think is appropriate for the specific technology in this specific place. There's already quite a bit of predictive technology out there. It kind of depends on the company and how advanced the company is.
[00:11:59] Aerospace obviously uses that as much as humanly possible, especially when you're talking about transporting humans anywhere. When you're talking about manufacturing, especially food products, some companies are good about it, some are, you know, I'll just say this out loud, slow to adopt doesn't mean they're not going to, they're just not sure how it works yet.
[00:12:22] I would personally love to see Ajinomoto get to that point. there's so much you can do just from connecting to the equipment itself through the Internet of Things. The other technology that, that is, and I will reference old equipment again, is can old equipment be connected to, you know, an external communication device that effectively provides information.
[00:12:50] If you're talking about a press that's 70 years old, can you do it? Sure. Is the company willing and able to invest? In the technology to bring that piece of equipment into the world that we live in today. It's, it's really up to them. But, I don't think that you would have a difficult time doing the ROI on that.
[00:13:14] It's normally pretty quick. I'd love to get there. I'd love to get there. It's been a, it's been a few years since I've been in that environment. I really think it's a lot of fun, because it, admittedly from someone that used to be a technician at one point, is it just kind of saves you a little bit of strain and heartache.
[00:13:32] You know, when you're a technician, when you're on the floor, when you have a wrench in your hand, all you want to do is fix it the right way. And you know, if you have more tools to help you do that the first time around, you get to leave work at the end of the day saying, ah. Knocked it out of the park today instead of saying, well, I hope that thing doesn't break down and don't call me at midnight. Uh, so there's a lot of opportunity there.
[00:13:59] Eric Cook: It's, it's interesting that you talk about, you know, IOT and especially even when you go beyond to older equipment, you know, because we've seen, sensors that are meant to bolt on, so to speak, you know, whether it's to do vibration detection, which is one of the more common ones or, or flow detection, you know, somewhere in the, in the system. And especially in the food manufacturing and oil and gas, as you've mentioned, right? Monitoring the flows of things and making sure that things aren't, aren't being slowed down anywhere. But one of the things that I always wonder about is how many sensors are out there that are already in people's environments that they're just not tapping into.
[00:14:39] Because if you have the data. You can learn a lot from it, right? Maybe there's a temperature sensor or you've got some sort of moisture sensor or something like that, and you look at, uh, a problem, right?
[00:14:51] And you say, you even have the AI or machine learning go and say, I wanted you to look at the data and tell me what, as you had mentioned, the high and the low points are, figure out what's normal so that you can alert me when something's abnormal. do you think you can also use that data at the, at the actual machine when you're making a repair to look at maybe the history, so you can make other decisions about how to repair it, how to replace it, and things like that. Do you think that those sort of things come into environments like yours, or is this something that is far off still for most companies?
[00:15:25] Todd Beckerdite: I don't feel like it's that far off. Um, we have equipment in our company that most of it's domestic built, but there is some equipment that comes from Japan. And it was built in Japan, shipped overseas to us, and And it's implemented and running. And a lot of that equipment doesn't necessarily live off of a processor.
[00:15:51] For example, it might be something as simple as relays. It's an on off relay. However, even from that one, that one relay, you can pull a lot of information, you know, on off, for example, or a start button, start, stop, just Those two data points can give you so much information over a 24 hour period. How long was power applied? Right? What time was power applied? How many times during that power on scenario was the start button and the stop button depressed, right? That starts to give you availability. That starts to give you reliability. I mean, just those two metrics alone are incredibly powerful. And you can also pull amp draw from a piece of equipment, for example, right?
[00:16:53] So if a piece of equipment starts to draw more power, there's a reasonable expectation that something in that piece of equipment is starting to degrade. So just those, those three things, for example. In any piece of equipment, nine times out of ten are already there. It's just how do you connect to those things.
[00:17:14] And I think that's very reasonable. There are some companies out there that are starting to realize that marketing to manufacturing equipment or manufacturing companies saying, hey, it doesn't matter how old this piece of equipment is, We've got this little thing. It costs you 300, but if you plug in here and plug in here, it connects wirelessly to your Wi Fi and now you've got all these data points to pull.
[00:17:41] And even some of them are so hungry to get their foot in the door that they'll like try it out for a period of time, things of that nature. That's the sweet spot right now for those older pieces of equipment is saying, is there something out there that can connect? And without having a human write down, well I started the equipment this time, I stopped the equipment this time, and this thing happened when I stopped it. Nothing against humans, by the way, because I'm one of them. There's bound to be errors. In that scenario, and an error, you know, the quality of your data in will equal the quality of your data out.
[00:18:26] So when it comes to that kind of information, you want it as accurate and as unbiased as possible.
[00:18:33] Richard Leurig: I think what's interesting, first of all, you're getting me very excited about talking about data and data and gathering data, right?
[00:18:39] Todd Beckerdite: I love data. Data is king.
[00:18:41] Richard Leurig: Yeah, there's so much that you can do with data. And I think what your, your example illustrates, though, is you can also take those data points and string them together to look at trends and correlations.
[00:18:53] You can determine where efficiencies may be gained or, to the point we were talking about earlier, where reliability issues may be, uh, coming up. You can also look at things like energy consumption, and it bridges the gap over to sustainability and energy usage and some of the things that a lot of companies are struggling with, right?
[00:19:16] And I'm just curious from your perspective that sustainability has been a big initiative globally, energy usage, which can be gained by looking at everything that you just talked about, combined with other sensors and other, other meter information. Do you see these things working in concert with one another or working against one another?
[00:19:37] In other words, gaining efficiencies, looking at predictive maintenance and reliability of an asset, plus how much energy is it consuming and what is kind of the footprint of the environment or manufacturing facility. Do you see those things working together or as separate initiatives from your perspective and from your company?
[00:19:57] Todd Beckerdite: I do think that they work hand in hand, but I'm not sure how many different entities consider those interlinked. So, for example, in, in the environments that I work in, I've got a lot of pieces of equipment. There's a lot of conveyors, freezers, steamers, you know, fairly large pieces of equipment.
[00:20:20] Anything on the grounds of a facility should be considered energy usage, right? If a facility is not operating and it's not in a wash down environment, or it's not in a, you know, we're shut down, but doing maintenance environment, now you start to think about, okay, well we want to reduce our energy footprint.
[00:20:44] How do you do that? Shut off pieces of equipment if you're not using it.
[00:20:49] to Richard's point, now you're just, you're pulling more juice, right, and companies have to pay for power. And when you get to that point, now you start to say, man, our utility bills last year were way higher than the year before.
[00:21:06] What happens? Well, you can start to look at downtime as a whole over the plan. You can start to look at availability. You start to look at, well, did we leave the lights on even though everybody went home? What is, what's going on there? I know for my company this year, we've got some pretty large sustainability initiatives going on.
[00:21:28] And that is one thing that we are looking at separate from mechanical breakdowns and things of that nature. And, you know, Ajinomoto as a whole is very big on having a small footprint from an energy usage standpoint across all of our facilities around the world. And I think they've got some pretty, I'm not going to say aggressive, but I think that they're, they're decent initiatives this year. , looking at things like natural gas, power, um, you know, We use trucking. How can you reduce the amount of trucking that you have? Because trucks burn fuel and A, B, C, and D, right? So there's a lot of different things that, that come into play. And I very much believe, to Richard's point, that they are intertwined.
[00:22:25] They shake hands constantly, whether people realize it or not. And that's going to be a very interesting thing that comes up this year. When equipment works as it's supposed to, it should, you should have a reasonable expectation that it's going to bring down your utility bills. When equipment does not work well, you should have a reasonable expectation that your utility bills are going to go up.
[00:22:51] All of these things are intertwined where you say, now you've got AI that's helping you troubleshoot and bring that downtime down. You should reasonably expect to see your utility bills come down. So if you're not spending as much, can you take some of that business profit and pump it back into the company to better improve technology and things of that nature?
[00:23:13] You're constantly trying to get more efficient and then you take some of that efficiency dollars and pump it back in to be more efficient, more efficient, it takes a while to get there, but it can happen.
[00:23:24] Eric Cook: You mentioned something about, um, the sustainability initiatives that Ajinomoto has. So, how is that coming downstream to you? So, for example, is it affecting not only energy requirements, which, which you've noted, do you have to do any sort of upward reporting on that?
[00:23:41] Or does it at least require reporting on equipment failures and replacements? Because obviously, when we talk about the environment, we usually think of energy usage as being the primary, but when you're talking about very large manufacturing environments, machine replacements can create a great deal of non recyclable material as well.
[00:24:03] Todd Beckerdite: You bring up a great point, so that is part of the initiative is saying, We recycle materials, right? We try to recycle as much as we can. We have what's called waste solids that come out of the plant. So that's waste, you know, food waste solids. There is waste water. Uh, there are packaging materials.
[00:24:27] What do you do with eight? You know, you, you want to recycle it instead of throwing it away, and one of the things they're saying is, well, start looking at just these areas, for example, and saying, do you have to have someone come pick it up once a week?
[00:24:40] Can you wait every other week? Can you wait every third week? If you're not getting it picked up as much, how do you reduce the amount that you're causing, right? So how do you reduce waste water? How do you reduce food solids? And those are, you know, when you're talking about a food manufacturing facility, those are very realistic things.
[00:25:03] Even, even if you're producing just tiny, tiny amounts, they still have to be dealt with. So that is a very big part of what we're doing this year. And we're looking at, we're actually looking at utility bills, uh, different things, negotiating with local utilities to see if they can. You know say, hey, you know, we're a pretty decent user.
[00:25:27] It all comes down to What can we do to be better without compromising what we do on a daily basis?
[00:25:35] Eric Cook: That's interesting. So Richard, when we talk to customers about, um, sustainability, especially, um, there's a lot of regulatory requirements around that, and those are becoming more and more stringent in different places around the world. What is it that? What is it that you think about, when you try to figure out how to help solve that problem?
[00:25:58] Because I know it's something that you work on quite a bit. What is it that you actually think about, uh, as part of your process there, um, when it comes to specifically regulatory requirements?
[00:26:09] Richard Leurig: You're absolutely right. Globally, there's huge initiatives and huge mandates and regulatory reporting requirements that exist, but they exist country by country. Uh, the EU, the UK, uh, the US, other countries have different standards and mandates, right? So, from our perspective, the key thing is trying to gather all of the data and the information to help us inform that reporting. Um, typically what you require is a lot of reporting around things that we've already talked about. Energy consumption, overuse of energy, um, That ties in directly though, and what's very interesting about it, not to take it too far off of that question, but what's interesting is, If a refrigeration door in a retail store is open, if there's an air leak, um, in a compressor that we were just talking about, these things draw huge amounts of energy when undetected.
[00:27:04] But the other side effect of that is the reliability or the issue on degradation of some kind of service being performed. Todd earlier said looking at lower tolerances and upper tolerances. of how equipment is functioning, how the plant is functioning to produce whatever product is being produced in that plant.
[00:27:24] In a, in a retail store or other places, it's storage of, you know, perishable items and things like that. In pharmaceuticals, it's, it's big perishable items, right? But at the end of the day, what's driving all of this is how is the energy being consumed.
[00:27:41] You know, we worked with companies where we took data and information off of the temperatures outside, inside, and in different places within the environment, and actually did proactive alterations of temperatures to reduce energy footprint and usage in summers. Where it was very hot in certain locations because they were not only trying to meet their ESG or their sustainability mandates, and their regulatory reporting requirements, but there's a cost efficiency to it.
[00:28:10] So I actually see these things all tightly bound together, which is why I think in a certain respect, replying and providing the data for regulatory reporting actually benefits most companies because it actually improves. Reliability of the, of the equipment and the assets, and it improves the efficiency.
[00:28:32] So these things are, are really tied together. You know, obviously, energy costs have gone through the roof, especially in the last couple of years. And so, there's more to it than just the regulatory reporting requirements. People think of that as a cost burden, but actually the data that we're gathering Uh, the data that I see being gathered is really helping on all aspects of running the business.
[00:28:56] Eric Cook: And I think that, you know, Todd, you've definitely pointed that out, that, that there is the, the ability to reinvest those efficiency gains into the business itself to make things more, not even more, not even just more efficient, but it also to maybe work in innovative ways you, you haven't before. So we, we find that quite interesting.
[00:29:16] Just a second. We're going to do a rapid fire question round, Todd, with you, if that's alright. Richard, I don't know if you want to jump in on some of these as well. But we're going to ask some questions. What I want to kind of understand is what your take is on a few things.
[00:29:30] But we start off with a question that we, we like to ask people. What's your favorite innovation and why?
[00:29:37] Todd Beckerdite: My favorite innovation The advent of the CMMS and SCADA type systems. So, there may be people out there that don't remember, uh, that not that long ago, a computerized maintenance management system was, was a futuristic thought. And most, medium to large size companies have them in place. Now you have a database that, that essentially, in some aspects, tells you what you need to do, and in some aspects, at the very least, takes information from others and informs you that someone, someone is saying that there's something that needs assistance and needs a set of eyes. Um, I just think that that's phenomenal and I am a massive fan of SCADA systems.
[00:30:25] So for, for those out there listening that don't know what SCADA is, it's Supervisory Controlled Data Acquisition. Uh, so essentially it's a, a database that takes some predictive technologies
[00:30:37] and it will alert you. So it's not, it's, it's not waiting for you to go see something. It's telling you, this piece of equipment's telling us something's wrong. Let's go look at it. The ability to interface with them and use them with upcoming AI, I think that's just, again, from a maintenance standpoint, kind of nerdy. I think that that's really just kind of a fun little pool to jump into.
[00:31:03] Eric Cook: Richard, you got one?
[00:31:04] Richard Leurig: Well, my favorite innovation probably isn't in our workplace. I mean, I, I think, um, I used to think, why do I need a cell phone, a mobile device that actually does email and everything else on it? Um, I'm gonna have to go with, I'm gonna have to go with that. A close second to that is the PC itself. You know, the origination of, of PC followed by Windows and Apple and all of the things that came from that. Obviously completely transformed things.
[00:31:33] Eric Cook: So I assume you're not, you're not one of the people who's going to go in on this new trend of dumb phones.
[00:31:38] Richard Leurig: I'm not going to dump phones. What I want to dump is tech. So I'm going to jump to, what's the innovation I wish would go away? Too many text messages, too many Teams messages, and too many emails. So my, my uh, side career at home is trying to come up with, My own AI product that helps me sift through all of the information that I receive on a daily basis and tells me what only what I need to know.
[00:32:03] Eric Cook: I love, I love that. I wish I had an AI that would respond to every time I have a cousin who has a birthday. Just says, happy birthday, cousin. It's not that I don't want to respond. I just, I just don't have enough hours in the day. So that's great. What about you, Todd? What's an innovation, just in general again, that you just wish would go away?
[00:32:20] Todd Beckerdite: Oh, I, as much as I love phones, I sometimes wish that they would go away. Just because I, I think that we as human beings have become leashed to this little metal square in our pockets. I think it's very handy. Very powerful. I mean, to, to Richard's point, you know, you can get text messages and email, and you can chat with someone.
[00:32:45] I can be on the other side of the world. And someone's like, Hey, I need a, I need help with this. Okay, hold on just a second, blah, blah, blah. You can jump on a team's message and literally see what's on their screen. I think that's very powerful. I also think that we, just as a, just human beings, we haven't taught ourselves what's good and what's not necessary, I guess you could say.
[00:33:11] Oh,
[00:33:14] Eric Cook: one that I, I, I thought of actually while we were talking about the, in the news segment. So, you guys have probably seen that, uh, OpenAI's, uh, chat GPT has a voice mode now, right? Where you can do a bot chat and that apparently it sounds like Scarlett Johansson.
[00:33:30] Um, so my question is to you and Todd, I'll get yours first, if that's all right. Um. you got to choose your AI's voice, whose voice is it?
[00:33:43] Todd Beckerdite: I would have to find something that obviously doesn't drive me crazy. Let me ask you this, can I change the voice depending on the scenario?
[00:33:55] Eric Cook: Sure, why not?
[00:33:56] Todd Beckerdite: Okay, so if it was AI giving me directions in my car, for example. Maybe I'd want something like Hulk Hogan, for example, like, you know, take a left here brother. Uh, but then also if it's, if, if I'm in the office, I'm probably going to want something a little more soothing, a little more, uh, personable.
[00:34:20] like Dame Judi Dench or, uh, Meryl Streep. You know what I'm saying? Something a little more, a little more mature, a little less, uh, You know, let's jump on it right now.
[00:34:33] Eric Cook: Okay. Richard, how about you?
[00:34:36] Richard Leurig: I would either want a voice that is, um, that you think of like as an authoritative voice. Type of voice that is, you know, not too high pitched or low pitched, but I don't know, maybe Winston Churchill or, the old time actors and actresses, you know, the Cary grants, the Frank Sinatra singers, like, you know, they just had, they, they didn't try to, to, to, you know, create voice patterns that were just the same across the board.
[00:35:06] Eric Cook: So when I'm, when, when I think about that, um, I kind of go probably with a little more, uh, pop culture reference and a little more, uh, uh, funny. Um, I want Ryan Reynolds as Deadpool as my AI voice.
[00:35:22] Todd Beckerdite: as long as it's not out loud.
[00:35:23] Eric Cook: He's going to, he's, he's, he's definitely going to have to get bleeped a lot though, but that's all right, but that's all right.
[00:35:28] Todd Beckerdite: yeah. I,
[00:35:29] Richard Leurig: a lot of, I could see that. I could see Ryan Reynolds, um, talking about Rexum and then interspersing like what he wants me to do. I
[00:35:40] Todd Beckerdite: I could get on board with Sinatra. That's a good one. Or even, uh, now you brought that up, even like a Johnny Carson as an AI
[00:35:48] Eric Cook: yeah, yeah,
[00:35:48] Todd Beckerdite: Because there's plenty of actual voice there where you could build an AI, an AI off of all of his Tonight Show rhetoric. Let
[00:35:57] Eric Cook: That's true. That's true. And they've got voice training now. So last thing I want to ask you, Todd, before we, before we wrap up here, what's one thing that we wouldn't know by looking at your LinkedIn profile?
[00:36:07] I like to travel a lot.
[00:36:10] Todd Beckerdite: When I went in the military and I got to travel around the world. I have been around the world at all 360 degrees of the globe in one form or another. And for whatever reason, that kind of sparked something in me
[00:36:25] The perfect opportunity came up. I would want to work overseas for just a handful of years. I'm not sure about my significant other, uh, whether or not she'd be on board with it, but, uh, I just think that would be so interesting. It would really kind of force me out of my, my space. Um, even, you know, especially if it was in another English speaking country.
[00:36:51] Still, I just think that that would be such a, a, a, a cool thing to help broaden me as a person even more.
[00:37:00] Eric Cook: Well, as an American who lives and works in the UK, um, I, I can highly recommend it. It's, it's amazing. And just being able to access other things that aren't necessarily available, uh, quickly in North America, like, you know, it really is just a 45 minute flight. If I want to fly to Paris, right. I mean, it's having access to those sorts of opportunities is also, you know, really cool.
[00:37:25] So Richard, I think. I want to sort of wrap up with you and maybe get your final thoughts on what we've been talking about, especially when it comes to reliability, predictability, AI. What do you think that the takeaway is from today's conversation?
[00:37:42] Richard Leurig: There are things that are available today. You can put a sensor on a switch and tell whether it's on or off, whether it's started or stopped. You can look at a lot of different energy and a lot of different data and pull it together. Um, clearly there's a pathway to better predictive maintenance and reliability through generative AI and AI capabilities.
[00:38:05] But I don't believe all companies are there, and I think That as companies are assessing older equipment, newer equipment, and how they want to interoperate with all of this. I think the real takeaway from all of this is, this is moving at a very rapid pace. Things are evolving quickly. And I think what we're going to see is that killer app. In other words, the thing that's going to revolutionize or change how manufacturing, how plants operate, isn't here yet. But I think the pieces and the foundation of that are being put together.
[00:38:39] Eric Cook: I think that's incredible. And I think, you know, as, as part of a company who creates technology to solve those sorts of problems, it's a very exciting time for us, but it's also a very hectic time because, uh, everybody wants to go fast now. And as you said, the technologies are emerging, but no one's put all the pieces together.
[00:39:02] But I think it's a really exciting time as both someone who's spent most of his life as a technologist and someone who's really interested in how that's going to impact companies, like a Ajinomoto and how we're going to be able to make a difference for them without also losing the human element.
[00:39:21] Cause that's always something that I, that I keep in mind is that the human element is the most important because at the end of the day, Todd said it best. We are all humans and we have to be able to work with technology and not let technology rule over us. So, thanks again, Todd. I really appreciate your contribution today and thanks so much for coming along.
[00:39:41] Thank you, Richard. We want to continue to have these sort of conversations and I really want to thank our sponsor, Accruent. Learn more at accruent. com and, uh, we'll see you later. Thank you.
.svg){kind=small}
