NDE 4.0 Podcast | Transcript | Dave Seto | Episode 9

NDE 4.0 Podcast Transcript

Episode 9 — Adopting the NDE 4.0 Mindset for Better Data and Decision Making — Dave Seto, Business Development Manager at UTEX Scientific Instruments

Floodlight Software: [00:00:00] Welcome to the NDE 4.0 Podcast, where we ask five questions for an NDE or NDT expert. This is the show for NDT and industrial inspection professionals, where we dig into the big questions about NDE inspections and digital transformation. Every episode we ask an NDT expert five questions that can help you do your job better.

Nasrin Azari: [00:00:27] Today we are honored to be speaking with Dave Seto. Dave is the business development manager for UTEX Scientific Instruments, based in Mississauga, Ontario, Canada. Dave has presented and published on the topics of fermentation of unconventional biomass, distillation as a means of waste reduction, and most recently on inspection automation for NDT.

During his career with UTEX, Dave designed novel ultrasonic probes, engineered automated scanning systems, and has developed customized software for automating inspections in metals production, aerospace, and the nuclear power generation industries. Dave has a bachelor’s degree in chemical engineering from the University of Waterloo, and he enjoys wholesome obsessions with bicycles and microbrewed beer.

Welcome, Dave to Floodlight Software’s NDE 4.0 Podcast series. And thank you for participating in our program.

Dave Seto: [00:01:22] Thank you for having me. It’s going to be fun.

Nasrin Azari: [00:01:26] Yes, it most definitely will. Our podcast poses five questions to an NDE 4.0 expert, and we focus those questions on the specific expertise of our guests. Dave, your company helps inspection companies automate their inspections with scanning systems, software, and high-performance ultrasonic instruments. To help our audience better understand what you do, let’s start with this introductory question. Our question one today is how do you define and how does your company contribute to the NDE 4.0 Initiative?

Dave Seto: [00:02:01] Actually, we see it a lot like some of your previous speakers like Johannes Vrana and Ripi Singh. We see it in two ways, one of them is this convergence of several technologies that enable the integration of NDE processes into automated manufacturing as a whole, and making the information that you get from an NDE process, more widely available throughout the organization. And then, as a fallout consequence, that maximizes the efficiency of the NDE process itself.

And the second way that we see it is as a way of thinking about NDE as an integral part of manufacturing, as opposed to something that happens after manufacturing, it happens to the product after you’re done making it. We see it as a more integrated whole. So we like to think of NDE 4.0 as just a shorthand way of saying that you’re taking the next logical step of making the data that you generate more available.

If you think about the sort of history of nondestructive testing, version 1.0 was really just test all the parts and throw away the bad ones. And then, in those terms, nothing was saved. You didn’t really have any data from those processes. It was just a go or no go decision. And that still happens today. There are still some test methods that do that.

And then you started to see paper CT scans in the sixties and seventies. So he could call that NDT 2.0. But again, that information is on paper and it’s buried in a filing cabinet somewhere, so it’s not really accessible and it’s hard to use it.

And then if you think about it another step would be NDE 3.0, and we think of that as the beginning of the digital age. Here, you had digital CT scans, you had images you could share, files if it was an X-Ray, for example. But even then they aren’t really that available.

They’re on a hard drive somewhere. And the hard drive is probably on a shelf because the computer that hosted that has died and it ran on Windows 3.1 or something like that. So to our way of thinking, it is really just taking the next logical step, making that information more widely available, making better use of it.

And, as a result, you’re gonna get feedback into your manufacturing processes that will make all of it more efficient.

Nasrin Azari: [00:04:18] Interesting. So is there a specific problem that your customers come to you to try to fix? And is there some specific benefit or problem they’re trying to solve or benefit they’re trying to achieve?

Dave Seto: [00:04:32] Yeah, it often has to do with the automation. We’re among the first companies, if not the first, to develop a Windows-based inspection system. And back when Windows 3.1 was a hot technology, we were developing our wind spec platform to automate the inspection of pressure tubes and in nuclear reactors.

And, the system actually ran with both. Eddy current and ultrasound and a couple of other sensing technologies gathering data all at the same time. So customers come to us for that kind of thing. There, they need to speed up the process or they need to maximize the efficiency of the process by adding things like Eddy current and ultrasound.

And we’ve been doing that for probably 15 years or 20 years. And it’s only, now that you’re starting to see those kinds of techniques talked about where companies who make instruments, for example, they tend to have very instruments, specific software. So you may be able to automate the collection of the data.

We like to think of that as motorized or mechanizing the data acquisition, but that’s not really what automation is. There’s a whole frontend to that. And then you get the data and then there’s a whole backend to that.

And, that’s generally what we do. So, since the 1990s, we’ve been providing people ways to better make use of their NDE data. And we’ve sometimes felt like we’re crying in the wilderness.

Probably just that people didn’t know what they were going to do with it. If they collected that kind of information, how would they make use of it? And that’s really where we try to be at the forefront, is that is the glue between the people who understand automation, that people who understand non-destructive test instruments and techniques, and we’re the bridge between those two worlds, which makes us, I think, inherently an NDT 4.0 I guess evangelist would be the right word.

Nasrin Azari: [00:06:30] Yeah. It’s a great feeling to work with companies and help them become more productive and effective. It’s one of my passions as well. Which moves us to a nice segue into our question number two, which is what do you see as the biggest potential benefits to NDE 4.0? And maybe, a little thought about how that looks in the short term and how that looks in the longterm.

Dave Seto: [00:06:55] Okay. so you know, the way we think of it, the whole point of doing nondestructive testing is to make better products. They need to be safer. They need to perform to higher standards. And that’s why you see all of the advanced research happening in places like nuclear power plants and aerospace and space launch companies and things like that.

So the whole point is to make a better product. So then the information that you get from the nondestructive testing can be used to help you improve your raw materials, improve your manufacturing processes, achieve higher performance, and, reduce the weight or increase the lifespan or whatever it is that, makes that product better.

We see that as the most obvious advantage. The sort of less obvious or follow on advantage of NDE 4.0 is having that data available. So if you think about societies in general, companies and even civil societies, they work better when there’s a free flow of information and free flow of resources, things like capital.

And so for us, that information from an NDE process is really just a resource. And if you can free it up and make it mobile. it’s almost hard to predict the number of benefits that you could get from having access to that. So improving the manufacturing cycle and improving your products and your raw materials is just, that’s the obvious one that we’ve always been using NDE for sure.

The following benefit is enabling that asset, that information, to move more and make it more accessible to more people. And then that information can be, in a simple case, you can data mine that information and get the statistical analysis done. What do you do with the stats? How do you improve, whatever it is?

Nasrin Azari: [00:08:50] And I’m sure that some of those benefits are things that we haven’t even thought about today.

Dave Seto: [00:08:54] Exactly right. So many technologies you can point out and in the early days of, prior to their adoption, people go, why would we need that?

Why, why would we need super thin, flexible, nearly shatterproof glass? I don’t know, maybe in 50 years when we have cell phones, that might be a useful thing. Why would anybody want a camera on a cell phone? Thinking about that back in the day.

I was one of those Luddites. I thought that the camera should be a camera and it was a bit of a purist about that. And the cell phone needed to have a keyboard because I’m a Canadian and Blackberry was our pride and joy in the tech world. And I’ve learned to use the cell phone for all kinds of things that I never thought I would, including reading serial numbers on motors that are buried upside down in a machine that I can’t get to with my, my plus 2.0 glasses. I’m never going to see that, but I can take a picture of it with my cell phone and then expand it and go, “Oh, okay. I know what that is.” So yeah, it’s amazing.

Nasrin Azari: [00:09:59] And I agree with you. I think there are definitely these obvious benefits, but then there are these hidden benefits that we try to think about. Obviously, safety is huge. And it’s the longer-term benefits that I feel like we can’t really even quantify today. Because we don’t know what, how technology will evolve, and what, how things will start working together. So I think there’s definitely some mystery there.

Dave Seto: [00:10:23] Yeah. And what, like one of the things that you can predict, you can imagine what would be making the information available in real-time. Factory MRP systems and ERP systems make products available that are lower costs and higher value because the productivity of the whole system goes up when you have real-time feedback.
NDE tends not to do that. It tends to be a separate process. So the information that would be available to you in real-time, if you were to adopt automation in the NDE process could improve the productivity of whatever it is, whether you’re making jets parts or airplanes, or, widgets that need some sort of safety component on them.

So if you think about it that way, there’s a cycle there of information from the back end of the process, feeding back into the front end of the process. Even just from a productivity standpoint, if at all times, which parts are in production, which parts are under inspection, whether or not you’re rejecting any parts through the NDE process.

Wouldn’t that be good to know, like instead of having parts in a box somewhere across town, waiting to be tested, if you knew at the end of your production line that, suddenly 5 or 10 percent of your parts are being rejected, that, that’s a huge benefit to you. And so one of the knock-on benefits you can imagine from that is that NDE becomes cheaper. It becomes lower costs because you’ve automated it. So now maybe you can be doing NDE on products and materials and objects that you weren’t doing it on before. So now you have the potential to offer a competitive advantage, right? Do a hundred percent inspection of so many things like semiconductors and pick a product, cell phones, there’s 100% inspection done on them in some way, at some point in their process.

So if it’s a, if it’s a component that could benefit from having NDE, it’s too expensive or no one’s implemented it because you have to put the parts in a box and ship them across town to an inspection facility and then wait for them to come back. That’s a fairly big speed bumping in your efforts to improve your product quality testing.

But if you could automate it and it can be right at the end of the production line, you could do a hundred percent testing. That would be a really useful thing. Think about something like bearings, for example. We make bearings probably by the billions. Millions and millions of bearings go out the door and the big ones like that end up in wind turbines and ship propeller shafts and pumps for nuclear power plants. When those get inspected, that’s part of what makes them expensive. If you could reduce the cost of the inspection, you could apply the same non-destructive test techniques to lower cost bearings, ones that ended up in cars or ones that ended up in smaller applications where, if you’re testing the bearings a hundred percent, maybe you could increase the lifespan. That’s a competitive advantage, right? Our product lasts twice as long as the competitor’s product. So yeah, there’s all kinds of follow on benefits to being able to automate your NDT.

Nasrin Azari: [00:13:26] Let’s talk about the other side of the coin. The benefits are there. Certainly, there’s a process to get there, but what do you see as the biggest obstacles to end success for companies?

Dave Seto: [00:13:37] I think like a lot of things, the biggest obstacle is a failure of imagination. You can’t invest in something if you don’t understand the potential benefits. And conversely, so if you can imagine the sort of happy future that occurs when you make these investments and improve these capabilities, conversely, you probably can’t imagine the downside if you don’t invest.

So either way, the failure to exercise imagination is probably the biggest barrier. A good example of this is, we’ve been talking a little bit about quality cycles. A good example is W. Edwards Deming. I don’t know how many people know about him, but, he was the quality guru who went to Japan in the 1950s.

And was invited to stay by the Japanese companies who were trying to rebuild their economy. So he taught them about statistical quality control and ways of thinking about product quality cycles, and, using control charts and things like that. And a way of thinking about the whole process of raw materials to production, to finished products.

And he was largely ignored in the United States. So he spent 30 years in Japan from the 1950s to the 1980s. And in the 1980s, American cars were being outsold by Japanese cars. American electronic companies were going bankrupt because people were buying Sony, and that didn’t happen overnight. Japanese companies spent 30 years rebuilding and figuring out how to improve their processes. And then in the eighties, especially in the automobile industry is the big one, the automobile industry woke up and realized, “oh, my goodness. We’re losing market share in a big way. What are we going to do?” And they began to engage Deming and, obviously he’d had many students over the years, so they very quickly had to pivot, invest hugely, start improving their processes and start improving their management-labor relations. That was a big part of what Deming taught was that we’re all involved in making it a really high-quality product. And there shouldn’t be any internal strife between those who manufacture and those who test and those who manage and those who execute the manufacturing process.

The American companies didn’t really imagine, in the 1950s and 1960s, that they would have any competition. And then all of a sudden in the 1980s they were given a rude awakening. So I think the same thing is going to happen with NDE 4.0. Forward-thinking companies realize, “Hey, this data is useful. We can do things with it immediately. And we can use the data as an asset that we can mine for all kinds of things that maybe we don’t even understand yet.”

A good example is everybody’s talking about A I. Nobody’s ready for it. We’re all just talking about it. But nobody’s actually ready for it. So that’s the first big obstacle is a sort of failure of imagination. And then I think right after that, once you begin to imagine that such a thing might be possible and that others might be doing it, there’s a fear component. I think that is the next biggest obstacle. We live in a society where people are pushing back, at times fairly hard, against big data. Privacy is a concern. Security is a concern. Making your NDE data more accessible. This is probably a scary concept. A lot of people think, “what happens when I lose control of the data?”

It is a valid concern and it can be managed. There are already technologies in place for it. Security so that those who are supposed to have access to, for example, just business, transaction data, If you’re not supposed to have access to it, there are ways of keeping you out.

And there are ways of making that information move freely within an organization and between organizations. So we’re already doing it. And another concern that a colleague of mine brought up when I was telling him that I would be chatting with you guys about this, liability has been a concern as well.

We actually have a couple of customers who asked us to automate their system so that you could render an analysis and render a judgment on the suitability of the component and then give it a go/no-go decision and then lose the data because they were worried about product liability concerns. For example, 10 years from now, if there’s a liability claim against the company because the component has some sort of failure. Firestone tires are good examples, right? Firestone in the 1990s was sued and their relationship with Ford fell apart over tires delaminating on the road and causing vehicle crashes. Companies, especially if they have ties to the automotive industry are very concerned about retroactively being judged for stuff that is a defect that they couldn’t detect 10 years ago.

And that’s a valid concern. However, I agree with, for example, Johannes and Ripi Singh, have talked about similar issues. If your NDE data is robust and secure, and you have processes in place that document how it was collected and how it was analyzed and how the decision that was made, that this is a suitable component or not a suitable component.

Then that shows due diligence. Ten years from now, we may be using different techniques. We maybe have tighter tolerances. That’s how all manufacturing works. All manufacturing. If you’re improving your manufacturing process, it works by testing what you have and making the tolerances tighter, right?

Cars, for example, the fit between the doors and the windows and all that stuff is way better than it was in the 1960s and 70s. Like a car is a quiet thing. Now at a hundred kilometers an hour, cars used to be noisy and rapidly and horrible. But they got better because people test them and they figure out a better technique and they improve it.

And then they test again and they figure out a better technique and they improve that. And NDE is no different. So we shouldn’t fear the future. We should exercise appropriate diligence, document what we’re testing, how we’re using it. What are the standards, for acceptance or rejection?

And then the data becomes its own validation of your efforts as a company to produce a high-quality product. And then if there is ever a liability claim, if you just do a quick Google search on liability claims the things that damage companies the most is the attempt to manage it as a PR exercise and not an engineering exercise.

So as soon as you start yeah. And you see it a lot. Yeah. Especially in the automotive industry. I don’t know why, but there seems to be a mini-industry of people who like to sue car companies. But, if a company is deemed to be trying to manage the story from a PR perspective, immediately people begin to get suspicious.

That’s worse for the company attempting to hide information from the public or hide information from regulators or downplay the seriousness of what could be a quality problem is a, is usually way more damaging to a company than, Just acknowledging that. Yep. we’re seeing a statistical trend here and we think it may be a quality-related, problem. And here are the steps we’re taking to fix it.

Nasrin Azari: [00:20:52] That’s a great observation. I think the car companies are concerned that people will stop buying their cars and go buy their competitors cars immediately. And that’s a big problem for their shareholders. So yeah, I can definitely see that sort of that liability.

We’ve talked about that with a couple of our other guests, too, about the fact is that if you’re using more advanced techniques, you’ve got the ability to detect more problems, which means you have the responsibility to fix more problems than you might have found before. So that’s definitely a really interesting component of this whole discussion of NDE 4.0 and how quickly can we get there? But, I know it’s not a process that happens overnight. Let’s move to our question three for you today, which is, can you describe a practical approach for companies that are ready to embrace NDE 4.0 or understand its importance and are ready to start moving that way. How might they progress from where they are today to a fully NDE 4.0 enabled environment?

Dave Seto: [00:21:52] Yeah, that’s actually a very important word, progress, because it’s, it is not a step-change from one state to another. You don’t suddenly become NDE 4.0 capable. I think it’s a process and you adopt principles and ways of thinking and begin to move forward in a certain direction. That’s probably the first step is start to thinking about as an integral part of your manufacturing process, all the way through the organization, from raw materials to finished products, and has the potential to dramatically improve all of those steps.

And then, as a lot of your customers are I’m sure aware of the life cycle after you’ve delivered the product, you’ve got a pipeline or a tank or a jet engine or some other asset that needs to have a long life because it was expensive to produce. And NDE is a big part of predicting how long you need to go between inspection cycles. How long you need to go between, maintenance and repair cycles, all that kind of stuff. yeah, the first step is to think about it as a person, and you’re going to progress from where you are now to a future state, but it’s a process you’re going to take baby steps at first.

So yeah. Yeah, the place where we’d like to encourage people is to, once you started thinking about it as an integration, so that doesn’t necessarily mean, throwing away what you’re currently doing with NDT. If you have let’s say an immersion tank process or something like that, where you’re scanning parts that are going to end up in, forgings for oil and gas installations or, jet engines or whatever, you can begin there.

Start collecting that data start automating that process, and think of it as integrated into your MRP and ERP systems. So instead of an operator potentially having to type stuff in by hand, you can barcode that information. Or you can just query a database. Query the factor database and say, okay, I’ve got part number a, XYZ one, two, three, is, I’m going to inspect it. And so then the inspection process becomes integrated into the factory. And then the next step is, once you’re doing that start saving your data. very often the data are not preserved, especially in a manufacturing environment. Once you’ve rented it, the decision that the part is, okay, you may have a report that may have an image and, a data file gets saved somewhere but it’s not necessarily accessible. So start thinking about building today, an archive of data, a goldmine that you’re going to use later down the road. So those are the first, those are the two big steps.

Think about it as a process, and begin integrating it. And then start thinking about your data as an asset and start preserving it like you’re going to save for college.

Nasrin Azari: [00:24:40] Yeah, and speaking of the data that brings us to our question five for today, which is around data analytics and AI. A lot of people think about some of the early benefits, saving data, and keeping track of all of this data. A lot of folks don’t know what to do with it today. And I think you alluded to that earlier, but speaking specifically about data analytics and AI, what do you think success looks like in a short term and how might it look in the longterm?

Dave Seto: [00:25:09] Okay. Two things. One we’ve already discussed: save the data, start preserving it now. I don’t mean that in a sort of dusty dead-end, put it in a box in the attic save the data. What I mean is to start building a database that you can use today and begin thinking about how you might also use it in the future.

Like for example, AI. Dr. Singh talked about how an AI needs to be trained in the same way that you train a human operator. Like a human analyst doesn’t inherently know what a bad weld looks like. And then once you’ve detected that there’s a difference in this material that you welded is that a crack, is that porosity is that some other form of defect?

So human analysts get trained. And as experts, then they train other humans. You need to train an AI in the same way. Like it needs to be able to recognize what a defect is in the first place. And then it needs to be able to classify defects, so begin saving your data. And we’re starting to see that some of those oil and gas companies are already starting to do that.

They’re starting to save databases with not just the raw data, but the analyzed data. So the data that’s been classified by an expert human who has made a judgment and done some measurements and said, okay, these portions of the scan are, indications that we care about.

And here’s how you discriminate between lack of fusion or flaw, some other type of flaw, the act of doing that is already beneficial because you can data mine that today. That’s just a statistical exercise. So if you have an archive of data, as soon as you’ve got a sample set of, let’s say more than five years, you can begin to build statistics about what your process is doing.

And that can happen today and you don’t need an AI for that. Any programmer worth their salt who works in ERP or MRP systems can write you a little algorithm to go mine your NDT data and figure out if there are any statistical trends worth looking at.

That can happen to that. And then, as that data builds, you can begin to decide, okay, if we want to speed up the NDE process, if we want to make it more reliable, maybe an AI is the right thing for us. We can go faster and it will make fewer mistakes than, say a human who’s tired at three o’clock in the morning and, trying to discriminate, between flaws. One of the things I like to point out to some of our customers is that in some cases we’re collecting tens of gigabytes of data, and they’re still asking a human to analyze that. And if you put it in the perspective of I’m going to make you watch a movie, which is about four gigabytes on a DVD, and I want you to tell me exactly when and where the dude in the red hat shows up in the crowd scene.

If you think about that, you’ve got to watch a two-hour movie. And there’s a brief moment where some guy in a red hat shows up in a crowd scene, pick it out, and don’t make a mistake because that’s the defect that you’re looking for in four gigabytes of data. When you start thinking about it in those terms, asking humans to come through that way, data is pretty, you’re putting a lot of faith in people which, is not unwarranted, but there are better ways of doing it.

And so that’s one of the big things that we do a lot of these days is helping customers automate not just the collection of the data, but the analysis of the data. A lot of people call it assisted defect recognition, cluster analysis, that kind of thing. So what you want is, it’s not really an AI.

It’s just an algorithm that comes through the data and goes, this stuff is different than that stuff. These blobs, you should maybe go look at and measure. So we do a lot of that, but then that’s the end of the benefit unless you save that data with a classification that says, okay, this was found in a scan.

An expert human analyst looked at it and said, this is surface irregularity. That’s not a defect because it’s going to get machined away. Anyway, this is worth looking at because it is an inclusion or porosity or something that is in fact a defect. So it’s only by doing that analysis today and then not just rendering a judgment. And then the, the part goes out for reconditioning or gets rejected or whatever the next step is the step that we’re not doing today. And that’s the step of saving, not just the raw data, but saving the data that a human has analyzed and said, yes, this stuff is worth looking at this stuff will show up, but it can be ignored.

Nasrin Azari: [00:29:50] And then those results can be fed back to a machine learning system to improve the AI.

Dave Seto: [00:29:55] Exactly, and think about it. We’re already doing it every day. Every time you say I am not a robot, and you have to identify all the pictures that have, I dunno, school buses in them, you’re training an AI.

We could be doing the same thing with our non-destructive test results. We could be saving what the human has decided is a true defect. And, here it’s measurements and here’s its classification, right? This is what porosity looks like. This is what the delamination looks like. This is what lack of fusion looks like, all that sort of stuff it’s happening.

I think at a research level, and there’s probably some companies who are, especially in the oil and gas industry, I think, who are beginning to deploy AI for analysis at least. And I think we’re going to start to see more and more of it in other industries like aerospace and power generation, things like that.

And once we have that capability, there’s the virtual cycle happening again, because if we can do that quickly enough and cheaply enough, we can inspect things many times during their lifespan without having to shut stuff down instead of the way we do it now, which is, when we’re coming up on our 25-year end of life, but we’ve decided we need to keep that asset going.

Let’s shut everything down and we’ll let you know this big inspection campaign and then start it back up again. And then, hopefully, be able to stretch another five or 10 years of life out of an expensive asset. The better we get at NDE, the better we get at NDE 4.0, we can build a history of that thing from its future today.

If we start collecting that data today in 25 years, you will have a bunch of data about the lifespan of that object. And you can use that. We don’t yet know what we might be able to do with that amount of data about the history of an object. Now, what happens to wind turbine blades at 25 years, if they’ve been exposed to more stress or less stress over their lifetime. We can design it better wind turbine blades, and that’s not going to happen tomorrow. That’s going to happen five years when we have 25 years of experience and that’s where we need to start today. So the data that data can be immediately beneficial, just from a statistical point of view from data mining what you’re already doing. And then in the future, as you try and better AIs that feeds back into your manufacturing and design processes and you end up with better wind turbines, more competitive engines, whatever it is that you’re building.

Nasrin Azari: [00:32:18] I love sort of the theme throughout this whole conversation today about the fact that using these technologies really helps us create better products, safer products, and just, continues to help improve. The other thing about the other comment you made about, as we start incorporating NDE 4.0 technologies and NDT testing will become cheaper and faster, and we’ll be able to do it more of it. So that should be encouraging for folks that are in the industry that are currently afraid that robots and technology are gonna take away their jobs. Certainly, there will always be jobs if there’s more work to be done. So it’s pretty fascinating.

Dave Seto: [00:32:59] I heard that, too, that the NDE industry, it’s a little bit like the medical community in the sense that we have this process where a technician makes an image and then another technician with a slightly higher level of experience and understanding maybe does the analysis and renders a judgment and then even higher levels of trained people will design the process that the two of those people use to render the image and then make an assessment. And that’s not going away. We’re actually adding jobs, I think because most of the work that we’re doing now involves robotics. Because like I said, you can’t scan something, that’s the size of a school bus by hand.

You just can’t, especially not in a timely fashion. And if you automate that, now you need to hire a robot technician and you need to hire, engineers who can manage the robot and you need to hire a higher level of educated NDE technician because now they’re not only moving probes by hand or operating a small machine.

They have to interact with a team of people who are operating a giant robot and they themselves will probably be expected to understand enough about programming the robot and manipulating the robot that, that’s a higher level of skill.

Nasrin Azari: [00:34:20] Yeah, that’s really interesting too. Just a fascinating thought for folks that are really future-oriented and interested in technology with a lot of people in this community are. I think there’s a lot of potential.

Dave Seto: [00:34:31] I think so too. And I think that’s how we’re going to get young people interested in NDE. I was at the ASNT Conference talking to people about the age demographics of the ASNT in particular, but I think it’s the same in Europe, right? It’s, we’re a sort of baby boomer industry. We’ve grown up with, the electronics age was how we began to be able to do things like ultrasound inspection and Eddy current inspection and so on and so forth.

And, young people are not necessarily interested in climbing down into a muddy hole and dragging a probe around on a pipe. Nope. No thanks. But hey, would you like to operate this million-dollar robot that’s going to crawl down 50 miles of pipe and come out the other side.

Or would you like to operate this remote control drone? That’s going to fly around and inspect, wind turbines that are the future of green energy. Like whatever the NDT application is, if we want young people to be involved, right they need to feel encouraged and engaged. And making it high tech, as opposed to the old school, low tech, is the way that we’re going to attract bright, young people into the industry and they’re going to drive it forward.

Nasrin Azari: [00:35:39] Definitely. Super interesting. This has been really incredibly informative today. Thank you so much, Dave, for sharing your knowledge and experience with everybody today.

Dave Seto: [00:35:50] Thank you for having me. It’s been a lot of fun.

Nasrin Azari: [00:35:52] Yeah. Good. If any of you listeners would like to reach out and connect with Dave, we’ve got his contact information on our podcast page, so you can learn more about Dave and UTEX Scientific Instruments there.

So thank you again, Dave, and thanks to the listeners for tuning in and we’ll see you all next time

Dave Seto: [00:36:09] My pleasure bye for now.

For more expert views on NDE 4.0, subscribe to the Floodlight Software blog at floodlightsoft.com.

Scroll to top