Prith Banerjee with Ansys
On this week's Industrial Talk we're talking to Prith Banerjee, CTO at Ansys about “Simulation Based and Hybrid Digital Twins”. Get the answers to your “Digital Twin Simulation” questions along with Prith's unique insight on the “How” on this Industrial Talk interview!
Finally, get your exclusive free access to the Industrial Academy and a series on “Why You Need To Podcast” for Greater Success in 2022. All links designed for keeping you current in this rapidly changing Industrial Market. Learn! Grow! Enjoy!
PRITH BANERJEE'S CONTACT INFORMATION:
Personal LinkedIn: https://www.linkedin.com/in/prith-banerjee/
Company LinkedIn: https://www.linkedin.com/company/ansys-inc/
Company Website: https://www.ansys.com/
PODCAST VIDEO:
THE STRATEGIC REASON “WHY YOU NEED TO PODCAST”:
OTHER GREAT INDUSTRIAL RESOURCES:
NEOM: https://www.neom.com/en-us
Hitachi Vantara: https://www.hitachivantara.com/en-us/home.html
Industrial Marketing Solutions: https://industrialtalk.com/industrial-marketing/
Industrial Academy: https://industrialtalk.com/industrial-academy/
Industrial Dojo: https://industrialtalk.com/industrial_dojo/
We the 15: https://www.wethe15.org/
YOUR INDUSTRIAL DIGITAL TOOLBOX:
LifterLMS: Get One Month Free for $1 – https://lifterlms.com/
Active Campaign: Active Campaign Link
Social Jukebox: https://www.socialjukebox.com/
Industrial Academy (One Month Free Access And One Free License For Future Industrial Leader):
Business Beatitude the Book
Do you desire a more joy-filled, deeply-enduring sense of accomplishment and success? Live your business the way you want to live with the BUSINESS BEATITUDES…The Bridge connecting sacrifice to success. YOU NEED THE BUSINESS BEATITUDES!
TAP INTO YOUR INDUSTRIAL SOUL, RESERVE YOUR COPY NOW! BE BOLD. BE BRAVE. DARE GREATLY AND CHANGE THE WORLD. GET THE BUSINESS BEATITUDES!
Reserve My Copy and My 25% Discount
PODCAST TRANSCRIPT:
SUMMARY KEYWORDS
simulation, twin, digital, data, people, model, talk, asset, predict, scott, industrial, industry, crack, jet engine, ansys, preventive maintenance, world, bridge, abb, data analytics
Welcome to the industrial talk podcast with Scott Mackenzie. Scott is a passionate industry professional dedicated to transferring cutting edge industry focused innovations and trends while highlighting the men and women who keep the world moving. So put on your hard hat, grab your work boots, and let's go
Alright, welcome to Industrial Talk once again, the number one numero uno industrial related podcast in the universe that celebrates industry heroes all around the world, and companies and companies put that down. Because you're bold, you're brave, you dare greatly. Absolutely, you innovate, you solve problems, and you're changing lives. You're changing the community in a well, changing the world that come on. That's why we celebrate you. That's why you're cool. All right, in the hot seat, we have a gentleman by the name of Prith Banerjee, CTO. ANSYS is the company we're talking digital twin, but we're talking about the ability to simulate using data analytics simulation, and honing in on that answer. Let's get cracking. innovation, innovation, it's cool stuff, right? And it is necessary. And if you're not in, if you're not doing your homework and getting in that, or talking to trusted sources about innovation, technology, and whatever it can do and how you can leverage it. I asked you to do it now, right? Industrial talk has trusted individuals go out there and just begin that journey. Have that quiet, have those conversations because you need it. You just do. We need for you to be resilient. The world is changing all the time. And you need to leverage technology and innovation and talk to trusted people to be able to create that business of resilient because we need you. We need you big time. Alright, when we start talking about innovation, I've got two sponsors that you need to check out. Big time. And because it's cool, so AI dash, right, go out to AI dash calm and they're able to use the technology satellite specifically to help you with your vegetation management, right utility, vegetation management. As an old crotchety, crusty, journeyman lineman, I knew how I had to trim trim trees, and it was all reactive. Trust me, it was reactive run out there trim, oh my gosh, there's another one. There was really you tried to make it sort of strategic. But now he never has he got run out there, grab the bucket truck, trim it boom, move on. Here is a here's a solution, a iDASH that looks down on your service territory and says, Yep, go right there. Go over here, trim that tree. Yep, go over there and not dunk over there. That's fine. And that good. I mean, that's that is deploying your capital going in your money in a way that really benefits us the consumer. So that, that the service stays up and running and you know, delivering the power that we need AI dash, they're leading the way that satellite AI, vegetation management, simplified baby. Alright, and the other one is Neil. So you go out to me on.com. And I have this conversation all of a time. I talked to innovators, I talked to technology delivers and service providers and everybody and we're talking about digital transformation. We're talking about AI when talking about quantum computing, we're talking about all of the topics that are really a buzz today. And the reality is, is that here's Neil, they're going to put it into action. They're creating a community of the future. And they're going you know what, you know what the best part about it is, they're going to make our lives better, they're going to go down that road, they're going to get the bumps and bruises, they're going to try to figure out solutions that really are meaningful to us. Neil, right. meaningful to us solve those problems. And we as a as a society, benefit greatly. I'm doing it's cool, don't you? Come on, you got to get all go No. jittery, happy with that stuff. That's me. on.com. They're creating the community of the future and they are doing it. They're doing it. Alright, neon count go out there. Alright. So one of the things that I've talked about is getting the most out of right getting the most out of it. Are you getting the most out of conferences? Are you getting the most out of your your engagements, whatever might be getting the most out of your campaigns, whatever it might be? Are you and ask that question and be truthful about it? Are you or is there ways of being able to improve that engagement, that conference, that whatever it might be, and I'm here to challenge you there? There are strategies, there are solutions that are out there, and one of them is okay. People ask me Scott, why do you why do you podcast Well, one, it's fun. Yeah. And and it's, it's available for everybody. But one, it's fun. And two, I get to talk to
leaders around the world. And these leaders are movers and shakers, and they are at the cutting edge of thinking and deciding what's good. You need to be there. You need to earnestly and and I mean, earnestly, there's ways that I'm going to continue to sort of add to this, but at a minimum, look at creating a podcast. And I have a I have a program out there that says, why you need a podcast, and it takes you putting every step of the way, right. But the reality is, is that if you want to get the most out of something, have multiple touchpoints, and an engagement, that that you can offer to your prospects, your customers, whatever it might be, bring life back to a blog. There's no better way. There's no better way of being able to do that. Yes, yes, there's a lot of podcasts out there. What we do at industrial talk is we we recognize that there is this fragmentation that exists out there, right, there is podcast over here, podcast over there, and so on and so forth, great content, they're great content, their solutions here, right? What do we do? Well, I'm I'm, I'm in a game of consolidating that I'm in the game of trying to bring all this great thinking together so that the industry, the community, the ecosystem, has a place to go and says, I want to know about cybersecurity. From a trusted individual, right, they're easy, right? And because typical topics tend to be very niche, a, you know, it's it's, the reality is you tap out the opportunities there. So you need a community to be able to be a part of flight information, to be able to be impactful. And have your dog God message be heard. He got to have it heard, right. It's got to people got to listen to if they don't like, does a tree make a sound when it falls in the forest? If you're not there? I don't hear it. I'll hear it right now. The same thing exists with your voice. You got to do it. And we'll talk a little bit about but I just tell you just go out to the well, this podcast and other podcasts will say why do you need to do a podcast? You got to do it. You got to figure it out. Contact me. I'm on I'm on a dock on book I am. I'm warm and fuzzy and cuddly. Maybe I can be more cuddling anyway. Alright, let's get going. So what's interesting and had a great conversation with Prith. And what's interesting is the the ability Yes, we can collect data, it's all good. Click, click, click, click Collect. But what about that simulation component? Right? It's great. Got all that data? Let's simulate what's taking place what simulate what might happen or whatever, and be able to, without a doubt, begin to converge the data and simulation into a true cogent response or information on when you need to perform maintenance, perform whatever corrective action, whatever it might be. Simulation is, it was great talking to Perez because he's you could tell he's pretty passionate about the whole thing and rightly so. Rightly so. Alright, let's get cracking. You don't want to hear me a moron. Here's Prith talking about digital twin but really simulation enjoy. Welcome to industrial talk. Thank you very much. Broadcasting from Palo Alto. That's where he is coming from. If you're out on the video, you'll see that there's plenty of sunshine happening there. Anyway, how you doing?
They're very good. Thank you very much, Scott for having me.
I'm excited about this conversation. Okay, listeners, we're gonna be talking digital twin. And you're saying Scott, I want to talk? Yes, you do you want to talk digital twin. But more importantly, I think Perth is going to bring a wealth of knowledge that you're just going to be dazzled by man. I know he is. So brief before we get going give us a little sort of 411 a background on who Perth is.
Sure so. So I'm Prith GMs Chief Technology Officer at ANSYS. ANSYS is a modeling and simulation company. We are a company of about $2 billion in revenue 5000 employees we help provide solutions to our customers in terms of when they are trying to design products. We enable them to design develop the most amazing products in areas such as automotive, or aircraft, aerospace, or manufacturing energy and so on. So that's what MCs does and my role as Chief Technology Officer is to set the long term technology strategy for the company around areas is AI machine learning high performance computing? digital twins?
Um, so you just you just threw out a hyper comment. What was that?
High Performance Computing high performance?
Oh, yeah. Yeah, I'd like that. See, you're you. You're sitting in the seat where a lot of people want to sit just because you get exposed. I mean, you must see a lot of innovation just coming. I mean, just constantly evaluating looking. You having a conversation around innovation and the impact it has on industry, society. I mean, it's pretty cool. You're living a dream, you have the
most fun job at MCs, I'll tell you, I, we have 45,000 customers. Every day, I talk to sort of CTOs of different customers across different verticals from aerospace. Like today, I had a meeting with a CTO of a medical device company yesterday, I remain meeting with a 5g company. So I'm learning about the sort of the future trends in these different verticals. And then from my own team members, right, I get to learn about the newest coolest technologies of AI machine learning applied to simulation or high performance computing, or digital twins or, or prime meshing, it is a fun job. And yet, here's
it and I've never seen it. So there's, there's, there's such a speed of velocity that's happening out there. And it's, it's, it's, it's exciting. But man you are, that's pretty cool. So let's start like diving into that. Let's dive into digital twin. Now, for the listeners out there, give us a little just what is a digital twin, and then we're gonna go into some of the challenges and opportunities that exist within that technology.
Absolutely. So in digital twin is a virtual replica of a physical asset. So let's say you have a jet engine of a plane, right. And that jet engine originally gets designed in a CAD tool, right? You say here is an engine, he has all these blades, and so on, and you design that jet engine on a CAD tool, then you actually manufacture that jet engineering gets installed on a United Airlines plane and this jet engines flying around. And so right, that is a particular jet engine, now you try to make a model of that engine. And oftentimes, people confuse, oh, you have a simulation model of the engine, that's a digital twin. That's not that is just part of it. So a model of a physical asset, like a jet engine, is one thing. The other thing is there has to be two way information flow between that physical asset, the jet engine, and the model. And that's what makes it a digital twin. So you have the jet engine, you put sensors on a jet engine, right, and you have you collect all those data for the sensors, you upload to the cloud, or whatever you are pulling that information. And using that data that you have from that jet engine, you try to make the model of the jet engine, as accurate as you can write with respect to what is in the real thing. So it's like a virtual replica. So I'm praying, I have a model of it. And this model of it keeps getting updates right here is with really excited, so the excitement level is going going there, right? Bridge temperature is what 98.4 Right? So it's constantly measuring what is happening on the real asset, and updating the model at the same time. So a digital twin is not a model of a jet engine. It's a model of a particular personalized model of that particular jet engine. Right? And how that is going on in time. So this is something that NASA used many many years ago when if you have seen the Apollo 13 The wonderful thing sort of the the picture that the movie about Apollo 13 Right when that that Apollo 13 had an accident, right? Yeah, people in NASA in Houston, right? Hey, Houston, listen to his to Right. Um, those engineers at NASA actually built a digital twin of Apollo 13. And he said, Oh, if you do these things will happen. So essentially the, of what was happening and sent all the instructions to the astronauts, and that's how they got in there. So she
did that. You just brought it to it. And I guarantee a lot of people are listening is like, yeah, Apollo 13. I remember him sitting in the capsule, switching and doing everything and making sure the amperage didn't, you know, whatever it was, whatever that target was Do this, do this do this. But that was that makes
complete sense. What was happening in Houston, right? Yes, a physical asset out there in space. And they had a model in Houston. And essentially all these scientists were trying to do stuff. What if this happened, this will blow up. They said okay, this would work. And they send those instructions and that's how the astronauts are brought back. So Go digital twin back in so long ago, right in the in a but essentially, the digital twin technology has come a long way since. But that's what digital twins are. It's a model of a physical asset with Two Way information flow between the user and the model.
And why is that important? It just gives me the individual able to look at the, the, the the characteristic of that asset that that real asset in a digital environment and see what's happening. And then I can I can, I can tactically move forward in some decision making or whatever it might be.
nicely. So let me give you an example. So we, we all rotate tires in our cars, right? We rotate tires every 5000 kilometers, miles. Why is that because all the tire companies have seen statistically, if you drive your tire out on a normal road, normally normal driving, your tire will wear this much any and this is going to happen. Except if it breaks driving crazily on us 101 in Santa Fe with lots of I probably need to rotate my tire every 3000 miles, right. So that's personalized driving. So you take the statistics of all possible cars, all possible drivers. And that sounds people say you should change your oil every 20 10,000 miles and changes. That's called preventive maintenance. Right? So you Why do you do preventive maintenance, you replace your oil, replace your tire, etc. So that your tire does not burn your engine doesn't conk out in the world of so there is this preventive maintenance that people do Now imagine if you could predict exactly when that engine would come. So just the day before you replace the oil, that would save a lot of money yet question is can you predict accurate? Yes, see, and that's what the digital twin comes. So the what the digital twin does is you have the physical asset, the engine or a car or medical device or whatever, right? And you are monitoring it and you are getting real time information, as opes is about to fall sick. So we humans, we fall sick, right? Before we fall really sick, you get a temperature, you can oh my day, I'm feeling really lousy, I'm having an A. So those are signal area about a false and then when you get one or three temperature you are boom, right? So imagine if you could sense things around you, you could do predictive analytics. And that would be huge. Because suppose you have the data center for Google or Amazon, imagine that, that data center going down, that would create tremendous waste, right? Five minutes of downtime on Google, or Amazon would be hundreds of millions of dollars of thing waste, right? If a power grid goes down lectricity goes on tremendous waste. So people in these essential industries, there are critical infrastructure assets, they don't want them to go down. So what they do is they try to predict when this is going to fail, and before it fails, do the corrective action. So the real deal here is can I predicted accurately
that even see that is because I lived that right? I was in that world, I was in the world of maintenance. And when you spend a maintenance dollar up here on the financials, it's $1 Down on the bottom, right. And if you could sweat the asset, and sweat it in such a way that you know when to perform that maintenance before it catastrophically fails. I've saved that much time, that much money. It's and you don't have a catastrophic failure that could put you out for who knows how long you know, whatever it is, how do we get there prep? There's that we've been having that conversation, people understand the value of that predictive capability. When do we really start to see it happening commercially?
So I will actually tell you sort of the first the technical challenges then the business challenges so the word again, again, there's this thing called the Digital dream consortium that I am a part of and there are some founding members I Microsoft and GE and handsome so right so I've been in this business for for some time now I've sort of lifted and so on or prior to MCs I did not tell you. I used to be CTO at ABB CTS right electric and when I was on those companies, I was helping build digital twins of the assets of ABB and Schneider Electric right so they have a transformer for ABB. You collect all the data right from sensors and you just by collecting data, you try to build a digital model of that transformer in the normal business havior and the apps as soon as it starts getting signals of abnormal behavior is out bad things are happening. Shut it down. Right? So the most of the world uses digital twins based on data analytics alone, right? So yeah. And they collect lots and lots of data, right sort of
tsunami, tsunami
of data from those data. They I mean, they use AI machine learning techniques. i The older the world knows about AI machine learning you you train those, sort of these, these convolutional neural network, CNN models of the 60s neural network. So there's a lot of work that people have done in this community using AI, ml, CNN, etc, to build these digital these digital twins based on analytics alone. Now, when I was at ABB, and Schneider Electric, we used to do all those things. And we found that you know what, it was not that accurate. So, yeah, it said, 70% accurate. So he said, Well, 70% is pretty good. You have a million dollar asset, you're 70% accurate means you made a 30% mistake. 30% mistake is $300,000.
That's, that's a lot of money. That's
a lot of money. So this is what the challenge was. So in the industry, people said, Oh, if you did pure analytics, right, it's not doesn't look good, does not always work. Now. Analytics works in cases, where you have enough data on the use case you're looking for, right? So four frequently occurring events, data and this goes well, for infrequent events, you cannot predict it, let me give you a very bad example. Space Shuttle Challenger exploring how many times has a space shuttle extended once. So you can get all kinds of data on space shuttles based on and so on, he would never be able to predict when that space shuttle would explode, right. So that is an infrequent event. So when an event is happening all the time, right? You have little things that are little non bolts or whatever, there are sort of breaking down in a in a this thing. frequently occurring events, a IML is great at catching infrequent events, you cannot go there. So this was my aha moment. Oh, okay. So what can we do? Therefore simulation, so now all around us, is governed by Singlish. governed by the laws of physics, which get modeled really, really accurate. Let's imagine you have a bridge. And you're trying to predict when that bridge will collapse, right? So that you can shut down the road, right? So you put all kinds of sensors on the bridge? And you say, hey, is there a crack on the bridge or not? So first of all, you can say, How long do bridges last 30 years. So he said, predict preventive maintenance on the bridge every 30 years, every 30 years, you go and fix the bridge or whatever. That's the preventive maintenance way. The IoT way is to put sensors on the bridge to see is there a crack on the bridge or not? Right? So let's imagine that there are cracks on the bridge, I said, Here is a crack that is 12 inches long today. Now you have no idea where that crack would go next week. Aha. But if you had simulation using say structural analysis, finite element analysis with ANSYS tools, I am telling you, Scott, I can predict with 1000 cars going on the road with this kind of speed and so on, what is the stress sprayed on the bridge? And if and if there's a crack that is with steel pillars and so on with is 12 inches, I am predicting through simulation that this crack will grow to 18 inches in a month through simulation.
So if I did that through simulation remand as I said it pure simulation, I would say this crack which has 12 inches long images tomorrow 44 inches at 36 inches in the crack, I will predict through simulation alone that the bridge will crack in a month. And that is a prediction. Through simulation alone, we have found out that the accuracy of this 70% goes up to 90%. No way. Okay, so now you have this million dollar part with physics based simulation. We have a tool called Twin builder, we were able to increase it from 70 to 80% over 90%. So that million dollar part your waist is now only 100,000. Here's the only 100,000 and still adorable. So that's where we have this aha moment Right? Where we said let's do hybrid digital twins. Let's combine that data analytics with the physics so now I'm going to bring it together whether example, here is this bridge, we had a track we had a crack which is 12 inches long, right to data. I measured it through to data analytics Right. And I predicted it will be 18 inches long a month from now. But a month from now I take the measurement, and the crack actually was only 60 meters long. So my simulation model was making a prediction, it was going too fast. So my data is a go slow, so it goes slow. Or, for example, that actual measured crack was 24 inches long. So that means my simulation was going slower. So by getting a, a updated information from that, that's why I say remember that the two way information flow, I pulled data from the bridge about the size of the crack. And I made my simulation model more accurate, synchronized with the real thing. So as I change the rate of crack growth, right, instead of making it 18 inches, I cracked it to eight, six inches, or 24 inches, whatever is going on the next month, when I read it, I will be a little more accurate. I'll say it should be 32 inches. And next time I say no, it should be 31. Oh, so it slowed down even more. So through this hybrid approach with a combination of data analytics, and simulation. Scott, we have shown data that the accuracy goes to 99%.
No way. Dan can, can you because this, this is just cool stuff. Don't get me wrong. I think this is cool stuff. Can you take that bridge analogy and you're you're you're honing, you're tweaking? You're you're zooming in on accuracy by data and simulations, and you bring it all together? Can you take that and say, Alright, this is it. This this bridge is in Texas, let's go to, you know, Florida, and and sort of take that learning that, that that simulation and apply it. And again, yes, tweak, tweak, tweak different weather patterns, a little bit more speed of the cars, whatever it might be, but be able to use it there too, as well. Absolutely. And
so just like sort of just the AI to like Siri, right, your your, your little tool. I mean, if I saw Siri, when it listened to you, Scott, you have an American accent, I have an Indian accent. But Siri, after listening to me for a month, it gets pretty good, right? When I say go do this, I mean, so the machine learning things get better and better, the more data it gets too similar is hybrid digital twin get. So there is something you can do with simulation with this basic version. But the more data you give, the better the model becomes. And this is sort of the state of digital twins that we are I am seeing from from ANSYS. And I'm talking to lots of lots of customers across different verticals, we have got sort of applications in industrial flow networks, working with oil and gas companies, applications in in, in electrification, with battery systems, and so on applications in HVAC, and buildings and so on tremendous opportunities across all verticals,
where does it end? And when I say that, let's say I am, I'm at 90% accuracy, I can measure it, I've looked at it my simulation, Mike data on there, and I'm tweaking, and then I put more time, energy and effort, I'm at 95, and so on and so forth. There's a point where it's like, Okay, we're good, we're good. And any any additional incremental benefit is just like, minuscule.
Yeah. And so that is sort of it is the law of diminishing returns, right? You plot a graph, and you see keep seeing, making chimey, at some point is flattening out. And so it really depends, right? For that particular engine, or wage or flow network, or valves or whatever, when you see the law of diminishing returns, you're not really getting incremental stuff, right? You sort of stop there like that, that's, that's what we have done.
And and, wow, it the applications are quite frankly, endless, because you get it to you get it to the point where let's say it's a motor, right? You do the same thing data simulate and and I and your analogy of Apollo 13, I see a guy in a capsule running a simulation, trying to get it just right. The same thing exists, you could be on that asset, no pump, motor, whatever it is, but you can apply it and it might be different in Texas as it is in Maine, right is because of the different weather whatever it might be different application. But
absolutely, yes. Now, let me tell you, I talked about physical assets. But you and I we are not we are not just physical asset. We are humans. Right. So let me talk about the opportunity of digital twins in the healthcare industry, right. So inside our body's most complicated organ, the human heart inside the heart, right you have a sort of this dielectrics signals that pump, right that generate the signals 3070 times a minute, right? So in response to electrical signals, the muscle heart muscles contract. So when the heart muscle contract, the atrium, the blood flows from the atrium to the ventricle, and then the ventricle for this thing, and the valves open up. And so it's just most complicated organ in the human body. Guess what I mean, ultimately, those human models are modeled by physics, right, there's electrophysiology, there's mechanics, there's fluid flow. So we are dancers have actually model the human heart. Now, there is a generic human heart that you can model and you can say, Okay, this is what is going on, we can also model treats human heart with as arrhythmia, right, and I have an irregular heartbeat problem. So it goes into a doctor takes an MRI scan, CT scan, whatever of the heart, and He measures all those things. And for prints specific heart, we can come up with well, if I have a B or whatever you can do that things you can do heart surgery or so on. Right? The other thing is, I mean, you can put in a implant, right? A pacemaker, right? So how will this pacemaker from a company like Medtronic interact with my heart, right? So a Medtronic company, that battery will have to do clinical trials of all kinds of pacemakers, and all kinds of patients and so on, right? But through a digital twin, of a human heart, right? You can create all these different models, and essentially, shorten that time compress it big time, or for trials of a pacemaker, right? And if so, that's kind of where we are heading, we are headed in terms of what is called in silico. Trials, right? Computer based simulations of 1000s of patients. I mean, the future is amazing.
Ah, see, see that that delivers, that deliver solutions, faster? Health, whatever, the health side, I just didn't even think about, but you're absolutely right. Create a little digital twin, correct. And you can run simulations, but it's, it's individually based, right? It's your heart. My heart is different than yours.
Like Fitbit, right? The Fitbit is collecting. Exactly you ever saw collecting data on you, Scott? 24/7. So imagine if your doctor had a model of Scott, sitting inside, right, and he's collecting all this data, right? And God bless them use my example, right print a model of prey, and just before previous about to have a heart attack, right, that doctor just two hours before gets to know a signal Hey, there's this the signals are going in the wrong direction previous governor calls of the ambulance and sends an alert before he has a look at them if the value of them
Yeah, yeah, that's that's cool stuff, my friend. Yeah. Why? What's what's, what's the, you know, the roadblocks? I mean, this is all good stuff. Let's make it happen today. Let's let's make it happen tomorrow. Let's put it that way.
As you know, Scott, I serve on the advisory board of the IoT solutions World Congress. In fact, we have a conference coming up in May of of this year, right.
And I put that on your calendars listeners big time.
I serve on the digital twin Consortium, and still there, this, it's a more than 300 members in this consortium and so on. So people are are excited about the potential of digital twins in different verticals, in manufacturing, in factories, in industry, for Dotto and so on so forth. The thing is got that shiny toy looks like a shiny toy. Oh, you have fantastic possibilities, right? But when you start actually pulling it together, right? You have to collect all this data and so on, and does it work in all the time and so on? So many customers are saying, oh, you know, I like the shiny toy. Let me use the shiny toy in a proof of concept. So you do a proof of concept number one, you waste like three, four or five months, maybe $100 million project and so on the like this. Oh, let's try this other games. Right. There's two POC number two then POC number three. So so people in this our Digital Consortium talk about the POC, Purgatory right. Before moving to scalable systems. So the digital twin thing is they going through this sort of inflection point where it is a lot of promise, but it has not quite seen scalable value across all industries in all opportunity. And that is I think what the business challenges today,
I agree with you, but it's gonna happen. It's gonna happen it's gonna be in commercial use. It's
like a $26 billion market by 2025. So yeah, it is going to happen is about a couple of billion, but it will happen.
It makes sense, right? There's no, I mean, it makes sense. You got to do something like that. That's make far right. We're gonna have to wrap it up. I hate to because I enjoy this conversation. How do people get ahold of your prep?
So they can contact me at my answers email address Prith.Banerjee at PRI th got Bannerjee ba NERJE. App. ansys.com. That's a NSY. Yes. Yeah, you can search me on LinkedIn, at LinkedIn Dash Grip Banerjee, or a Twitter at Big Banerjee.
Alright, listeners, you don't have to worry about it. I'm going to have it all out in industrial talk, you'll be able to get a hold of this. This incredible professionals with no problem and don't come to me then Sen. Scott, I can't get a hold for it. You're lying. You're lying out there. Just don't do that. All right, Fred, you were absolutely spectacular. Yeah, you your energy's gone. It's exciting. I love watching you just get all giddy about this stuff, man. It's cool stuff. Good job. Your job is to be a
professor for 20 years. Really? Yeah,
that's a hard job, by the way.
Oh, that's where I, I learned how to help lecture.
Wonderful, wonderful. All right, listeners. We're gonna wrap it up on the other side. Once again, we'll have all the contact information for prep, and then some. So stay tuned.
You're listening to the industrial talk Podcast Network.
All right, thank you very much for joining. Once again, industrial talk. Thank you. Thank you very much for saying yes. And being on industrial talk. Love the conversation. You listeners need to reach out to prif go out to a stat guard go out to LinkedIn, you will not be disappointed or go to industrial talk, find the link, reach out to him. You need trusted professionals that are at the forefront of thinking through all of this Firth is one of them, reach out, do not hesitate to do that. All right. I'm going to continue to challenge you once again. That is are you getting the most out of the bar. The bottom line is we need you. You need to be involved. You need to have that business of resiliency and you need to do a podcast. You need to create one you need to figure it out. Again, go out to industrial talk. I'll give you the step by step everything you need right there right now. No excuse you need to be a part of that. Alright. People Be bold, be brave, dare greatly hang out with people that are bold, brave and daring, greatly industries full of it. You're going to change the world. Thank you. Once again, we're going to have another great conversation right around the corner.
Transcript
Welcome to the industrial talk podcast with Scott Mackenzie. Scott is a passionate industry professional dedicated to transferring cutting edge industry focused innovations and trends while highlighting the men and women who keep the world moving. So put on your hard hat, grab your work boots, and let's go
Alright, welcome to Industrial Talk once again, the number one numero uno industrial related podcast in the universe that celebrates industry heroes all around the world, and companies and companies put that down. Because you're bold, you're brave, you dare greatly. Absolutely, you innovate, you solve problems, and you're changing lives. You're changing the community in a well, changing the world that come on. That's why we celebrate you. That's why you're cool. All right, in the hot seat, we have a gentleman by the name of Prith Banerjee, CTO. ANSYS is the company we're talking digital twin, but we're talking about the ability to simulate using data analytics simulation, and honing in on that answer. Let's get cracking. innovation, innovation, it's cool stuff, right? And it is necessary. And if you're not in, if you're not doing your homework and getting in that, or talking to trusted sources about innovation, technology, and whatever it can do and how you can leverage it. I asked you to do it now, right? Industrial talk has trusted individuals go out there and just begin that journey. Have that quiet, have those conversations because you need it. You just do. We need for you to be resilient. The world is changing all the time. And you need to leverage technology and innovation and talk to trusted people to be able to create that business of resilient because we need you. We need you big time. Alright, when we start talking about innovation, I've got two sponsors that you need to check out. Big time. And because it's cool, so AI dash, right, go out to AI dash calm and they're able to use the technology satellite specifically to help you with your vegetation management, right utility, vegetation management. As an old crotchety, crusty, journeyman lineman, I knew how I had to trim trim trees, and it was all reactive. Trust me, it was reactive run out there trim, oh my gosh, there's another one. There was really you tried to make it sort of strategic. But now he never has he got run out there, grab the bucket truck, trim it boom, move on. Here is a here's a solution, a iDASH that looks down on your service territory and says, Yep, go right there. Go over here, trim that tree. Yep, go over there and not dunk over there. That's fine. And that good. I mean, that's that is deploying your capital going in your money in a way that really benefits us the consumer. So that, that the service stays up and running and you know, delivering the power that we need AI dash, they're leading the way that satellite AI, vegetation management, simplified baby. Alright, and the other one is Neil. So you go out to me on.com. And I have this conversation all of a time. I talked to innovators, I talked to technology delivers and service providers and everybody and we're talking about digital transformation. We're talking about AI when talking about quantum computing, we're talking about all of the topics that are really a buzz today. And the reality is, is that here's Neil, they're going to put it into action. They're creating a community of the future. And they're going you know what, you know what the best part about it is, they're going to make our lives better, they're going to go down that road, they're going to get the bumps and bruises, they're going to try to figure out solutions that really are meaningful to us. Neil, right. meaningful to us solve those problems. And we as a as a society, benefit greatly. I'm doing it's cool, don't you? Come on, you got to get all go No. jittery, happy with that stuff. That's me. on.com. They're creating the community of the future and they are doing it. They're doing it. Alright, neon count go out there. Alright. So one of the things that I've talked about is getting the most out of right getting the most out of it. Are you getting the most out of conferences? Are you getting the most out of your your engagements, whatever might be getting the most out of your campaigns, whatever it might be? Are you and ask that question and be truthful about it? Are you or is there ways of being able to improve that engagement, that conference, that whatever it might be, and I'm here to challenge you there? There are strategies, there are solutions that are out there, and one of them is okay. People ask me Scott, why do you why do you podcast Well, one, it's fun. Yeah. And and it's, it's available for everybody. But one, it's fun. And two, I get to talk to
leaders around the world. And these leaders are movers and shakers, and they are at the cutting edge of thinking and deciding what's good. You need to be there. You need to earnestly and and I mean, earnestly, there's ways that I'm going to continue to sort of add to this, but at a minimum, look at creating a podcast. And I have a I have a program out there that says, why you need a podcast, and it takes you putting every step of the way, right. But the reality is, is that if you want to get the most out of something, have multiple touchpoints, and an engagement, that that you can offer to your prospects, your customers, whatever it might be, bring life back to a blog. There's no better way. There's no better way of being able to do that. Yes, yes, there's a lot of podcasts out there. What we do at industrial talk is we we recognize that there is this fragmentation that exists out there, right, there is podcast over here, podcast over there, and so on and so forth, great content, they're great content, their solutions here, right? What do we do? Well, I'm I'm, I'm in a game of consolidating that I'm in the game of trying to bring all this great thinking together so that the industry, the community, the ecosystem, has a place to go and says, I want to know about cybersecurity. From a trusted individual, right, they're easy, right? And because typical topics tend to be very niche, a, you know, it's it's, the reality is you tap out the opportunities there. So you need a community to be able to be a part of flight information, to be able to be impactful. And have your dog God message be heard. He got to have it heard, right. It's got to people got to listen to if they don't like, does a tree make a sound when it falls in the forest? If you're not there? I don't hear it. I'll hear it right now. The same thing exists with your voice. You got to do it. And we'll talk a little bit about but I just tell you just go out to the well, this podcast and other podcasts will say why do you need to do a podcast? You got to do it. You got to figure it out. Contact me. I'm on I'm on a dock on book I am. I'm warm and fuzzy and cuddly. Maybe I can be more cuddling anyway. Alright, let's get going. So what's interesting and had a great conversation with Prith. And what's interesting is the the ability Yes, we can collect data, it's all good. Click, click, click, click Collect. But what about that simulation component? Right? It's great. Got all that data? Let's simulate what's taking place what simulate what might happen or whatever, and be able to, without a doubt, begin to converge the data and simulation into a true cogent response or information on when you need to perform maintenance, perform whatever corrective action, whatever it might be. Simulation is, it was great talking to Perez because he's you could tell he's pretty passionate about the whole thing and rightly so. Rightly so. Alright, let's get cracking. You don't want to hear me a moron. Here's Prith talking about digital twin but really simulation enjoy. Welcome to industrial talk. Thank you very much. Broadcasting from Palo Alto. That's where he is coming from. If you're out on the video, you'll see that there's plenty of sunshine happening there. Anyway, how you doing?
They're very good. Thank you very much, Scott for having me.
I'm excited about this conversation. Okay, listeners, we're gonna be talking digital twin. And you're saying Scott, I want to talk? Yes, you do you want to talk digital twin. But more importantly, I think Perth is going to bring a wealth of knowledge that you're just going to be dazzled by man. I know he is. So brief before we get going give us a little sort of 411 a background on who Perth is.
Sure so. So I'm Prith GMs Chief Technology Officer at ANSYS. ANSYS is a modeling and simulation company. We are a company of about $2 billion in revenue 5000 employees we help provide solutions to our customers in terms of when they are trying to design products. We enable them to design develop the most amazing products in areas such as automotive, or aircraft, aerospace, or manufacturing energy and so on. So that's what MCs does and my role as Chief Technology Officer is to set the long term technology strategy for the company around areas is AI machine learning high performance computing? digital twins?
Um, so you just you just threw out a hyper comment. What was that?
High Performance Computing high performance?
Oh, yeah. Yeah, I'd like that. See, you're you. You're sitting in the seat where a lot of people want to sit just because you get exposed. I mean, you must see a lot of innovation just coming. I mean, just constantly evaluating looking. You having a conversation around innovation and the impact it has on industry, society. I mean, it's pretty cool. You're living a dream, you have the
most fun job at MCs, I'll tell you, I, we have 45,000 customers. Every day, I talk to sort of CTOs of different customers across different verticals from aerospace. Like today, I had a meeting with a CTO of a medical device company yesterday, I remain meeting with a 5g company. So I'm learning about the sort of the future trends in these different verticals. And then from my own team members, right, I get to learn about the newest coolest technologies of AI machine learning applied to simulation or high performance computing, or digital twins or, or prime meshing, it is a fun job. And yet, here's
it and I've never seen it. So there's, there's, there's such a speed of velocity that's happening out there. And it's, it's, it's, it's exciting. But man you are, that's pretty cool. So let's start like diving into that. Let's dive into digital twin. Now, for the listeners out there, give us a little just what is a digital twin, and then we're gonna go into some of the challenges and opportunities that exist within that technology.
Absolutely. So in digital twin is a virtual replica of a physical asset. So let's say you have a jet engine of a plane, right. And that jet engine originally gets designed in a CAD tool, right? You say here is an engine, he has all these blades, and so on, and you design that jet engine on a CAD tool, then you actually manufacture that jet engineering gets installed on a United Airlines plane and this jet engines flying around. And so right, that is a particular jet engine, now you try to make a model of that engine. And oftentimes, people confuse, oh, you have a simulation model of the engine, that's a digital twin. That's not that is just part of it. So a model of a physical asset, like a jet engine, is one thing. The other thing is there has to be two way information flow between that physical asset, the jet engine, and the model. And that's what makes it a digital twin. So you have the jet engine, you put sensors on a jet engine, right, and you have you collect all those data for the sensors, you upload to the cloud, or whatever you are pulling that information. And using that data that you have from that jet engine, you try to make the model of the jet engine, as accurate as you can write with respect to what is in the real thing. So it's like a virtual replica. So I'm praying, I have a model of it. And this model of it keeps getting updates right here is with really excited, so the excitement level is going going there, right? Bridge temperature is what 98.4 Right? So it's constantly measuring what is happening on the real asset, and updating the model at the same time. So a digital twin is not a model of a jet engine. It's a model of a particular personalized model of that particular jet engine. Right? And how that is going on in time. So this is something that NASA used many many years ago when if you have seen the Apollo 13 The wonderful thing sort of the the picture that the movie about Apollo 13 Right when that that Apollo 13 had an accident, right? Yeah, people in NASA in Houston, right? Hey, Houston, listen to his to Right. Um, those engineers at NASA actually built a digital twin of Apollo 13. And he said, Oh, if you do these things will happen. So essentially the, of what was happening and sent all the instructions to the astronauts, and that's how they got in there. So she
did that. You just brought it to it. And I guarantee a lot of people are listening is like, yeah, Apollo 13. I remember him sitting in the capsule, switching and doing everything and making sure the amperage didn't, you know, whatever it was, whatever that target was Do this, do this do this. But that was that makes
complete sense. What was happening in Houston, right? Yes, a physical asset out there in space. And they had a model in Houston. And essentially all these scientists were trying to do stuff. What if this happened, this will blow up. They said okay, this would work. And they send those instructions and that's how the astronauts are brought back. So Go digital twin back in so long ago, right in the in a but essentially, the digital twin technology has come a long way since. But that's what digital twins are. It's a model of a physical asset with Two Way information flow between the user and the model.
And why is that important? It just gives me the individual able to look at the, the, the the characteristic of that asset that that real asset in a digital environment and see what's happening. And then I can I can, I can tactically move forward in some decision making or whatever it might be.
nicely. So let me give you an example. So we, we all rotate tires in our cars, right? We rotate tires every 5000 kilometers, miles. Why is that because all the tire companies have seen statistically, if you drive your tire out on a normal road, normally normal driving, your tire will wear this much any and this is going to happen. Except if it breaks driving crazily on us 101 in Santa Fe with lots of I probably need to rotate my tire every 3000 miles, right. So that's personalized driving. So you take the statistics of all possible cars, all possible drivers. And that sounds people say you should change your oil every 20 10,000 miles and changes. That's called preventive maintenance. Right? So you Why do you do preventive maintenance, you replace your oil, replace your tire, etc. So that your tire does not burn your engine doesn't conk out in the world of so there is this preventive maintenance that people do Now imagine if you could predict exactly when that engine would come. So just the day before you replace the oil, that would save a lot of money yet question is can you predict accurate? Yes, see, and that's what the digital twin comes. So the what the digital twin does is you have the physical asset, the engine or a car or medical device or whatever, right? And you are monitoring it and you are getting real time information, as opes is about to fall sick. So we humans, we fall sick, right? Before we fall really sick, you get a temperature, you can oh my day, I'm feeling really lousy, I'm having an A. So those are signal area about a false and then when you get one or three temperature you are boom, right? So imagine if you could sense things around you, you could do predictive analytics. And that would be huge. Because suppose you have the data center for Google or Amazon, imagine that, that data center going down, that would create tremendous waste, right? Five minutes of downtime on Google, or Amazon would be hundreds of millions of dollars of thing waste, right? If a power grid goes down lectricity goes on tremendous waste. So people in these essential industries, there are critical infrastructure assets, they don't want them to go down. So what they do is they try to predict when this is going to fail, and before it fails, do the corrective action. So the real deal here is can I predicted accurately
that even see that is because I lived that right? I was in that world, I was in the world of maintenance. And when you spend a maintenance dollar up here on the financials, it's $1 Down on the bottom, right. And if you could sweat the asset, and sweat it in such a way that you know when to perform that maintenance before it catastrophically fails. I've saved that much time, that much money. It's and you don't have a catastrophic failure that could put you out for who knows how long you know, whatever it is, how do we get there prep? There's that we've been having that conversation, people understand the value of that predictive capability. When do we really start to see it happening commercially?
So I will actually tell you sort of the first the technical challenges then the business challenges so the word again, again, there's this thing called the Digital dream consortium that I am a part of and there are some founding members I Microsoft and GE and handsome so right so I've been in this business for for some time now I've sort of lifted and so on or prior to MCs I did not tell you. I used to be CTO at ABB CTS right electric and when I was on those companies, I was helping build digital twins of the assets of ABB and Schneider Electric right so they have a transformer for ABB. You collect all the data right from sensors and you just by collecting data, you try to build a digital model of that transformer in the normal business havior and the apps as soon as it starts getting signals of abnormal behavior is out bad things are happening. Shut it down. Right? So the most of the world uses digital twins based on data analytics alone, right? So yeah. And they collect lots and lots of data, right sort of
tsunami, tsunami
of data from those data. They I mean, they use AI machine learning techniques. i The older the world knows about AI machine learning you you train those, sort of these, these convolutional neural network, CNN models of the 60s neural network. So there's a lot of work that people have done in this community using AI, ml, CNN, etc, to build these digital these digital twins based on analytics alone. Now, when I was at ABB, and Schneider Electric, we used to do all those things. And we found that you know what, it was not that accurate. So, yeah, it said, 70% accurate. So he said, Well, 70% is pretty good. You have a million dollar asset, you're 70% accurate means you made a 30% mistake. 30% mistake is $300,000.
That's, that's a lot of money. That's
a lot of money. So this is what the challenge was. So in the industry, people said, Oh, if you did pure analytics, right, it's not doesn't look good, does not always work. Now. Analytics works in cases, where you have enough data on the use case you're looking for, right? So four frequently occurring events, data and this goes well, for infrequent events, you cannot predict it, let me give you a very bad example. Space Shuttle Challenger exploring how many times has a space shuttle extended once. So you can get all kinds of data on space shuttles based on and so on, he would never be able to predict when that space shuttle would explode, right. So that is an infrequent event. So when an event is happening all the time, right? You have little things that are little non bolts or whatever, there are sort of breaking down in a in a this thing. frequently occurring events, a IML is great at catching infrequent events, you cannot go there. So this was my aha moment. Oh, okay. So what can we do? Therefore simulation, so now all around us, is governed by Singlish. governed by the laws of physics, which get modeled really, really accurate. Let's imagine you have a bridge. And you're trying to predict when that bridge will collapse, right? So that you can shut down the road, right? So you put all kinds of sensors on the bridge? And you say, hey, is there a crack on the bridge or not? So first of all, you can say, How long do bridges last 30 years. So he said, predict preventive maintenance on the bridge every 30 years, every 30 years, you go and fix the bridge or whatever. That's the preventive maintenance way. The IoT way is to put sensors on the bridge to see is there a crack on the bridge or not? Right? So let's imagine that there are cracks on the bridge, I said, Here is a crack that is 12 inches long today. Now you have no idea where that crack would go next week. Aha. But if you had simulation using say structural analysis, finite element analysis with ANSYS tools, I am telling you, Scott, I can predict with 1000 cars going on the road with this kind of speed and so on, what is the stress sprayed on the bridge? And if and if there's a crack that is with steel pillars and so on with is 12 inches, I am predicting through simulation that this crack will grow to 18 inches in a month through simulation.
So if I did that through simulation remand as I said it pure simulation, I would say this crack which has 12 inches long images tomorrow 44 inches at 36 inches in the crack, I will predict through simulation alone that the bridge will crack in a month. And that is a prediction. Through simulation alone, we have found out that the accuracy of this 70% goes up to 90%. No way. Okay, so now you have this million dollar part with physics based simulation. We have a tool called Twin builder, we were able to increase it from 70 to 80% over 90%. So that million dollar part your waist is now only 100,000. Here's the only 100,000 and still adorable. So that's where we have this aha moment Right? Where we said let's do hybrid digital twins. Let's combine that data analytics with the physics so now I'm going to bring it together whether example, here is this bridge, we had a track we had a crack which is 12 inches long, right to data. I measured it through to data analytics Right. And I predicted it will be 18 inches long a month from now. But a month from now I take the measurement, and the crack actually was only 60 meters long. So my simulation model was making a prediction, it was going too fast. So my data is a go slow, so it goes slow. Or, for example, that actual measured crack was 24 inches long. So that means my simulation was going slower. So by getting a, a updated information from that, that's why I say remember that the two way information flow, I pulled data from the bridge about the size of the crack. And I made my simulation model more accurate, synchronized with the real thing. So as I change the rate of crack growth, right, instead of making it 18 inches, I cracked it to eight, six inches, or 24 inches, whatever is going on the next month, when I read it, I will be a little more accurate. I'll say it should be 32 inches. And next time I say no, it should be 31. Oh, so it slowed down even more. So through this hybrid approach with a combination of data analytics, and simulation. Scott, we have shown data that the accuracy goes to 99%.
No way. Dan can, can you because this, this is just cool stuff. Don't get me wrong. I think this is cool stuff. Can you take that bridge analogy and you're you're you're honing, you're tweaking? You're you're zooming in on accuracy by data and simulations, and you bring it all together? Can you take that and say, Alright, this is it. This this bridge is in Texas, let's go to, you know, Florida, and and sort of take that learning that, that that simulation and apply it. And again, yes, tweak, tweak, tweak different weather patterns, a little bit more speed of the cars, whatever it might be, but be able to use it there too, as well. Absolutely. And
so just like sort of just the AI to like Siri, right, your your, your little tool. I mean, if I saw Siri, when it listened to you, Scott, you have an American accent, I have an Indian accent. But Siri, after listening to me for a month, it gets pretty good, right? When I say go do this, I mean, so the machine learning things get better and better, the more data it gets too similar is hybrid digital twin get. So there is something you can do with simulation with this basic version. But the more data you give, the better the model becomes. And this is sort of the state of digital twins that we are I am seeing from from ANSYS. And I'm talking to lots of lots of customers across different verticals, we have got sort of applications in industrial flow networks, working with oil and gas companies, applications in in, in electrification, with battery systems, and so on applications in HVAC, and buildings and so on tremendous opportunities across all verticals,
where does it end? And when I say that, let's say I am, I'm at 90% accuracy, I can measure it, I've looked at it my simulation, Mike data on there, and I'm tweaking, and then I put more time, energy and effort, I'm at 95, and so on and so forth. There's a point where it's like, Okay, we're good, we're good. And any any additional incremental benefit is just like, minuscule.
Yeah. And so that is sort of it is the law of diminishing returns, right? You plot a graph, and you see keep seeing, making chimey, at some point is flattening out. And so it really depends, right? For that particular engine, or wage or flow network, or valves or whatever, when you see the law of diminishing returns, you're not really getting incremental stuff, right? You sort of stop there like that, that's, that's what we have done.
And and, wow, it the applications are quite frankly, endless, because you get it to you get it to the point where let's say it's a motor, right? You do the same thing data simulate and and I and your analogy of Apollo 13, I see a guy in a capsule running a simulation, trying to get it just right. The same thing exists, you could be on that asset, no pump, motor, whatever it is, but you can apply it and it might be different in Texas as it is in Maine, right is because of the different weather whatever it might be different application. But
absolutely, yes. Now, let me tell you, I talked about physical assets. But you and I we are not we are not just physical asset. We are humans. Right. So let me talk about the opportunity of digital twins in the healthcare industry, right. So inside our body's most complicated organ, the human heart inside the heart, right you have a sort of this dielectrics signals that pump, right that generate the signals 3070 times a minute, right? So in response to electrical signals, the muscle heart muscles contract. So when the heart muscle contract, the atrium, the blood flows from the atrium to the ventricle, and then the ventricle for this thing, and the valves open up. And so it's just most complicated organ in the human body. Guess what I mean, ultimately, those human models are modeled by physics, right, there's electrophysiology, there's mechanics, there's fluid flow. So we are dancers have actually model the human heart. Now, there is a generic human heart that you can model and you can say, Okay, this is what is going on, we can also model treats human heart with as arrhythmia, right, and I have an irregular heartbeat problem. So it goes into a doctor takes an MRI scan, CT scan, whatever of the heart, and He measures all those things. And for prints specific heart, we can come up with well, if I have a B or whatever you can do that things you can do heart surgery or so on. Right? The other thing is, I mean, you can put in a implant, right? A pacemaker, right? So how will this pacemaker from a company like Medtronic interact with my heart, right? So a Medtronic company, that battery will have to do clinical trials of all kinds of pacemakers, and all kinds of patients and so on, right? But through a digital twin, of a human heart, right? You can create all these different models, and essentially, shorten that time compress it big time, or for trials of a pacemaker, right? And if so, that's kind of where we are heading, we are headed in terms of what is called in silico. Trials, right? Computer based simulations of 1000s of patients. I mean, the future is amazing.
Ah, see, see that that delivers, that deliver solutions, faster? Health, whatever, the health side, I just didn't even think about, but you're absolutely right. Create a little digital twin, correct. And you can run simulations, but it's, it's individually based, right? It's your heart. My heart is different than yours.
Like Fitbit, right? The Fitbit is collecting. Exactly you ever saw collecting data on you, Scott? 24/7. So imagine if your doctor had a model of Scott, sitting inside, right, and he's collecting all this data, right? And God bless them use my example, right print a model of prey, and just before previous about to have a heart attack, right, that doctor just two hours before gets to know a signal Hey, there's this the signals are going in the wrong direction previous governor calls of the ambulance and sends an alert before he has a look at them if the value of them
Yeah, yeah, that's that's cool stuff, my friend. Yeah. Why? What's what's, what's the, you know, the roadblocks? I mean, this is all good stuff. Let's make it happen today. Let's let's make it happen tomorrow. Let's put it that way.
As you know, Scott, I serve on the advisory board of the IoT solutions World Congress. In fact, we have a conference coming up in May of of this year, right.
And I put that on your calendars listeners big time.
I serve on the digital twin Consortium, and still there, this, it's a more than 300 members in this consortium and so on. So people are are excited about the potential of digital twins in different verticals, in manufacturing, in factories, in industry, for Dotto and so on so forth. The thing is got that shiny toy looks like a shiny toy. Oh, you have fantastic possibilities, right? But when you start actually pulling it together, right? You have to collect all this data and so on, and does it work in all the time and so on? So many customers are saying, oh, you know, I like the shiny toy. Let me use the shiny toy in a proof of concept. So you do a proof of concept number one, you waste like three, four or five months, maybe $100 million project and so on the like this. Oh, let's try this other games. Right. There's two POC number two then POC number three. So so people in this our Digital Consortium talk about the POC, Purgatory right. Before moving to scalable systems. So the digital twin thing is they going through this sort of inflection point where it is a lot of promise, but it has not quite seen scalable value across all industries in all opportunity. And that is I think what the business challenges today,
I agree with you, but it's gonna happen. It's gonna happen it's gonna be in commercial use. It's
like a $26 billion market by 2025. So yeah, it is going to happen is about a couple of billion, but it will happen.
It makes sense, right? There's no, I mean, it makes sense. You got to do something like that. That's make far right. We're gonna have to wrap it up. I hate to because I enjoy this conversation. How do people get ahold of your prep?
So they can contact me at my answers email address Prith.Banerjee at PRI th got Bannerjee ba NERJE. App. ansys.com. That's a NSY. Yes. Yeah, you can search me on LinkedIn, at LinkedIn Dash Grip Banerjee, or a Twitter at Big Banerjee.
Alright, listeners, you don't have to worry about it. I'm going to have it all out in industrial talk, you'll be able to get a hold of this. This incredible professionals with no problem and don't come to me then Sen. Scott, I can't get a hold for it. You're lying. You're lying out there. Just don't do that. All right, Fred, you were absolutely spectacular. Yeah, you your energy's gone. It's exciting. I love watching you just get all giddy about this stuff, man. It's cool stuff. Good job. Your job is to be a
professor for 20 years. Really? Yeah,
that's a hard job, by the way.
Oh, that's where I, I learned how to help lecture.
Wonderful, wonderful. All right, listeners. We're gonna wrap it up on the other side. Once again, we'll have all the contact information for prep, and then some. So stay tuned.
You're listening to the industrial talk Podcast Network.
All right, thank you very much for joining. Once again, industrial talk. Thank you. Thank you very much for saying yes. And being on industrial talk. Love the conversation. You listeners need to reach out to prif go out to a stat guard go out to LinkedIn, you will not be disappointed or go to industrial talk, find the link, reach out to him. You need trusted professionals that are at the forefront of thinking through all of this Firth is one of them, reach out, do not hesitate to do that. All right. I'm going to continue to challenge you once again. That is are you getting the most out of the bar. The bottom line is we need you. You need to be involved. You need to have that business of resiliency and you need to do a podcast. You need to create one you need to figure it out. Again, go out to industrial talk. I'll give you the step by step everything you need right there right now. No excuse you need to be a part of that. Alright. People Be bold, be brave, dare greatly hang out with people that are bold, brave and daring, greatly industries full of it. You're going to change the world. Thank you. Once again, we're going to have another great conversation right around the corner.
About the author, Scott
I am Scott MacKenzie, husband, father, and passionate industry educator. From humble beginnings as a lathing contractor and certified journeyman/lineman to an Undergraduate and Master’s Degree in Business Administration, I have applied every aspect of my education and training to lead and influence. I believe in serving and adding value wherever I am called.
