Robbrecht van Amerongen and Remco Bunder

SUMMARY KEYWORDS

assets, digital, industrial, data, twin, escalator, passenger, model, station, palo alto networks, company, drone, remco, elevator, platform, combine, netherlands, organization, 3d model, robert

00:00

Industrial Talk is brought to you by Palo Alto Networks. You've heard me talk about my friends at Palo Alto Networks Zero Trust OT security delivering a comprehensive security solution for all OT assets, networks and remote operations. But did you know that the Palo Alto Networks solution provides over 1100 app IDs for OT protocols, over 500 profiles for critical OT assets, and over 650 OT specific threat signatures, now that's best in class security delivered, learn more about Palo Alto Networks, zero trust IoT security solutions, and how you can achieve 351% ROI over five years, learn more about the Palo Alto Networks solution by going out to Palo Alto Networks.com. That's Palo Alto Networks.com.

00:56

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 Hey

01:14

there, and welcome to Industrial Talk. Thank you very much for joining a platform that is dedicated to industrial professionals all around the world. Because you're bold, you're brave, you're daring greatly. you innovate, you collaborate, you solve problems. And therefore you are changing the world as we speak. Thank you very much for what you do. We have a great conversation, it is digital twin in action. It is how the Netherlands is using digital twin, and collecting that data in a way that they can make the experience as a whole, much better for all of the wonderful writers of that service. So that's on and I mean, we just barely scratched the surface, we're going to be doing a webcast on that. So that you understand a little bit more get dive a little deeper. So let's get cracking. Yeah, I've turned this one around quickly. It was such a great conversation, it was so interesting how they are using and collecting data really doing it in such a way that it's it's real time. It's insightful. And they have some challenges. But they're, they've overcome it. It's just it's it's a wonderful story and and exciting. So that's why we're going to be doing a webcast too. Just because we need to dive deeper, podcasts are fine, but you can't really dive deep into it. So. But anyway, this is still, this is meaty, with a lot of great little tidbits. So make sure that you have paper and pencil and take some notes because it's pretty cool. All right. Now, a business note, you need to amplify your voice. God, just get it out there. You need to amplify your voice. And Industrial Talk is here for you. You just go out to Industrial Talk.com It's easy. And then you see the button. And then in that button, you click Yeah, let's collaborate, let's talk let's do whatever. So that we can have you and I or your company and I or whatever it might be to have a conversation about how we can amplify your voice, expand your network and create more opportunities for your organization. How about that? It's all there at Industrial Talk. It's easy, easy, navigable, is navigable. Yeah, navigable. That's a word, right? I didn't I just didn't pull that one out. But anyway, it's there. And you do we need to increase your your, your reach, expand, get into new markets globally. Fortunately, Industrial Talk is in a position to do that we, we've invested a lot of time, energy and effort to be able to do that to make sure that industry as a whole is well represented. And you need to be well represented. It's all out there go out to Industrial Talk.com. And let's let's come on, let's just have a conversation. It's easy peasy. All right. On to the conversation that we have now, again, this is this is a relationship between the rail system and the companies that is out there in Netherlands and an organization that helps provide the ability, the platform the the insights into the data that is needed. And it's just great on how they decided that they wanted to go down this road and be able to do this for their assets and their business in general and to be able to have that insight to truly improve the experience of the riders. It's it's a it's a really great story. And, again, be on the lookout for the webcast just because it's a it's got it there. There's a lot more. I love the digital twin in action, and it's in action. And it is moving. And it's doing stuff in the Netherlands. So let's get on with the conversation. Enjoy it. All right, gents. Welcome to Industrial Talk. Thank you very much for finding time in your schedule to be on Industrial Talk. How're you guys doing today?

05:20

I'm very well. Thanks.

05:21

Very well. Robert. Good doing well.

05:24

Grant. Yes.

05:25

All right. Very good, man. All right. For the listeners out there. Let's get right into it. Give us a little background, Robert, on why you're such an incredible professional and then we'll go to Remco.

05:35

Yeah, thanks a lot. Yeah. Well, I'm Robert phenomenon. I work with Amazon collusion, that's the Dutch IT company when we specialize in real time data processing, specifically for industrial solutions. And more specifically, on the real railway and real solutions. We're working here for 17 years is it might be very interesting job to be here for so long and constantly busy with involved in innovative digital projects within a company would you say

06:07

digital projects would give us an idea of what what digital projects are

06:12

not everything where decision making this needs to be done almost real time based on accurate real time data. That means in energy management in management of large buildings or railway stations and airway terminals, and also based on feed industry, is everything where data is derived from process decisions are being made. And then the decisions are being fed back towards the operational systems.

06:47

Are you specifically saying that data around facilities like like rail facilities, airport terminals, or do you get into the actual assets themselves, in the case of, of rail, that you put in devices, reading the data, looking at it from a real time, sort of, like set up? Yeah, and

07:10

most of the assets, which we connect to are already digital, or have some some form of software in terms of PLC and SCADA systems on top of it. So we connect to those systems and make these data this data available in a centralized system. So you can run some logic on top of it and combine it. And perhaps we will get later on in some examples where you talk about the digital twin, combined is data and make decisions and then also operate via activators actually steer this these devices again, the run faster or slower, and in different different modes.

07:47

So Amis conclusion is, I'm taking existing infrastructure, data collecting devices, like you said, SCADA, PLCs, and so on, and being able to pull that in and then being able to put it into your platform to be able to draw conclusions and see, you know, or derive insights into that.

08:08

We actually bring the physical data which is spread across our whole country, to a centralized place where the people are the experts that were working in offices, can make an analytics on do analytics on top of it, make decisions, and then actually program these decisions as these logic in the system and then derive the conclusions back to drive the conclusions back to the actual systems again, the essence again,

08:38

okay, I got the picture. All right. All right. Remco, bring it give us background. Well, that's a hard act to follow. By the way, just FYI.

08:50

Amsterdam, in the Netherlands, where you told us you already left for some while Well, I work for Dutch Railways. But I work in this specific area, I work for the real estate department. And we have like 40 to 40 stations. I have several roles. But one of them is product owner for a team that's concerned with Internet of Things, solutions. And the other end there's a part where I'm involved in the drone organization of the Dutch Railways, where we because we also use drones to harvest information. Which we can make 3d models, such as we did for introduction route, as you know. What can I ask, can I say I'm 54 years old, and well, with

09:43

a flowing header hair, that's what I see. Of course, I don't so that's that's it is what it is. Just let me ask you this real quick from from the Dutch railway perspective, this whole digital transformation of your assets It's your facilities, whatever it might be. Take us through sort of that thought process of where you were prior to all of this, and why this is such an important component to your strategic, you know, business?

10:18

Well, it started off like a simple Id like

10:28

we bought this thing once. And this was the first drone. We we got. That's a drone. Yeah, well. It's a very small one. But it's, it's cool. I pitched the idea of, okay, why are we still using letters? And why don't we buy drones? And then well, I showed them this. And I said, Well, let's, and they said, well give it a go. And we bought drones. And when we bought drone, we also saw the possibility to create 3d models, from our assets with the software that was involved in these rounds. And pretty soon, we came up with the Id like, Okay, if we have a 3d model of our assets, and we could also connect it to real time data. It this could be very useful, because we have like people who have to manage more than 10 stations, but they cannot be at the same place in one time. And that's where I thought, Well, let's start very easy and small, with a very tiny station somewhere and with one elevator, and where my partner and colleague, John kilchis. said, No, we take the biggest station we have. And we're going to make that 3d. And we're going to put sensors in it. And we're going to fix the real first 3d, digital twin of a train station. And well, that's what happens. And the so we didn't have a business case in the beginning. We just wanted to prove we could do it. And because we had no idea of the value, yes. Yeah.

12:15

Because but you, boy, talk about taking a big bite. Which one is your largest train station?

12:25

interdiction route? We have like 2 million passengers. And that's big for Holland. Every day going there. So 2 million every day. Yeah. Passengers. Yeah. That's a really big hub for

12:44

Okay, I have to pull on that. Because we're going to be talking about digital twin. I got it. I got it. And what did did it surpass? I mean, you just started using drones, you started taking the pulling information off of a 3d type of environment, digital twin it right. It will take us through the process of just say, We did it. How long did it take? I've just swirling with questions. How long did it take? Because that doesn't happen overnight?

13:16

No, no? Well, first we started like, Okay, we do need to harvest the data from the inside and the outside. So you have several techniques to create the 3d model. So let me take you to the 3d model. If it's okay with you. Yes, please roll with it, baby. Okay. We we hired a drone that was equipped with a camera, and we added flew over Utrecht, and mate pictures, which could later be used to use the technology called photogrammetry. And which you can make a 3d outside of an office building. Then for the insights, you cannot fly with a drone. We used LiDAR, which is a technology like Shona, you send the bills, it gets back you have API sets, and then that connected to a geometry, geo reference. You can make a from a point cloud, you can make a 3d model and then we add the challenge to put this outside and the inside on each other. So because we have the outside of course, where what's what the passenger see, but on the inside we had the elevators, the escalators, the stairs, the whatever the shops. And well I think we in total, we spent and it's not like non stop, but It took about three months from out any call it delays, the processing of the model took a lot of times, because it was like 20 billion dots, which had to come together as a point clouds, and then you have to decimate it. Because what because it's so big, you have to scale it down till it's visible enough, but small in size. And then we have the 3d model, then, what's the next step was we have the escalators and elevators? Because that was for us? Very well, they were already there. And they were already equipped with sensors, how can we get the data from these devices towards our 3d model. And then with the help of the team have a nice, we created a IoT environment with a event, etc, where we had maybe a ROB ref can explain a bit more about it. But okay, data comes in with the status of the elevator or escalator combined with a status combined with a location. And because we had the location, we could point it on the 3d map. And that's where westward are, are, in our opinion, digital twin was accomplished. Okay,

16:35

so with that said, Robert, I want to sort of for the listeners out there, let's let's sort of level set around a 3d model versus a digital twin what give us give us sort of a, you know, we talked about 3d Going into digital twin. Talk to us about that.

16:55

It's there's a lot of different definitions of digital twin. And, and it's sometimes the infused from marketing perspective, or commercial perspective are very technical perspective. In this case, there is a physical model of a building or an a machine in disguise a building, where you actually plot the actual location of assets, who are creating data, because they're still immutable, or at least fixed for a couple of months until you replace a wall or put an escalator somewhere else or put your life somewhere else. And then the trick for a digital twin is that it's being fed by real time data or near near real time data. In this case, it was sufficient to send data every 15 minutes that was sufficient to to assess the status of the station, we also have digital twins, where the data is being sent into the model every second, but that's more an energy, energy management and energy systems. So the combination of the location, the structure of the building, the physical attributes of the building that's being simulated in the model, and the actual data, you can see what is happening in this building in real time in your office. So in this case, the station manager was actually the conscious of the station that too, was responsible for everything within the station. And in such a big building, he can be everywhere at the same time. So we put some alerts in this model just to show Well, are there specific locations where the surface of electrical equipment or other assets are below normal, so that they don't reach their service level. And sometimes, it's acceptable. So when you have two escalators, and one is not working, you always have the other one. And, but when there's two escalators broken, then that's that's a high priority incident. So you have to act on that one, immediately. That's being shown in the model itself. And having this this this data, you can also combine it with other data like energy usage, with the passenger flow, and then create models to optimize for for example, sustainability, or passenger quality. So experience in the station, a high value with experience are optimized for retail revenue. So when you have a very busy platform, it might be very interesting to open another place where you can buy extra coffee

19:42

and especially this elevator, escalator, problematic. problematic if you are, have a disability and you arrive at a platform where both the escalator and the elevator is broken.

19:57

Well, that's not good. If you don't want, you don't want that it's not good. Yeah, sadly. But let me ask you this Aramco, when when, when we've got this, this vision, this this vision of your assets, polling data on a real time basis, did it help you to perform the maintenance that as necessary, the degradation of those assets in a way that you can proactively minimize the outage of maybe that escalator or that elevator, compress that time? They're going to? There's going to be events throughout the year, whatever. But did it help give that clarity into those assets? To do that?

20:42

Well, it did. But the power of it more lies in the visibility of like, if you have a platform, where boats, like platform 12, were about the escalator, any elevator are broken, this does something to the urgency of fixing that problem. Because that means there is no other way for passengers who have a disability to get where they want to go. That gave the station manager the opportunity to proactively chase that case, like Okay, listen, I know the priority normally for escalators priority to whatever. But now I see that I have the escalator elevator on the same platform being broken. Yeah, it makes it a higher level. And that's, that's visibility. So that that you see it in the context of that environment made it very valuable. So the only erase the priority, more or less, the head, Robert.

21:53

And there's also an added complexity in especially in kind of organizations like this railway, the ownership and responsibility of assets is kind of complicated, depending on what location you are. Sometimes there are two major companies, one which runs the trains that were in close is working is also responsible for the real estate, the rest of the train station, there's another company called ProDOS responsible for the actual tracks. And sometimes assets are owned by the track company, sometimes assets are owned by the rail company. And sometimes assets are owned by the local municipality, so the local city, so I wanted to spread his ownership, you also have probably different kinds of systems. So you have to connect to all kinds of different OEMs. And it's kind of difficult, it's quite difficult to get clear insight into total status of complicated station like routers, intro, we have to combine all this data in one platform. So in before we made this digital 3d digital twin model, people had to view four or five different systems perhaps even more, just to check what the status of the station is. And then you still have to consider that the data you're looking at is has a different time frequency, sometimes that is being refreshed every month, sometimes every day, sometimes every hour, and now it's all been harmonized and centralized. So you can see, well this is broken now. It's on the same platform, the two different OEMs and now I have to talk to these two OEMs to make sure that they are fixed.

23:41

See, how do you deal with with all the responsibility and ownership challenges? How do you train how do you educate on let's say, I know I'm just I own this section of track, whatever it might be. And you come in and you say hey, we're gonna give you greater visibility into the health of that asset because we're going to be pulling data and it's going to be real time or every 10 minutes or whatever the parameters are in. And and I have to I have to learn how to how to Ramco how to how to deal with the the education component and say this is important for you and it has value for you. Therefore you need to take this serious or whatever it might be but take us through that I

24:33

well it's not a nice story when we first started with the idea of the digital twin our business said well oh there you have it guys and they're always playing and yeah we what does the What does what what value does it have etc and I always make the joke to my colleagues who are with the maintenance department. Yeah, well, you will, you still like to work with the pencil, but we have to move on. And so in the beginning, without people knowing what digital admin could mean for them, there was always, always already some resistance. Especially when we got the award in Barcelona got worse.

25:34

That's so funny. I, it's, you know, every time, and this is just from, from my experience, and all the conversations I have with innovative organizations and companies like yours and others. It always gets down to the human equation, I don't care. And it's global, right? It doesn't, you know, one segment of the country in the world. It's always it's, it's global. And it's always people. And it's always that and that that ability to succeed or fail, or challenge, it's always gets down to the people. But where are we at today? I mean, from from, from bringing it in and using it, give us that maturity component,

26:26

I can shed a little bit light on that, that it actually depends when she says on the type of people you have in your organization. And I think every person, every organization is adding value to this organization. And they need to work together to make it really make it happen. So probably in every organization, you have, several people like me, Aramco are running a little bit ahead of the troops, creating new ideas and doing stuff, which isn't really existing already showing that it can be done. And this guy's there, we just made the digital twin model of the station, just show them now this is what you can do. And this is what is additional value. And now the next step is make it make it so that it's being reliable, trustworthy, and that you can rely on this model every day and every second that you have need to make decisions and not that it's a proof concept. And that's actually where we want our specialization is to accelerate and execute on innovative ideas, and to expand them to real productive systems with large scale skill with large scale and large number of assets. So it's not a proof of concept or a pilot, it's a really productive system, which is able to be the basis and the reliability foundation for every person in his organization to make decisions upon. Probably that's the gap between innovation and actual, large scale production that's quite difficult and complicated to make. Because then all the complicated stuff, which you ignore, when you do innovation comes on the table like security, privacy, scalability, reliability, and so on. And that's just as important as being innovative and creating cool new ideas. This is the cherry with which makes this this really valuable for Yeah,

28:26

here's the funny thing about your, your rail approach, and the assets. And it's the, it's not, it's not a clear, you know, it's it's the asset basis extends out and it goes everywhere, and it involves so many different people. I'll just have to say, what is the real value of going or entering into this digital twin world with your real assets? What's the real value? What am I seeing the bottom line? Better maintenance better, better, better this better that financially? This I can you bear the money was on that one?

29:14

I think we better maintenance is surely one of the things in response like that we can quickly quicker respond to the disturbance we have in the country. And this doesn't have to be in 3d this might as well be in 2d. Plus, you can also make a better assumption on the impact. And I think that's also of great value. Because like for a train company, our main goal is to make sure our passengers get from A to B shape. And therefore, these escalators And the elevators are? Well, not it's not our core business, but they are part of our business. And therefore, I think they are not how much euros Can you attach to having a passenger getting on the right place, I don't know. But in the end, if people feel they're comfortable by traveling by train, and because all the escalators and elevators are working, I think that's beneficial. And

30:33

those are high use assets. I mean, they just are high use, or you're running all the time I use assets require constant monitoring.

30:42

The business cases in this area are that's also a complicated part, because the business guys arise from different departments, and different strategic goals of the organization. So and the rail company is a publicly held company. So they have a public service to the Netherlands to make sure that it's up the trains are operational, and also that the service is high quality. However, they also have committed to the various goals of carbon reduction. And they make the need to make their operations efficient enough, so they have enough money to invest in new projects and business continuity. So whenever you have a platform where you can combine data from all kinds of assets, you can optimize in all these three areas, with property, the same data or a little bit more data, adding different kinds of logic. So in this model, we are now building models where we can add energy uses data on top of this, this model and escalators and elevators are assets with a high energy footprint, then you can combine it, for example, with the passenger flow and say, Well, is it really necessary to have these assets running when the amount of prestige is very low? That could be saving in an energy where you can see every escalator is the same. But number three is using two times as much electricity as the rest, they are functionally the same operations. The problem is something is wrong is and at the moment, that's the data is actually monitored, that is monitored by the OEM. And they act when there is an degradation of service in the service level agreement. But they probably don't act on Texas energy users because they're not responsible for that part. Yeah,

32:39

it'll be an interesting case. Because if the power consumption is higher, that this could mean that the doors are having to make push more open and close. So it could be a predictive index, either whether or not broken or not, but you have early warning system with that.

33:01

Yeah, no, that's a, that's a clear indication that some, there's stress going on in that asset that that needs attention. And maybe you can proactively take the asset out, do the maintenance is necessary, and not run into the situation where you got an elevator out and an escalator out. And and then the quality of the the riders experiences shot because it only takes one time, only that person comes out. And it's like, Oh, what am I going to have to do? I know, it's just it's, oh,

33:33

no, I can tell you. We had, unfortunately, a few cases this year, where passengers are being well stranded at the platform and not getting down. And we make headlines and that headlines we want to make so

33:51

yeah, I agree. That's something so Remco Where do you see where are you at? Where do you see the next step? Where what what are you looking forward to as from a strategic perspective with the rail? Well,

34:03

I want to add more sensors in the area of safety. And like with AI combined with with audio, maybe because we also had a proof of concept earlier on the gravity problem we have in the Netherlands. For some reason people find it very funny to put gravity on trains and train stations and it costs us a lot of money to clean this every time. So we made a sensor, which was able to detect this kind of actions. And if I can put that kind of sensors also in my digital model, or like you could say to that is a big station. If it would respond to crowd cheering or a gunshot. It would make the first responders it could be helpful first for the first responders. To indicate where they have to go to be without having to run through the entire station. So I think that's the area in the area of safety, I think the biggest challenge, new ideas would come.

35:13

Very good. Can you add to that, Robert?

35:21

Also, some undetected, slow degradation can be added to this model. So then the other thing is network a little bit less, the smaller stations, most of them also have a kind of basement or the foundation, when there's water leakage, or when there's high water in the Netherlands, sometimes happen that the basements are filling with water. And when you place the center, and in that area, you don't have to go there every week, or every month to check if the basement is filled with water. So you can detect it the perimeter prematurely, and see if you can stop it before electrical wires being damaged, which is placed in these basements, and other stuff. And you can see what happens. Also in simulation, what happens when you change something in the passenger flower and design of the building, you can see, well, if you move kind of a couple of benches a little bit more to the left, what will happen to the passenger flow? Okay, can we still process this with this amount of people? And will there be a difference in the the stress which been put on the escalator, and so on? Is there sufficient lighting to support these people? And all this, this is quite sounds quite simple, but it's derived from four or five or 10 different systems. And now you can combine it for simulation in one big centralized system, which helps to answer questions in terms of sustainability, but also investment in answers to various regards to revenue or passenger. Passenger quality. There

37:00

you go, Hey, we're gonna have to wrap this up. Remco? How do if somebody wants to find out more about what's happening? How do they get a hold of you?

37:12

Well, they could drop me an email, of course. Yeah, no. Well, I will ship my email, you could put it in the underbar wemco.funder@ns.nl.

37:26

Are you out there on LinkedIn? Of course. No, have the LinkedIn contact. Robert, what about you? How do they get a hold of you? Hey,

37:34

you can also check out conclusion Dalton LLC or other services and search for decides on the IoT and digital twin solutions. You're gonna have, of course, you can connect with me on LinkedIn and send me messages. Whenever you have any questions about this or other IoT solutions you want to invest with me? I

37:52

love it. And I'm telling you, I feel like I've just scratched the surface with this conversation. I want more but but you know, we're limited by time. You both were just absolutely wonderful. Thank you very much for being on Industrial Talk.

38:06

Well, thanks. Nice to meet you. Again, after Barcelona in the book. We need you again.

38:13

All right, listeners, we're gonna wrap it up on the other side, we're going to have all the contact information for these two jets out on Industrial Talk. So stay tuned, we will be right back.

38:22

You're listening to the Industrial Talk Podcast Network.

38:32

Yeah, meaty conversation a lot, a lot of bits and tidbits and all that good stuff and the digital twin in action, by the way, and action. I highly recommend that you reach out to them. We're gonna like, again, all out on Industrial Talk, just, you know, find them out on LinkedIn and and begin a conversation of how they can definitely help you in your digital transformation journey, your digital twin journey. They're doing it they they've been in the trenches. So it's a that's a good thing. All right. Again, follow up right now. Your voice needs to be amplified. Your opportunities need to be expanded. You need to increase that network. That's what Industrial Talk is all about. For you. It is a platform, an ecosystem dedicated to you celebrating industrial professionals here again. Be bold, be brave, dare greatly hang out with these two jets, change the world. We're going to have another great conversation shortly on Industrial Talk, so stay tuned.