How IoT Prevents Railway Accidents

Subscribe to IoT Leaders

Subscribe now

Subscribe now

Subscribe now

Transcript

Intro:
You’re listening to IoT Leaders, a podcast from Eseye that shares real IoT stories from the field about digital transformation swings and misses, lessons learned, and innovation strategies that work. In each episode, you’ll hear all conversations with top digitization leaders on how IoT is changing the world for the better. Let IoT leaders be your guide to IoT, digital transformation, and innovation. Let’s get into the show.

Nick Earle:
Hello. My name’s Nick Earle, CEO of Eseye. Welcome to this episode of the IoT Leaders podcast. In this one, we’re going to talk all things rail. It’s a very nice story about a very old industry. Let’s face it, rail is 200 years old and it’s still essentially the same system of an engine pulling carriages with steel wheels on steel rails. It’s the same system we’re using 200 years later. It’s not an industry that has really embraced technology very much other than in the actual rolling stock itself. But in this case, we’re talking about IoT and the ability to do environmental monitoring to predict accidents which can be terrible, as you’ll hear, when they happen. It is a great IoT use case with probably one of our very first customers from 14 or 15 years ago who asked us, could we help with a design product that didn’t exist at a low price that could actually monitor railways.

Nick Earle:
You’ll hear Declan, who’s my guest this week. You’ll hear him talk about what they did, the journey he went through, the pivot he had to do to solve the problem and where he is now. It’s a story we can all relate to about railways and it’s also a climate change story about how the nature of climate change is producing more localized weather, which increases the need to solve this problem in the first place. So lots here, including for those of you who really wanted to know, an explanation of why train tracks buckle in the UK but not in Australia where it’s even hotter. If you ever wanted to know that, continue listening. With that, I’m going to hand you over to my chat with Declan O’Brien of TrenTrace. Enjoy this. Declan, welcome. Welcome to the IoT Leaders podcast.

Declan O’Brien:
Thank you very much. It’s a pleasure to be here. Thanks for being asked.

Nick Earle:
Sure. Let’s dive in and we’re going to talk about all things railways, safety and many things. But before we do so, I always like our guest on the IoT Leaders podcast to just introduce themselves, people get to know you before they hear your story. So maybe what’s your quick story of how you got to where you are today?

Declan O’Brien:
Well, my name is Declan O’Brien. I’m a mechanical engineer. I have a master’s in engineering, received many years ago in Ireland. I started off, I suppose, I got into the medical device industry after about two or three years working where I got exposed to writing software and software for process control, for quality management and for stuff like that. I suppose I got more into engineering systems and stuff like that in maintenance systems. I suppose around 2010 was when I started my Internet of Things story when I don’t think the internet of things phrased was around at the time. It was more remote telemetry or whatever it was.

Declan O’Brien:
It’s been kind of a meandering route to get to where we are today. We now have a system called TrenTrace that does mainly environmental monitoring and it’s based around the railway industry. It can also transport infrastructure in general, but railways, I suppose, where it’s most at all.

Nick Earle:
Right. We’re going to obviously get into that and what you’ve done with TrenTrace where you are a Director. Whenever you talk about railways, you have to talk about safety and whenever you talk about safety, unfortunately, you have to use examples of where it can go wrong because that’s really what we’re trying to avoid here. Where I’d like to refer to is, just to give an example from the UK which was in 2020, it was the Stonehaven incident in Scotland, wasn’t it? Maybe just to recap what happened there because that’s a great example of what your company is trying to stop happening in the first place.

Declan O’Brien:
Yeah, that’s right. Stonehaven was an accident that happened during the COVID lockdown, which I suppose is probably a little bit lucky because-

Nick Earle:
Yeah. Not too many people on the train, right?

Declan O’Brien:
That’s right. There were only actually nine people on this train and it ended up with three fatalities unfortunately and six people were injured, three of whom were injured relatively seriously. I suppose it happened due to what we’re all experiencing, which is climate change really, whether you believe that the science are not behind climate change, whether events are becoming more frequent and nastier in their consequence, unfortunately. Just to give a background of what happened at Stonehaven was a commuter train was traveling on the track at about 70 miles an hour and it hit a landslip. The train was derailed. The front carriage left the train and rolled down the embankment and catching fire. The other two carriages came off the track. When they looked at what were the causes of the accident, the first one was that there was around 50 millimeters of rain fell in the previous seven radars.

Declan O’Brien:
Even for Scotland or for Ireland as well where we’re well used to rain, it’s a hell of a lot of rain to fall in a very, very short space of time. When they look back at the causes for the accident or what they might have seen or been able to foresee before the accident, they discovered that there were actually two previous landslips occurred, one of them in 1915, which caused a fatality and another one that happened around 2020 where they saw that there were problems with this again, or sorry, around 2015.

Nick Earle:
Yeah.

Declan O’Brien:
When they looked and they looked at the root causes for why the landslip occurred, because things like this, they can be very difficult to prevent all the time, but they look back to see what actually were the root causes of the accident, the first one was obviously the severe weather. So they got 50 millimeters of rain in a very short space of time. That caused flooding. The flood damage actually caused the landslip, but it also happened because of poor drainage around the area. When they look back at the previous contractor who had been tasked with making the area a little safer along there, didn’t put in the drains as they were supposed to be and they weren’t monitored properly either. So a combination of those two things caused the actual landslip to occur and hence the accident.

Declan O’Brien:
You can say it was an unfortunate, unforeseeable event, but when you look back at the information that was there, they knew that an awful lot of rain had fallen. They knew that there were issues with the area. Is something like that preventable? You can’t say for sure with 100% certainty, yes, we could have absolutely prevented this accident, but you can certainly increase your odds by putting it, monitoring it in a place like that because all of the evidence pointed to the fact that there’s something could happen on this stretch.

Nick Earle:
Exactly. We’ve talked about a lot of different use cases on this IoT Leaders podcast in over 20 episodes, I think, and one of the common themes is that by monitoring a lot of things, you can spot patterns. As you say, we know how to monitor rainfall, we know how to monitor drainage systems, we know how to monitor where the trains are and how fast they’re going, and putting these data together to predict the likelihood of something is one of the big advantages of IoT, is making sense out of lots and lots of data. We know this is a huge issue and as you rightly say, with climate change happening and certainly in the UK. The last, I would say just month, we’ve gone from the hottest day ever, over 40 degrees centigrade, over 100 degrees Fahrenheit to some of the worst rain certainly I’ve seen, tremendous storms and that appears to be the new normal.

Nick Earle:
This is a really important issue, especially as people’s lives and safety are concerned. Let’s try and unpack this story. Clearly, this is not a new thing. Railways have been around for 200 years and there have been a lot of tenders in this space, haven’t there? But they tend to be not really based on an IoT model as it seems. It tends to be small amounts of very big equipment as opposed to IoT, senses all over the place and lots of measurements. So maybe explain that model from your perspective.

Declan O’Brien:
Yeah. I suppose what we’ve seen is that the tenders come out for lots of the problem procurement processes. They seem to be based more around, I suppose you might call them slightly older models where weather stations have been monitored remotely for a long time as have some water levels as well. It seems to be that it’s almost the people who are doing the specifying are looking at the type of costs that are going to come in for these things and they specify very, very expensive equipment to do the things. For example, if they want to measure water levels for predicted flooding or something like that, they want to measure to maybe 0.1 millimeters accuracy with the pressure sensors, sorry, vented pressure sensors that take account at the changes in atmospheric pressure and things like that, but they want to do it to a very, very high degree of accuracy.

Declan O’Brien:
Whereas what we look at is more less accurate instrumentation. We might be able to measure just for example that one millimeter accuracy rather than 0.1 millimeter accuracy, but there’s an order of magnitude difference in the cost of the instrumentation. When you specify very expensive equipment, you can only afford to put it very, very spaced out. You can’t afford to put it, say once every five kilometers. You have to do it once every 50 kilometers. What we tried to look at is, wouldn’t it be better if we could measure two plus or minus one millimeter every five kilometer distance or every 10 kilometer, depending on the project, rather than spacing out your equipment an awful lot more and being able to measure very, very accurately?

Declan O’Brien:
Especially with monitoring things like rainfall amounts, things like water levels, is it really an advantage to know to within plus or minus 0.1 millimeter how much rain has fallen or is it plus or minus one millimeter sufficiently accurate? If you look back at things like the way climate change is happening, what’s happening is the weather event, the severe weather events are happening more frequently, but severe weather events tend to be, at times, very, very localized. So if you get convective storms, something like that, like in the case of Stonehaven, you could have 50 millimeters rainfall in the space of two hours in a five square kilometer radius and 10 kilometers down the road, it may not have rained at all.

Nick Earle:
Right. That’s a really interesting point, isn’t it?

Declan O’Brien:
The power of what we can do by putting in slightly less accurate instrumentation over a much wider area, I think is telling us an awful lot more data.

Nick Earle:
That is so true. I was talking about just the weather we’ve had the last couple of months or so, but I was thinking of some of the much needed rain, certainly from my garden’s perspective, much needed rain that I went into work the next day and said, “Oh, God, thank goodness we had all that rain.” People who lived less than 10 miles away from me said, “What rain?” I said, he must be kidding. It rained for four hours. It was thunderstorms. Not a drop. So that pattern is changing, which has implications for environmental monitoring and measurement as you point out. So let’s talk about TrenTrace and what you did. You’ve had a couple of attempts at getting the right solution, haven’t you? Let’s talk about it from a product point of view. As I understand it, you started off with something that would go…

Nick Earle:
No. Let’s take a step back even before that. You wanted to solve this problem. We’re going to get into how Eseye helped you solve this problem and how you found us, but you had it go initially at putting something under the rail. Is that right? Sort of putting something under the rail to monitor it.

Declan O’Brien:
That’s right, yeah. We started off like a lot of companies start off with a little sob story where myself and a partner, we decided or we were through a tender process. We started a job for Irish Rail to measure the rail temperature. We started off with a product that was sitting underneath the rail. Attached to the bottom of the rail was a Pt100 temperature probe that measured the temperature. There was a battery and a modem and all was self-contained on a little clip that sat underneath the rail. We had enormous problems with connectivity. I looked after the software end of it so that we could get the information into a database and display it to the user and my partner was responsible for all of the hardware.

Declan O’Brien:
Everything was blamed. Even the Vodafone network was blamed that we were connecting to for the issues with the data, issues such as not being able to connect and when it did connect, issues with the timestamp where you can get into the technicalities of how the hardware actually picks up its time. If you don’t have reliable time stamping on your data, it becomes very difficult to do anything with it. And around that time, I met with Ian Marsden, one of your directors.

Nick Earle:
One of our co-founders, yeah.

Declan O’Brien:
Yeah. Surely by fluke, I think I’m probably one of your oldest customers, I would think.

Nick Earle:
I think you are, I did a little shout out. I think you’ve been with us since very early days. I don’t know whether you all have been the longest of our customers, but you’ve certainly been with us for a while, yeah.

Declan O’Brien:
I think I’ll claim that title anyway.

Nick Earle:
Okay. Because we can’t prove it, so might as well claim it.

Declan O’Brien:
Yeah.

Nick Earle:
You met him? Where did you meet him?

Declan O’Brien:
I met Ian. I was working close to the airport in Dublin and he happened to be passing through and I had made contact with him previously about this project but he missed the boat. But we initially made contact then and he was in the back of my mind as somebody who could probably help us out if we ever needed help. As it turned out, we did actually need help. A little time after that, I said about working with Ian and Paul Marshall, one of the other co-founders, to design a product that we could use to actually measure the rail temperature and do it in a different way to the previous incarnation we had of the product. We got a much better product. It was much more reliable, but it still wasn’t reliable enough for what we wanted to do. So as a next iteration of our product, we decided to split it because we discovered that most of the issues with reliable connectivity were due to the fact that our modem was sitting underneath a big block of metal, which was…

Nick Earle:
Yeah. Doesn’t handle it. Not the best place to put a modem underneath a big chunk, about three inches of metal.

Declan O’Brien:
Paul, who I was working pretty closely with on the new iteration, suggested that we split and move the modem to the side of the rail and connect wirelessly to the sensor that will sit underneath the rail. So we use the ZigBee protocol to make a wireless connection, which is pretty much the same as Bluetooth-directed. A different protocol, same idea. We ended up with the unit, with the modem sitting at the side of the track powered by a rechargeable battery in solar panel that connected to our temperature sensor sitting underneath the rail. Once we started with that, and now the connectivity issues vanished. We could have our modem connected all the time. There were no issues with time. There were no issues with power because we had solar panel and suddenly, we had a good system for being able to measure the rail temperature.

Nick Earle:
If I can just hit pause there. Again, I want to draw the lesson from that. It’s interesting the way you phrased it. We have a lot of customers, prospects, people who talk to us who say, “Oh, the connectivity of your SIM card isn’t working because of the connectivity. If it’s connectivity, it must be the problem with the SIM card.” Actually, the SIM doesn’t do the connectivity, the modem does and it’s the interaction, the firmware details. A lot of people say, “Oh, I don’t want to get involved in the firmware,” but the firmware’s absolutely key and you mention just one example there, time stamping. If you’ve not got the right firmware settings and you’ve not got a holistic device design by somebody who knows what they’re doing, then actually, even if you do get the data and in the first iteration you couldn’t get the data, but then when you moved it away from three inches metal, got the data, then sometimes, you can get the data but the timestamp isn’t working for reasons that are pretty technical that most people don’t understand and so the data is useless.

Nick Earle:
One of the things that we do, and it’s for those of you watching on YouTube, there’s a reason I’ve used the picture that I have behind my head, which you could just listen to this, it’s the circuit board picture, this week’s picture, is that we always say as Eseye, it’s all about the device. It’s all about the device. Even on our website recently, and I don’t know whether you are aware of this, Declan, but we had a survey by Kaleido Research. I think it’s the largest IoT survey that’s been done. It was 750 enterprises globally, about 60% in the US, the rest, around the world. These are companies with over 1,000 employees. They asked them, if you’ve done a project in retrospect, what was the single hardest point of the project? And 84% of people who said they’d already tried IoT, because as you said, it’s been around for at least 12 years, 84% said, in retrospect, I totally underestimated how difficult the product is. I just thought it was a case of putting a SIM in because we’ve all been trained that’s how our phones work.

Nick Earle:
And so we’re always hammering home this message. You have to start with the product and you have to work with people who understand the product and the firmware settings and the way networks work and the clock and the timestamp because that is the number one inhibitor for IoT. In fact, we’re going back to hardware being important from thinking that hardware wasn’t important because the cloud came and we’ve seen the back of hardware. It’s all about software and it’s all in the cloud. Our view is that no, when it comes to IoT, both devices and edge aggregation devices with 80% of applications going to be processed at the edge, actually, we’re now swinging the pendulum back and the hardware is hugely important. Yours is a great case study in that sense because that was the help that you needed. You’re still, I guess, using Vodafone, but it wasn’t Vodafone’s issue. It was the device, right?

Declan O’Brien:
That’s exactly right, yeah. And just to come back, some people think that oh, well, we’ll use the cloud for our stuff and it’s a panacea or an answer to all our problems. As I said before, the cloud is just a computer in somebody else’s office.

Nick Earle:
I got to make sure all this has caught that because when we first did a sync up prior to this podcast, you told me that. I wrote it down. I thought it was the most succinct explanation of cloud computing I’d ever heard, so I got to repeat. Cloud is just a big computer in somebody else’s office. All the books in cloud computing, you summed up with one phrase, but essentially, of course it is. You uncoupled the two pieces. You still need to monitor the rail but you put little poles as I understand at the side and that helped with the comms. Then you’ve got this other model that you mentioned which is smaller, cheaper devices that can be put closer together, especially because of the local nature of climate change, and slippage doesn’t happen over a 10-kilometer area. It happens in a small area.

Declan O’Brien:
That’s right.

Nick Earle:
Just in terms of how it works, if we can, three things. You mentioned temperature. We all have images of rails buckling, which were very dangerous, water, so water flooding, bad drains, and slippage. Just to explain to listeners how we measure those things with your solution.

Declan O’Brien:
Yeah. The TrenTrace solution, essentially, what we’ve tried to do with our product, and I think we’ve got it nailed now, is that we can sell people the product. It’s a box that they can literally put on the side or they can put it on a pole on the side of a track. They can turn it on, connect the battery, and we have all of the infrastructure done. You guys have done the hardware design for me. I have done all of the software design. All of the intricacies are taken out for the customer, so it’s literally pretty much plug and play if you want to look at it that way. We can measure temperature, obviously. The rail temperature is done via a wireless ZigBee connection.

Declan O’Brien:
The water level is done, we use a pressure sensor that’s vented to the atmosphere and it’s literally mounted in a mounting tube. You see in the image behind me, even just in the corner here, if I point this, I can’t see it now, but in the background there, you can see one of the solar panels and there’s a water level sensor mounted just over the edge of the ditch there. Basically, what it does is it measures the pressure of the water and that’s how you calculate your water level. It’s very useful. We record. When we display it on screen, you can see how the water level rises and falls and you can very, very easily build up a picture of what it should do based on the amount of rainfall that you have.

Nick Earle:
Right. So you can look for unusual patterns of data.

Declan O’Brien:
Exactly.

Nick Earle:
So if it rains this amount, the water pressure should be that amount. But if the water pressure’s greater than that, then it’s indicative of-

Declan O’Brien:
There’s something else happening, yes.

Nick Earle:
Yeah, like the drains.

Declan O’Brien:
Yeah. We know, for example, with this one that we usually see the water level tends to rise pretty much linearly if we get a heavy rainfall event the previous day. The picture that’s in the background there, it’s actually on a floodplain in the northern half of Ireland, I suppose, and it’s close to the Shannon, the big river that runs through the middle of Ireland. What can happen is if you get a lot of rain, there’s a floodplain around here that’s naturally occurring and the water drains into there, but we know that if it rains on a Tuesday, on a Wednesday, we expect the water level to rise. Depending on how much rainfall happens, we can also see the shape of the drop-off in that. We know that if it doesn’t follow the normal pattern, that there’s something else going on.

Nick Earle:
You measure that little and often as you said, not just every now and again. You build up a whole pattern and dashboard, which is your early warning system for the-

Declan O’Brien:
That’s right.

Nick Earle:
… rail networks to long. They go over long distances. The one that fascinates me is slippage. I understand water. I understand temperature. How do you measure slippage or actually a better word, predict slippage? Perhaps it’s starting or likely. It’s almost like sensing an avalanche, it seems to me. So it seems intuitively really difficult thing to do.

Declan O’Brien:
Yeah. What we’ve done with the slippage stuff is that we use a load cell to measure the tension of a cable that’s pinned to the side of the surface that we think may slip or whatever. Again, if we go back to the hardware design, to save money so that we’re not sending data every 15 seconds on every parameter that we measure, we send in general every 10 minutes. But for our slippage detection, we send every 15 sec or we can send it once every 15 seconds if we find an issue. What happens is we measure the value on the load cell and every 15 seconds, we have a hook to say that, well, if it goes 1% higher than it was in the previous reading, we know there’s an issue here. So we send the data. We can get data back to the central station within 15 seconds of any issue happening.

Declan O’Brien:
In general, what happens is if you get a lot of rainfall, it can happen or it can also happen at times of very cold weather when the soil actually freezes. When it freezes, the water expands and any water that’s in cracks and things like that will push them apart. Very often, you can get instability that’s caused by that, but we can pick it up because of the change in tension in the cable. It’s sensitive enough actually to pick up when the sun comes out and when the sun goes in because it actually warms up the cables slightly and we can see the difference in temperature with that. We can compensate for that because we put a temperature sensor on the cable and again, it’s because it’s the same technology as we’re using to measure the rail temperature, so we’re almost getting it for free.

Declan O’Brien:
The beauty of doing what we’re doing is that it’s probably an order of magnitude are even more cheaper than other methods they’re doing. It can be measured with a fiber optic cable where you can build up fantastic pictures of what’s happening, but the problem is that you can only do it over a very small area. If you spend maybe a million euros or whatever it’s going to be measuring the differences on an embankment, well, would you not be better off spending a million plus another million maybe just to fix the issue and you don’t have to monitor anymore? Whereas with our one, we could probably measure it for €2,000 and you don’t need to do any remedial work until you think there’s actually an issue.

Nick Earle:
It’s a nice story, isn’t it, because some of the IoT use cases are based on massively sophisticated technology and artificial intelligence and ML and satellites, but actually, when it comes to the environmental monitoring, in your case, if you put a cable along the bank, peg it in, then you can measure that. As you said, the temperature of the cable, you can compensate for it, but if also one part of the bank moves, the tension on the cable changes and therefore you say it is a high-tech solution, but it’s not because it’s a cable, but more importantly, it’s this point you keep on coming back to, it’s a low-cost solution that can be implemented many, many times.

Declan O’Brien:
That’s right, yeah.

Nick Earle:
And especially with public finances, the way they are right now and inflation and debt and a whole series of things, a lot of the budgets, they’ve always been tight, but they’re particularly tight and people saying, “Oh, well, I know I need to do this, but perhaps there hasn’t been an accident for a while,” and there’s always somebody else wanting access to the money. Whereas if you can say, “Look, if you’ve got an order of magnitude or even two orders of magnitude difference in the cost,” and you say, “Well, just for a small amount of cost, you can put these things in and yeah, you’re not going to get the data to five decimal places, but you’re going to get the data. It’s the perfect enough model than-

Declan O’Brien:
That’s exactly it.

Nick Earle:
It’s a lot better than doing nothing and then the human cost and the actual cost of an event. So I think it is a classic IoT because what you’re actually doing is relatively simple. In the case of the slippage, the cable measuring temperature. These things have been around for years, but your business model is optimized for what’s happening both from a budget point of view and from a climate change point of view. As you said, the trick on this was getting the device and the connectivity, which is where we come in and that’s why we’re talking. Let’s look at the rail industry. You have this solution, rolling it out, able to do what we’ve described. Now from a rail perspective, there’s lots of things that they need to look at.

Nick Earle:
As an outsider to the rail industry, if I look in, they’re spending a lot of money on new rolling stock, electronic tickets in the UK finally, the big innovation of electronic tickets, the Wi-Fi on the train, the weather monitoring and things like that. But you’re actually coming at this, it seems from that’s the train and monitoring some big things like weather, but you’re actually linking environmental monitoring into this. I guess what you’re saying is you need to bring all this information together over wide areas in order to make sense of the big picture. There’s a very big dashboard opportunity here, especially considering prices of components are coming down, prices of connectivity is coming down, prices of hardware is coming down. I assume that from your perspective, you’re saying, look, given what’s happening and all of these trends, prices coming down miniaturization, it is an opportunity to add environmental monitoring to the mix on a mass scale. Would I be right in saying that?

Declan O’Brien:
That’s absolutely right. The more monitoring you can get in, the more data you build up. Obviously, data history is very important for things, for example, like rail track. If you monitor the max temperatures and the minimum temperatures and you see what your averages are, when they lay rail, they use a thing called a rail neutral temperature. You see a lot of talk about, in the UK, “Well, my god, we’ve got a little bit of warm weather here. Why do the trains have to slow down? Can’t they run these in Australia and India and places like that, that have even hotter temperatures?

Declan O’Brien:
I suppose a little bit of knowledge of the dangerous things they might say. You know that well, it expands and they can have hot temperatures somewhere else. But the real issue is that it’s the difference between the maximum temperature and the minimum temperature that really makes the difference when you lay rails. Because when you hit a maximum temperature and you think the maximum is going to be, say 50 degrees, the track temperature is going to be 50 degrees, you have to look at what the minimum temperature is going to be and you might decide that the minimum temperature is going to be minus five degrees or minus 10 degrees. So essentially, what you’ve got to do is you’ve got to calculate. Well, where’s the middle point? So that’s the temperature that I’m going to use to lay my track. Obviously, if you pick that the rail neutral temperature is 27 degrees, it’s not going to be 27 degrees at night when you’re laying a new piece of track. So what they do is they either compress the track or they stretch the track as they put it down.

Declan O’Brien:
What they do is they stress it with the amount that says, right, well, if it was 27 degrees, it’s now 12 degrees. This is how much stress should be in the track. So when you come to places like Australia or India, they don’t have very low temperatures in most of the places, so they don’t have the same range in their temperatures so they don’t have these issues.

Nick Earle:
That’s an added bonus. I didn’t think, as part of this podcast, we’d answer the question which I’ve asked, which is, okay, it got hot, but rail lines in Australia don’t buckle so are we using different? I just thought we were using different metal or whatever and that the explanation of the highs and lows in the range, that was interesting. I never realized that that’s what they did. I guess it links back to your system, doesn’t it? Because-

Declan O’Brien:
It does, exactly, yeah.

Nick Earle:
Six weeks ago or seven weeks ago, whatever it was here in the UK, we said, oh we had… Two years before that, we had the previous highest record temperature of 38.4 or whatever. That was a one in a thousand year event or whatever, but it turned out to be a one in a two-year event. Two years later, it went to above 40, so even 40 degrees centigrade. So even if you set the range, with climate change, certainly at the top end, those ranges are rapidly out of date.

Declan O’Brien:
That’s right. Exactly.

Nick Earle:
Rapidly out of date. When you set a new high bar within two years, it is unthinkable for it to be 40 degrees centigrade in the UK.

Declan O’Brien:
That’s right, yeah. Because you’ve got to remember, with all these high temperatures, if we have a really cold winter time where we get maybe minus 10, if you lay your track and you’re laying it with an eye on your highest temperatures, when you hit minus 10, we’ll have problems with the rails that crack because when they contract, they get into too much tension and they’ll break. That also happens.

Nick Earle:
Well, we’re learning a lot. It’s absolutely fascinating and it’s a great story. We could talk about this for ages. As you know I was on the periphery of the rail industry prior to this. Spent time at Hyperloop, actually trying to kill the rail industry, selling Hyperloop around the world, but I did spend an awful lot of time studying the rail industry as a result because that’s the people, the regulators and all the different people. There’s a lot of systems in place. It’s fair to say that technology hasn’t advanced at the same speed in rail as it has in some other industries, but what you are showing is that it is possible now with IoT to actually get data at low cost with a high density to actually predict things that were impossible to predict previously like back to where we started, the Stonehaven accident.

Nick Earle:
Unfortunately, clearly, the bank slipped onto the tracks and it probably had been there for quite a while. It was a remote area and the trains slammed into it and unfortunately, the accident happened. But the ability to now monitor a whole infrastructure like this, it’s becoming realistic thanks to companies like TrenTrace, your solution, just because of what IoT can do and what the data can do. You then combine that with the satellite monitoring and the rainfall, all the other things. We actually have a chance to really predict this at relatively low cost without having to buy large amounts of high equipment and put it in every now and again. It’s a very big area and a big opportunity in a very big industry around the world, which is a problem that needs solving. You’re doing something which really does contribute in terms of IoT making a difference. It must be quite heartening for you when you actually see these systems in place and the fact that your data is now being used. We really need to use that data more, don’t we?

Declan O’Brien:
Yeah, for sure. There’s another added benefit, especially for the rail industry from a safety perspective, which is the safety of the rail personnel. I know Network Rail and they operate a policy of keep boots off ballast, which basically the ballast is the stones in which-

Nick Earle:
The stones, yeah.

Declan O’Brien:
… the rails are laid. What they want to do is they want to keep the personnel boots off that because…

Nick Earle:
Because trains hit them. Yeah, it’s dangerous.

Declan O’Brien:
They’re in the red zone. There are trains and traffic happening and accidents happen. You don’t need to have a guy go around and monitor the track temperature by sticking on a magnetic temperature gauge or you don’t need to have a guy go around and inspect a culvert to see if it’s blocked or whatever, with debris or leaves or branches and stuff like that just to see, was the water flowing and draining away correctly? You can do it automatically with our technology, which keeps people safe. It keeps people on the train safe and it keeps the personnel who are looking after it safe.

Nick Earle:
Yeah. Absolutely. It’s definitely an area that IoT’s going to make a major contribution and we’ll all, as individuals, feel effects of it because the trains will run better and more efficiently,-

Declan O’Brien:
Sure.

Nick Earle:
… less delays. Declan, we could talk for a long time and I’d be in danger of sharing war stories from my Hyperloop days, which by the way, I failed. I kept saying to the UK government, you really shouldn’t do HS2, the high speed rail because there’s a different way of doing it, but it was too late. It has its own momentum, but one day, it’ll happen. So that was really, really interesting. Our time is up, I’m afraid. Thanks for being one of our, I won’t say oldest, that has a different connotation, one of our longest customers right in the sweet spot of what we do, solving the device issue and the connectivity issue and in an area that, as we said, is massively important and everyone listening to this or watching this will actually be able to relate to the problem that you are solving. It’s a really nice case study and great for the IoT Leaders podcast. So thanks very much for joining us on this episode.

Declan O’Brien:
Thanks very much for having me.

Outro:
Thanks for tuning in to IoT Leaders, a podcast brought to you by Eseye. Our team delivers innovative global IoT cellular connectivity solutions that just work, helping our customers deploy differentiated experiences and disrupt their markets. Learn more at eseye.com.

Outro:
You’ve been listening IoT leaders featuring digitization leadership on the front lines of IoT. Our vision for this podcast is to be your guide to IoT and digital disruption, helping you to plot the right route to success. We hope today’s lessons, stories, strategies, and insights have changed your vision of IoT. Let us know how we’re doing by subscribing, rating, reviewing, and recommending us. Thanks for listening. Until next time.