Latest / The Intermodal Logistics Podcast / Tracking the Trains

Transcript

Tracking the Trains · The Intermodal Logistics Podcast

• 0:04
  SPEAKER_00: Thanks very much for joining us on this edition of the Intermodal Logistics Podcast.
• 0:09
  SPEAKER_00: I'm Kevin Baxter with my co-host today, IntegLogistics CEO, Rick Lagore.
• 0:14
  SPEAKER_00: Rick, good to be back.
• 0:17
  SPEAKER_00: Only one off.
• 0:18
  SPEAKER_00: Only one so today we are uh talking about an area of uh the supply chain that that maybe people don't feel like they have that much visibility on, and in the way that uh there is uh visibility and intel coming in from uh the railroads and uh the railroad tracks, and there's miles and miles of them everywhere.
• 0:41
  SPEAKER_00: And uh unless you're in in a town, you don't necessarily see what's going on uh on those, but uh but rail estate is uh the company that we are uh discussing uh today with uh DP of Sales and Marketing at Dan DeVoe.
• 0:54
  SPEAKER_00: Dan, thanks very much for uh joining us.
• 0:56
  SPEAKER_01: Thank you.
• 0:57
  SPEAKER_01: Yeah, excited to be here.
• 0:59
  SPEAKER_00: So uh first off, we'd like to uh get a little uh info about our our guests.
• 1:04
  SPEAKER_00: Uh so so tell us uh how you uh came to work with Rail State and uh and then we'll dive uh deeper into uh to what the the service actually is.
• 1:14
  SPEAKER_01: Yeah.
• 1:15
  SPEAKER_01: Uh so it's been a uh securitous path uh to get here and won't go into too much detail.
• 1:22
  SPEAKER_01: Uh but uh had a background in um marketing, um so had been working at um some SaaS companies as well as on the consumer side, um, including in a company that uh was logistics-based.
• 1:37
  SPEAKER_01: Uh I was an early employee at Drizzly, which was an alcohol delivery app that was later um sold to Uber.
• 1:45
  SPEAKER_01: Um and about five years ago was introduced um to the uh founders of uh Real Estate.
• 1:54
  SPEAKER_01: Uh well, I was looking for a new challenge, um, and they needed someone uh with a background in marketing.
• 2:02
  SPEAKER_01: Um so that's how we connected.
• 2:04
  SPEAKER_01: Uh, have been at the company now for five years.
• 2:08
  SPEAKER_01: Um but you know, before I was even a marketer, uh did some other work.
• 2:14
  SPEAKER_01: I was an attorney for a while, uh, worked in policy, um, and also worked in the financial services industry.
• 2:21
  SPEAKER_01: Um, but throughout all of those roles, um, have been very focused on data and analytics and connecting the dots between different areas to uh basically get creative solutions for uh customer problems.
• 2:38
  SPEAKER_01: Um so when I was introduced to RailState, um it was a very new problem um with um some very interesting customer challenges.
• 2:47
  SPEAKER_01: So joined yeah, five years ago um and have been building um with the team since then.
• 2:54
  SPEAKER_01: Um and now we are about 20 people, um so still quite small.
• 2:59
  SPEAKER_01: Um and although I am uh leading sales and marketing um at a small startup like this, uh that basically means I do uh pretty much um everything across Word, um, which I'm sure yeah you're familiar with.
• 3:16
  SPEAKER_01: Um so yeah, work a lot uh with customers uh to help sure um ensure that they're getting the the best data for um their particular needs.
• 3:28
  SPEAKER_03: So you say startup.
• 3:30
  SPEAKER_03: How long have you been uh has Real State been in existence?
• 3:34
  SPEAKER_01: So um we've been active um for about five years.
• 3:39
  SPEAKER_01: Um it started initially as an idea um from our founders in about 2019 um and took some time to you know really get um off the ground.
• 3:52
  SPEAKER_01: Um but our founders, Jamie Heller and John Schmitter, um, they are have long careers in uh transportation and logistics, um, so 30 plus years.
• 4:04
  SPEAKER_01: Uh Jamie mainly focused on consulting and um in the energy sector.
• 4:10
  SPEAKER_01: Um and John has a background in railroads, worked at a number of um class ones um and some smaller railroads, and also has been yeah, consulting um with shippers for a long time.
• 4:24
  SPEAKER_01: And the two of them had been consulting with major shippers, many of them um in Canada on the energy side, doing lots of work um like mediation and arbitration.
• 4:37
  SPEAKER_01: Um and found that you know when they're in those engagements um representing a shipper, it's very hard to um make a good case um in those situations because data just hasn't been available um to really support um what the shippers were um arguing.
• 5:04
  SPEAKER_01: And so what Jamie and John uh would often do is try to collect um information that was otherwise not available and do that directly.
• 5:14
  SPEAKER_01: So for years uh that might involve uh actually going out and collecting things in person.
• 5:21
  SPEAKER_01: Um, so sitting on the side of a hill and watching trains go by and counting uh the cars to see what's really moving on a particular corridor.
• 5:31
  SPEAKER_01: Um and you know, that is something that you can do for a few hours, a few days, um, and you get some good information, uh, but it's uh not uh sustainable um you know for an entire industry.
• 5:47
  SPEAKER_01: Um so having had that experience of seeking out um information on what is really moving on uh particular lines, um came up with the idea of rail state to basically operationalize uh that work of someone being out there and like counting uh trains and cars.
• 6:08
  SPEAKER_01: Um and so that's what the idea um came about around yeah, 2019.
• 6:16
  SPEAKER_01: Um took some time to build out how actually go about collecting um that data in a a way that could scale um and be automatic.
• 6:28
  SPEAKER_01: Um and then around yeah, 2021 um came out with our first uh collection uh device.
• 6:37
  SPEAKER_01: Um so a camera-based uh sensor uh that we install outside um adjacent to uh the tracks uh to collect images of uh trains that are passing by.
• 6:51
  SPEAKER_03: Which is much different than normal track and trace.
• 6:54
  SPEAKER_03: Normal track and trace is it's you don't have an image of the car, you don't have an image of the trailer, the container.
• 7:00
  SPEAKER_03: So you uh you guys are actually tracking and taking pictures of the cars behind the train.
• 7:08
  SPEAKER_01: Yeah.
• 7:09
  SPEAKER_01: Um, so um, and part of the reason for that is uh to do this work independently, we need it to be outside of the rail right-of-way.
• 7:20
  SPEAKER_01: Um if we were closer uh trying to collect that information um, you know, and within the the right of way, uh we would need permissions.
• 7:29
  SPEAKER_01: Um there would be restrictions likely on the on the data.
• 7:32
  SPEAKER_01: Um and unlikely that that would happen.
• 7:36
  SPEAKER_01: Um so in an attempt to keep everything independent, um we yeah, built these uh camera-based sensors that we can install um outside the rail right-of-way and take images of um the trains as they pass.
• 7:52
  SPEAKER_03: So are you getting permissions from the landowners next to the trains, or how does that go about?
• 7:58
  SPEAKER_01: Yeah, so we um have a team that goes up and finds uh locations, but basically, yeah, we look for uh properties that are adjacent to the rail lines.
• 8:09
  SPEAKER_01: Um, so could be residential, could be small businesses.
• 8:12
  SPEAKER_01: Oftentimes it's you know, someone's garage or you know, a shed um in the back that abuts um the rail line.
• 8:22
  SPEAKER_01: We make an agreement with those uh property owners and then install one of our cameras um on their property.
• 8:30
  SPEAKER_01: Um so we we do access uh power and internet um at those sites.
• 8:37
  SPEAKER_01: Um although we do have some capabilities for cellular and even solar for more remote locations.
• 8:44
  SPEAKER_01: Uh but generally, yeah, we're tapped into local um internet and power um so that we can run the devices.
• 8:51
  SPEAKER_01: Um and maybe yeah, could explain yeah, a bit on the the technology.
• 8:58
  SPEAKER_01: Um our boxes um are um about the size of like a microwave.
• 9:06
  SPEAKER_01: Um and they include a camera, um, also have a infrared um illuminator that we developed.
• 9:16
  SPEAKER_01: Um, and that's because we need to take images at night.
• 9:20
  SPEAKER_01: Uh so uh to illuminate something from say 200 feet away um at night and not using visible light, um you need another way.
• 9:31
  SPEAKER_01: So we're using yeah, those infrared illuminators so we can take pictures um at all times.
• 9:38
  SPEAKER_01: Um we have some processors on the device as well, um, so we can um go through the images and our machine learning algorithms um basically look at all the images of each individual train and extract um data from those images.
• 9:56
  SPEAKER_01: So that's uh locomotive information, uh the type of rail car it is, the uh rail car ID, uh the hazardous material placard, um, and then in the place uh of uh containers, we're identifying the orientation and size of each individual container on uh on a well car as well as the uh container IDs.
• 10:22
  SPEAKER_01: Um with all of those images of uh a train um basically put together the entire composition.
• 10:32
  SPEAKER_01: Um so from the lead locomotive to the last car, um exactly what each train um is is made up of um all the cars and containers.
• 10:46
  SPEAKER_00: And so so what are you pulling?
• 10:48
  SPEAKER_00: I mean, you're you're pulling obviously what what the train is made up of, what are what are you measuring and what are you what are you providing in terms of uh that that finished product afterwards then?
• 10:59
  SPEAKER_01: Yeah, so it's um we are you know for a particular corridor, say, um we're getting all of the train volumes, um all of the train types, when they are passing, uh the speed, the train lengths, um, and then all of that car and container information.
• 11:18
  SPEAKER_01: Um so with that, you know, we know how many uh tank cars of LPG have gone by a particular location, or um yeah, how many domestic containers of you know JV Hunt containers have gone past a particular location.
• 11:36
  SPEAKER_01: Um so we're collecting all of that, and then depending on the customer need, they're looking at different things.
• 11:44
  SPEAKER_01: Um so we have customers that are looking at train volumes through a particular area, uh, wanting to know um how frequently they're going through, what type of trains they are, uh the train length, the spacing in between trains.
• 12:01
  SPEAKER_01: Um then we have other people that other customers that are looking for commodity information.
• 12:08
  SPEAKER_01: Um, so we want to know, again, maybe yeah, how how much LPG um is going through a particular um area.
• 12:16
  SPEAKER_01: And then we have uh other customers that are looking for um information on their own shipments.
• 12:24
  SPEAKER_01: Um, so uh tracking uh their own cars.
• 12:28
  SPEAKER_01: Um although we're not generally, you know, we're not built to do the exact same things as a track and trace um system, but uh do have yeah, customers that are um looking at their specific cars and um how they're you know faring um against other traffic um on the network.
• 12:49
  SPEAKER_00: Oh, you mentioned um sort of the genesis was was your your founders working with shippers.
• 12:55
  SPEAKER_00: So so shippers are obviously one customer type.
• 12:58
  SPEAKER_00: What are some of the other customer types that you mentioned?
• 13:00
  SPEAKER_00: That's some of the different information that they're looking for, but but what other kinds of uh companies are are are your customers?
• 13:08
  unknown: Yeah.
• 13:08
  SPEAKER_01: We we work with a lot of government um and public sector uh customers.
• 13:13
  SPEAKER_01: So um work with federal agencies, um, state and provincial agencies.
• 13:20
  SPEAKER_01: We do have a large presence in uh Canada.
• 13:23
  SPEAKER_01: Um and you know what they are looking for um also varies depending on uh their mandate.
• 13:31
  SPEAKER_01: Uh but in the example uh for the example of one of our major customers in Canada is Transport Canada, essentially the Department of Transportation um in Canada.
• 13:42
  SPEAKER_01: And they want uh to know what's going on in the rail network overall, um in the same way that they're looking at other modes of transportation and wanting to know what's moving, um, if there's um congestion issues, if there are disruptions, uh what uh type of recovery is happening, um, if there are needs for new or expanded infrastructure, um having that data behind uh to support those decisions is what they're looking for.
• 14:18
  SPEAKER_01: We have some other government customers that are looking at hazardous material movements uh for emergency planning, uh, others that uh want to know just about locomotives uh because they are uh have a mandate for uh emissions um monitoring.
• 14:37
  SPEAKER_01: And so government is a is a big area uh for us.
• 14:41
  SPEAKER_01: Um also have work with a number of states um in the US who have to do rail planning.
• 14:49
  SPEAKER_01: Um so every state every four years has to submit a rail plan to the FRA outlining you know the impact um and needs um of the rail network within a state and identifying uh infrastructure projects uh to prioritize over the next you know five, 10 years.
• 15:09
  SPEAKER_01: And we're helping them with um the information on what is really moving uh currently through uh the state um and what uh levels of performance are actually being seen um on quarter by quarter uh across the state.
• 15:30
  SPEAKER_01: For the past uh 20 plus years, uh how that planning work has generally been done is using the Waybill um the Carlo Waybill sample, um, which is a great source of data um for carload movements and origins and destinations, uh not the greatest uh source of data for uh actual train movements.
• 15:57
  SPEAKER_01: And also all of the data from the Carload Waybill sample is at least two years old.
• 16:03
  SPEAKER_01: Um so um we're helping those states have current data on what's really uh moving um throughout throughout those states.
• 16:14
  SPEAKER_01: Um so government, yeah, big area for us.
• 16:18
  SPEAKER_01: Another um area outside of shippers um for uh rail state has been um companies looking for various types of market intelligence.
• 16:28
  SPEAKER_01: Um commodity traders, um, hedge funds looking at sort of macro issues of what's moving, um you know, different uh export volumes, uh cross-border volumes, but also some companies that are um either analyzing or looking to invest in particular stocks.
• 16:52
  SPEAKER_01: Um so we even have some people that are looking at uh companies like Boeing, um which their fuselages are shipped by rail um from Kansas to um Washington, and we're tracking um those those movements um when we see them move from um across the network from Kansas to Washington.
• 17:17
  SPEAKER_01: So there are yeah, a number of uh use cases uh that we're seeing uh from customers.
• 17:26
  SPEAKER_01: Um but yeah, generally it's the government um market intelligence uh focused applications and um shippers.
• 17:39
  SPEAKER_03: So do you have full coverage across the the uh the rail network?
• 17:43
  SPEAKER_03: Are there spotty areas?
• 17:44
  SPEAKER_03: It's uh you've mentioned Canada several times.
• 17:46
  SPEAKER_03: It kind of sounds like Canada is pretty pretty much covered.
• 17:48
  SPEAKER_03: Is there how does it look in the US and Mexico?
• 17:53
  SPEAKER_01: Yeah, so we have um all of Canada, the Class One network, uh covered um and have had that covered for a couple of years now, um, as well as the Canada-US border.
• 18:05
  SPEAKER_01: Um and that's from partly from our history of having worked with um some of the agencies up there who helped um do that rollout with us in Canada.
• 18:15
  SPEAKER_01: Uh the US, we are right now about um 40% of um coverage across the Class One network, uh, all of the West Coast, uh Southern California to Chicago, uh some coverage in the east and uh the Gulf, and we're building out currently um the Mississippi Valley um area uh over the next couple of months.
• 18:42
  SPEAKER_01: We do have Mexico and the US border covered, uh don't yet have any others uh locations within Mexico.
• 18:53
  SPEAKER_01: Um and we're on a path now to have the entire Class One network covered in at least the US and Canada by the end of next year.
• 19:05
  SPEAKER_00: Okay.
• 19:06
  SPEAKER_03: It's aggressive.
• 19:07
  SPEAKER_00: Yeah, that's uh a lot more coverage than uh a few people sitting on a hill, then I guess, right?
• 19:12
  SPEAKER_00: Yeah.
• 19:13
  SPEAKER_01: Uh yeah, we have um and the number may not sound uh huge, uh, but it's about 350 sensors um now.
• 19:24
  SPEAKER_01: Uh but that does cover uh so much of the network uh because there are many places where you can go hundreds of miles um and nothing really happens.
• 19:36
  SPEAKER_01: Like there's no facilities, um, there's nowhere for, you know, there's no yards, there's nowhere for trains to go.
• 19:42
  SPEAKER_01: Um so may only need one sensor in a location like that to capture all of the flows on that corridor, whereas other places we have a lot more density.
• 19:54
  SPEAKER_01: Um around port areas, so Port of Vancouver or LA and Long Beach have significantly more um dense coverage.
• 20:05
  SPEAKER_01: Um so we're capturing all of those movements.
• 20:08
  SPEAKER_00: Okay.
• 20:09
  SPEAKER_00: So how do you determine how how many you need in a a given area?
• 20:13
  SPEAKER_00: I mean, what what's is there a formula uh that that goes into this or uh wish there was a formula.
• 20:21
  SPEAKER_01: Um through experience, uh we've you know developed um some internal guidelines, but each location um does have quirks um and some particular needs.
• 20:35
  SPEAKER_01: Uh but generally we try to you know cover every um subdivision in a in an area.
• 20:43
  SPEAKER_01: And uh depending on uh junctions and locations of yards, uh we'll try to get you know on either side of them.
• 20:52
  SPEAKER_01: Um so in some areas that can be quite easy, others it's yeah, um a lot more complicated.
• 21:01
  SPEAKER_01: But through experience, we're we've gotten pretty good at um at it.
• 21:07
  SPEAKER_00: Well, so so let's get into uh to more of what you learned.
• 21:12
  SPEAKER_00: I I think actually one of the things that that put uh real estate on on my radar in particular was uh a couple of LinkedIn posts recently.
• 21:22
  SPEAKER_00: Um there were there were a couple of incidents where where something happened uh to uh to a train and uh and then there's a question of okay, so so how quickly does the the train get back up and running after a disruption.
• 21:38
  SPEAKER_00: So so take us through actually uh a couple of uh of those events.
• 21:43
  SPEAKER_00: I mean, uh I think uh the the two that were some recent LinkedIn posts, there was a NSF uh issue in uh the NECL subdivision and then um uh UP uh derailment in uh New Mexico.
• 21:58
  SPEAKER_00: So what are you?
• 22:00
  SPEAKER_00: How are you tracking those and seeing like when the disruption occurs and then when it gets up and running?
• 22:06
  SPEAKER_01: Yeah.
• 22:07
  SPEAKER_01: So when in areas where there's a good amount of volume, and Southern California being an area where there's a lot of volume, when we see drop-offs in um volume from the norm, we're internally alerted about that.
• 22:29
  SPEAKER_01: So we'll have that information of that there has been something.
• 22:52
  SPEAKER_01: We'll be alerted of that and then investigate it to see what the impact has been.
• 23:01
  SPEAKER_01: So in these cases, we um were able to identify that when we saw the slowdown through this region, um, could look to see what trains have gone through and which trains had not passed through multiple of our sensors.
• 23:20
  SPEAKER_01: Um, so then could identify which ones you know have um been stopped and aren't continuing through, and um which ones are the last ones that have made it through a particular um location.
• 23:36
  SPEAKER_01: Then oftentimes what happens is we'll get uh some local news about uh there being a derailment um or some other issue, and then can put those together to understand, you know, okay, there was a derailment here, but we now know you know what the impact is, what trains were stopped, um, and then as soon as we see trains moving again um through our sensors, we know you know when that recovery has started and can look uh for how long it's taking and what the shape of that recovery looks like.
• 24:13
  SPEAKER_01: So in the case of UP, um they were down for yeah, a bit less than a day.
• 24:20
  SPEAKER_01: Um, but then the on the the day the day after um with the recovery had significantly more volume um than they typically do on a day.
• 24:31
  SPEAKER_01: Um so it was really moving those trains that had been um backed up um through quickly and then got back to normal um very soon after that.
• 24:43
  SPEAKER_01: Um and generally, you know, that has that is what we have found um with uh major disruptions, um derailments, but also some other issues which um we have some history on in um Canada of impact from wildfires or extreme cold um or even labor disputes that have um led to a shutdown of the rail lines um or you know particular uh port locations.
• 25:16
  SPEAKER_01: But the shape of a recovery generally seems to uh be the same, that uh there's a level of traffic that we see before, uh we see the drop-off, and then we'll get uh back to that uh status quo level very quickly.
• 25:35
  SPEAKER_01: Um, so there may be an increase um in above normal um for a couple of days, uh, but then generally get back um to that status quo level quite quickly, um, which I think is different um than what a lot of people assume uh may be happening that uh the railroads may be bringing on you know additional trains uh to move volume that has been backing up during a an extended um extended disruption.
• 26:08
  SPEAKER_01: But from what we've measured, we haven't really seen that.
• 26:11
  SPEAKER_01: Um it's been more um yeah, get going from a status quo, seeing a drop-off, and then getting back to that same level of traffic.
• 26:21
  SPEAKER_00: Aaron Powell Well, when when people hear about uh a derailment, they think that uh this is gonna take who knows how long.
• 26:28
  SPEAKER_00: But but so so that's really interesting to see like how quickly the the network is able to recover in in these situations.
• 26:37
  unknown: Yeah.
• 26:37
  SPEAKER_01: And yeah, derailments um, you know, they obviously it can vary in uh size um and uh you know impact.
• 26:48
  SPEAKER_01: But they are not gonna say common, uh, but they they're maybe more common than uh what a lay person um would understand.
• 26:57
  SPEAKER_01: Uh, because the only ones that make you know big news are the major ones that we all know of.
• 27:05
  SPEAKER_01: Um but there are small ones um where a few cars may come off um the tracks.
• 27:13
  SPEAKER_01: Um, and that happens almost on a daily basis.
• 27:16
  SPEAKER_01: And those are often fixed within hours.
• 27:19
  SPEAKER_01: Um just yeah, moving the the cars back on um to the track um or removing uh the cars and moving trains through.
• 27:28
  SPEAKER_01: Um and even if there is damage to tracks, uh often yeah, that's done incredibly quickly as well.
• 27:37
  SPEAKER_01: Um so they yeah, the railroads are quite good at um making those repairs.
• 27:43
  SPEAKER_01: Um and yeah, provided there's not, you know, other issues or leaking um hazardous materials, um, we see yeah, uh derailments um clean up very quickly.
• 27:59
  SPEAKER_00: So that's interesting how you how you put all this together.
• 28:02
  SPEAKER_00: So you you are obviously using a lot of automated uh uh you said the machine learning, uh and I'm sure AI, uh the evolution of that has played a role as well, but then you're still ground-truthing with uh old-fashioned uh sort of investigative work, like watching the news or maybe making a few phone calls or things like that.
• 28:22
  SPEAKER_00: So so it's an interesting way of kind of pulling all this information together.
• 28:26
  SPEAKER_01: Yeah.
• 28:27
  SPEAKER_01: And I think you know, the thing that I think we add that um others, yeah, there's not other sources for is um measuring the the impact of the disruptions um and uh so how long they take, uh, yeah, again, what the recovery looks like, and also what um companies or commodities um may have fared um worse um in a particular um disruption than others, um, as well as having um a history of what happens with uh different types of disruptions.
• 29:09
  SPEAKER_01: So um, you know, can't predict derailments, but can uh predict some other things um or have a better sense of their level of risk.
• 29:21
  SPEAKER_01: Uh so for example, um what we see in Canada um every year are uh stretches of extreme cold.
• 29:32
  SPEAKER_01: Um so could be a week, uh, two weeks, and during those times, um train volumes decrease uh and train lengths decrease as well.
• 29:43
  SPEAKER_01: Um so we're measuring the impact of that um as it's happening.
• 29:50
  SPEAKER_01: So yeah, when it's negative 25 or negative 30, um and they can't use uh typically length uh typical length trains, um, have to use um shorter ones so that the the air braking system um can still work during uh the those temperatures.
• 30:10
  SPEAKER_01: Um we can see you know the impact and the history of um how that um happens.
• 30:17
  SPEAKER_01: Um and so then looking forward to the following year um when companies are planning or governments are um trying to um you know understand and analyze what you know economic output is going to be, um we can provide that additional data to them to understand you know the contours of what a disruption um and recovery look like in cases of something that is predictable um and regular, like um extreme cold.
• 30:52
  SPEAKER_00: Well, you have uh you also have the stretches of the open plains where you have the high winds happen sometimes.
• 31:00
  SPEAKER_00: Uh have you seen any kind of patterns in terms of like when when you're getting the double stacks versus when uh maybe we can't do that uh uh or is that more just variable based on uh weather on a on a given time frame?
• 31:16
  SPEAKER_01: Yeah, that one's a bit uh tougher.
• 31:19
  SPEAKER_01: And um, for us, we do need a bit more uh data history um to see those patterns um more clearly.
• 31:30
  SPEAKER_01: Uh but oftentimes we do know that uh that type of um wind impact uh is a lot more isolated than uh what you get with extreme cold.
• 31:42
  SPEAKER_01: Right.
• 31:43
  SPEAKER_01: But it is a definitely an area um that can be studied um and um with some additional data history uh I think could yeah provide some insights on that.
• 31:57
  SPEAKER_00: So uh about a five-year history uh of rail state, we've had uh a freight market that's kind of followed an interesting path in those five years where seems like right around when you got up and running was when things were getting super hot and heavy out there.
• 32:14
  SPEAKER_00: And then things kind of settled down for a little bit.
• 32:17
  SPEAKER_00: Are you are you seeing um now that we are in a situation this year where you know a lot of information out there is that um, you know, intermodal, for instance, has has been pulling a lot of uh truckload volume uh from uh that side of the market.
• 32:32
  SPEAKER_00: Have you have you seen some differences in in the market this year as opposed to in in those last couple of years where things were maybe a bit softer?
• 32:41
  SPEAKER_01: Yeah, definitely see um increases um in in volume in a number of areas.
• 32:49
  SPEAKER_01: Um so I know I think the the ARR has put out some overall um statistics recently, I think up with about a 4% increase um year over year.
• 33:00
  SPEAKER_01: Uh but looking at some particular locations, um, we're seeing higher increases um in some locations.
• 33:09
  SPEAKER_01: So for example, uh Southern California domestic intermodal.
• 33:15
  SPEAKER_01: Um so what we're seeing there is about a 9% year-over-year increase um currently.
• 33:22
  SPEAKER_01: And a place like Vancouver um in Canada, uh international um intermodal, we're seeing about a 20% year-over-year increase um there.
• 33:35
  SPEAKER_01: And something else that we've identified um that explains you know some of the um how the additional um volume is rolling out this year is an increase in train size.
• 33:50
  SPEAKER_01: Um, so not seeing a a large um or noticeable increase in train volume, uh, but are seeing larger trains.
• 34:00
  SPEAKER_01: Um so yeah, UP um in Southern California about 11 trains are about 11, um 700 feet this year, uh, which is about five percent above last year.
• 34:15
  SPEAKER_01: Um so carrying more containers um on the same number of trains, um, which is something that you know is is I think unique to the intermodal world, um, where you can really um adds a significant amount of um boxes and length to a train without uh too much addition in power or uh cruise needed.
• 34:42
  SPEAKER_01: Whereas other traffic, unit train traffic, um can't really add too much um length without increasing the as much need for um that power.
• 34:54
  SPEAKER_01: So that um those changes in train length um and also dispersion of um different volumes from uh through different locations has been something that we've identified.
• 35:07
  SPEAKER_01: Um something else that has been interesting this year is we've seen some ocean carriers switch um volumes to different railroads um without any sort of uh public announcement that they've you know um engaged in some new partnership.
• 35:29
  SPEAKER_01: Um have seen that uh from Maersk um out of Southern California, um switching to UP from BNSF, um and really that's only been since the beginning of July.
• 35:43
  SPEAKER_01: Um also seeing uh CMA CGM uh move to CP out of uh Vancouver and away from CN.
• 35:54
  SPEAKER_01: And that's something that uh have just identified really in the past uh few weeks.
• 35:59
  SPEAKER_01: Um have you know like some additional insights on uh the impetus for those changes, uh, but have noticed those in the measurements um and unique thing um that we haven't seen in in in years past.
• 36:18
  SPEAKER_00: Well, so so the uh independent nature uh of real estate is interesting because obviously the railroads aren't doing things to to measure a lot of this information themselves, but uh uh how does that uh you know change the equation when you're talking to you know potential customers about uh you know the independence of the the data that you're gathering?
• 36:43
  SPEAKER_01: Yeah.
• 36:44
  SPEAKER_01: Um our customers appreciate uh the independent independence of the data uh because they know that we are objectively measuring what we see.
• 36:57
  SPEAKER_01: Uh we're not um taking sides.
• 37:01
  SPEAKER_01: Um you know, objective data like this is not something that yeah, takes a side.
• 37:06
  SPEAKER_01: Um we are uh just yeah, measuring what we see and gives them um additional evidence um and support for the work that they are doing so that when they are going into negotiations um with a carrier or you know looking at new facility locations, they have uh evidence and data uh that helps them make better decisions.
• 37:38
  SPEAKER_01: And you even if it's in a maybe a um a negotiation situation, there is not a major imbalance in um in data or in evidence that uh typically has existed.
• 37:54
  SPEAKER_01: So they can come with equal footing and really get to the heart of uh an issue without you know basically being um in a situation where one side has all the information and they're left with little information and trying to trust that um we're providing them that additional level of trust that they they can use for a variety of purposes.
• 38:24
  SPEAKER_00: So how do you interact with with the railroads, though, in terms of uh, I mean, obviously you're you're not in situations where you need to be right on the tracks or anything, but but how does real estate interact with with the class ones, for instance?
• 38:38
  SPEAKER_01: So we don't have yeah, um any sort of agreements um in place uh for the data collection.
• 38:47
  SPEAKER_01: Um how we're we've been interacting with them generally is that they have interest in the data that we're collecting as well.
• 38:57
  SPEAKER_01: Um so from a you know customer standpoint, uh they know what is going on in their network.
• 39:06
  SPEAKER_01: They don't know what's going on on other networks.
• 39:09
  SPEAKER_01: Um so being able to have visibility into that for um you know cooperation, cooperative uh arrangements that they have with the other railroads, um, as well as you know, understanding things from a competitive standpoint, um it's visibility that they are interested in.
• 39:29
  SPEAKER_01: Um and also we've been able to provide some additional insights that even the railroads didn't know about their own networks.
• 39:39
  SPEAKER_01: Um and it really gets down to that um it our system being um image-based, uh, so we have evidence of something that happens.
• 39:52
  SPEAKER_01: So if uh a part of a a rail car um goes missing, um, or if uh we're finding uh containers that have been opened um en route um as potential uh theft, we are able to provide them with evidence of uh when something um like that happened.
• 40:17
  SPEAKER_01: And not they're not generally able to um have those insights from you know just their uh normal systems that are uh you know collecting information about operations, but rarely from uh uh an image standpoint.
• 40:35
  SPEAKER_00: Oh yeah, I mean uh bringing up the cargo theft thing with with images, that's uh that's a big deal as uh as obviously cargo theft is uh unfortunately continue to be uh a big deal out there.
• 40:48
  SPEAKER_03: So the images you're capturing, that's that's interesting to me because so the images you're capturing, you're able to capture uh the side as well as uh the back of the container in some form or fashion.
• 40:59
  SPEAKER_03: So you're getting a couple shots at it, or or uh how is that how are you capturing that?
• 41:04
  SPEAKER_01: Yeah, so we um we get the sides, um and basically, yeah, we're uh our cameras are set up um in a way to collect the side.
• 41:14
  SPEAKER_01: Uh but as the train moves, um, we can see um parts of um different angles.
• 41:22
  SPEAKER_01: Um so seeing somewhat into the back of um or front of rail cars.
• 41:29
  SPEAKER_01: But in the case of what we've been seeing with theft, uh it's really the doors on a container are open.
• 41:36
  SPEAKER_01: Just flapping open.
• 41:37
  SPEAKER_01: Okay.
• 41:39
  SPEAKER_01: Um we found uh uh last year um there was a train that um we saw at one of our sensor locations uh and there was no container doors open.
• 41:53
  SPEAKER_01: Um and then we saw it at the next sensor location, which was only 10 miles away.
• 41:58
  SPEAKER_01: Um and the train also made that distance in 10 minutes, um, so averaging about 60 miles per hour.
• 42:06
  SPEAKER_01: And at that second location, we we found uh 15 um open containers.
• 42:10
  SPEAKER_01: Wow.
• 42:11
  SPEAKER_01: How in the heck?
• 42:12
  SPEAKER_01: How does that happen?
• 42:14
  SPEAKER_03: Um yeah, it's not like they're riding up on their horse like in the westerns and coming up on the on the rail car at 60 miles an hour.
• 42:22
  SPEAKER_03: Um yeah, how how does that happen?
• 42:25
  SPEAKER_01: Uh yeah.
• 42:27
  SPEAKER_01: Yeah, there well it when a train stops, obviously, you know, things um can happen, but uh have heard, and we're not experts on this, that um you know there are people accessing some of the containers um earlier on in a journey um and will then um come out uh of the containers um while the train is in motion.
• 42:55
  SPEAKER_01: Um so that's what yeah we have heard.
• 42:58
  SPEAKER_01: But it's uh it's a yeah, a wild uh thing to um imagine being on a speeding train like that.
• 43:06
  SPEAKER_00: And um but uh have you had any uh have you had any images where you've actually seen uh people doing anything uh like uh in in in any of these areas?
• 43:19
  SPEAKER_01: We yeah, we have um seen some um we've have captured, yeah, some people on mid-action basically.
• 43:30
  SPEAKER_01: Yeah, it in when things are happening.
• 43:33
  SPEAKER_01: Uh but our our images um are high resolution enough uh to read things on the sides of the rail cars, um so even down you know to container numbers, uh, but not anything that's um high resolution enough to you know recognize a a person in any way.
• 43:58
  SPEAKER_01: And you know, we're not Set up to do that.
• 44:01
  SPEAKER_01: It's not a good thing.
• 44:02
  SPEAKER_00: Yeah, no, I was just curious more, yeah, if I mean as as you're you're following all of these trains going, if you you can see the difference between the doors being open and closed in an area, I would assume that occasionally you actually are catching some uh whether you can actually see their faces or not, but you're actually catching some uh activity uh even occurring.
• 44:23
  SPEAKER_01: Yeah.
• 44:24
  SPEAKER_01: We we've more seen things like doors open with uh boxes um coming out of container doors.
• 44:34
  SPEAKER_01: Yeah.
• 44:35
  SPEAKER_01: There's another thing that you know it's we're capturing things at track speed, so 60 miles per hour.
• 44:43
  SPEAKER_01: Um and you know, we're only at select the locations, so we have to get lucky, uh really lucky to capture some of those specific things.
• 44:53
  SPEAKER_01: Um but we have captured some very interesting uh things around the network um over time, for sure.
• 45:02
  SPEAKER_03: Well, Kevin was saying that you bet you put some stuff up on LinkedIn.
• 45:05
  SPEAKER_03: Is there stuff that's up on your site that gives just general information as going as what's going on in the real network, or is that there's no real public information that you guys put out there other than you have to be a customer to grab some of this data?
• 45:18
  SPEAKER_01: No, we um release reports um uh through our website, but also through our email list um and on LinkedIn.
• 45:29
  SPEAKER_01: Um oftentimes those are related to um something very timely.
• 45:36
  SPEAKER_01: Um but also we do have other reports um on general trends that we're seeing in across the network.
• 45:44
  SPEAKER_01: Um so yeah, if anyone is interested, um they can go to railstate.com um and sign up um for our email list, um, and yeah, we'll receive some um regular reports.
• 46:02
  SPEAKER_00: And that's where somebody would go to if they were interested in uh becoming a customer as well, I would assume.
• 46:07
  SPEAKER_01: Yes, yeah.
• 46:08
  SPEAKER_01: So railstate.com would be the easiest uh place to start.
• 46:12
  SPEAKER_00: All right.
• 46:13
  SPEAKER_00: Well, Dan, we appreciate you taking the time today, and uh yeah, a lot of good information there.
• 46:20
  SPEAKER_00: Thank you.
• 46:24
  SPEAKER_00: I know you're a fan of data and yeah, data-driven decision making, and and that was interesting to to hear uh one of the just the uh use cases for this, how um how shippers wanted to be able to go into these negotiations with some data of their own that maybe they didn't have before.
• 46:44
  SPEAKER_03: Yeah, that my head is spinning on the amount of data that they could capture.
• 46:48
  SPEAKER_03: And as they were talking about, well, shippers gathering this information, particularly as he's he's talking about, say, energy.
• 46:54
  SPEAKER_03: Um, well, how much is my competitor moving?
• 46:56
  SPEAKER_03: I mean, you'd be Yeah.
• 46:58
  SPEAKER_03: Or you're or or you're one of the railroads and you're trying to figure out, well, how much is the competition moving in, say it's the Southern California lane versus what I have moving on the Southern California lane.
• 47:08
  SPEAKER_03: There's just a plethora of things that this this data could deal with.
• 47:12
  SPEAKER_03: Um and as you and I were talking just before we jumped back on here again, um, I didn't realize their technology was was image-based.
• 47:19
  SPEAKER_03: So um, you know, most of the stuff is is you're just capturing data as uh a car is coming by and you're capturing car number, you're not really capturing the physical car itself.
• 47:30
  SPEAKER_03: Yeah.
• 47:30
  SPEAKER_03: Um and so I'm I mean, it it the independence of that data is is is amazing as well, and the things that can be done with it.
• 47:39
  SPEAKER_00: Well, I mean, uh visibility is such a thing.
• 47:42
  SPEAKER_00: I mean, we we just talked with Ann on uh Ann Renke on the last episode.
• 47:47
  SPEAKER_00: Uh there they've just done this this shipper survey, and visibility has been something that that people have been clamoring for when they're doing intermodal.
• 47:56
  SPEAKER_00: So if you can do your track and trace and have this piece.
• 48:00
  SPEAKER_00: I mean, this is truly visible.
• 48:01
  SPEAKER_03: Yeah.
• 48:02
  SPEAKER_03: Definitely visible.
• 48:03
  SPEAKER_03: We're not talking just uh here's here's the scan of the uh at this point in time of of this uh intermodal container.
• 48:10
  SPEAKER_03: It's it's the physical look at it.
• 48:12
  SPEAKER_03: And then and then it was talking about um theft.
• 48:15
  SPEAKER_03: And you know, with um within a certain section, uh I can identify where the theft happened.
• 48:20
  SPEAKER_03: So there's there's all kinds of really great use cases for this stuff.
• 48:24
  SPEAKER_00: Yeah, yeah.
• 48:25
  SPEAKER_00: Very interesting stuff.
• 48:26
  SPEAKER_00: And thanks again to you, Dan, for uh joining us to uh talk about real estate.
• 48:31
  SPEAKER_00: And thank you for joining us again on the Intermodal Logistics Podcast.
• 48:34
  SPEAKER_00: Please do rate and uh review us wherever you get your podcast.
• 48:39
  SPEAKER_00: Subscribe, follow.
• 48:40
  SPEAKER_00: Uh automatic downloads, all that stuff is uh great.
• 48:44
  SPEAKER_00: And uh you can also drop us comments, questions, or email us at podcast at intechlogistics.com and find out more about Intech Logistics the Company at www.intechlogistics.com as well.
• 49:00
  SPEAKER_00: For Rick Lagore, I'm Kevin Baxter.
• 49:03
  SPEAKER_00: Thanks again, and we'll be back soon with another episode.