View the March 18, 2020 seminar recording
Jennifer Symoun
Good afternoon or good morning to those of you to the West. Welcome to the Talking Freight Seminar Series. My name is Jennifer Symoun and I will moderate today's seminar. Today's topic is A Primer on the Operational Characteristics and Impacts of Precision Scheduled Railroading.
Before I go any further, I do want to remind you to call into the teleconference for the best audio quality. If you are listening to the audio over the computer and experience any issues, I am unable to fix them as audio quality will vary based on your network connection, computer, speakers, and other factors. Please also keep in mind if you are calling into the teleconference for the audio, you will need to mute your computer speakers or else you will be hearing your audio over the computer as well.
Today's seminar will last 90 minutes, with 60 minutes allocated for the speakers, and the final 30 minutes for audience Question and Answer. If during the presentations you think of a question, you can type it into the chat area. Please make sure you send your question to "Everyone" and indicate which presenter your question is for. Presenters will be unable to answer your questions during their presentations, but I will start off the question and answer session with the questions typed into the chat box. We will also take questions over the phone if time allows and I will provide instructions on how to do so once we get to that point.
The PowerPoint presentations used during the seminar are available for download from the file download box in the lower right corner of your screen. The presentations will also be available online within the next few weeks, along with a recording and a transcript. I will send a link to the recording in the next day or so and will also notify all attendees once all materials are posted online.
Talking Freight seminars are eligible for 1.5 certification maintenance credits for AICP members. In order to obtain credit for today's seminar, you must have logged in with your first and last name or if you are attending with a group of people you must type your first and last name into the chat box.
PDH certificates are also available for Talking Freight seminars. To receive 1.5 PDH credits, you will need to fill out a form. Please see the link in the chat box. Certificates will be emailed one week after the seminar. A seminar agenda has been included in the file download box for those who need to submit an agenda to their licensing agency.
Finally, I encourage everyone to please also download the evaluation form from the file share box and submit this form to me after you have filled it out.
Today we'll have three presenters:
Our first presentation will be given by Carl Van Dyke. Carl has spent 40 years focusing on the optimization of freight railway networks through the application of software tools and operational expertise. He founded MultiModal Applied Systems in 1992, which became part of Oliver Wyman in 2006, where he served as a Partner through 2013. Past work included the development of the MultiRail service design software system, and working on the redesign of the operating plans for every North American Class I railway, plus a number of railways in Europe, Africa, and Asia. Most recently he has been working on the design of the next generation production systems at two different Class I railroads.
Carl Van Dyke
Thank you. Good morning or afternoon depending on where you are. Today I will take you through a somewhat railway-focused presentation on precision scheduled railways. I apologize in advance for the use of jargon here and there, it is railway specific, hopefully it will be clear enough. I also want to take a minute to give special thanks to the Association of American Railroads and Oliver Wyman for the use of selected slides in this presentation.
The PSR (Precision Scheduled Railway) as identified by Hunter Harrison, father of the whole concept, is driven by five core principles, which are: Asset Utilization, Cost Control, Service, Safety, and People. We start with Asset Utilization because this is perhaps the most important driving force in the PSR process. It is the focus on utilizing all assets, both fixed and rolling, in a way that is at their absolute maximum. And that drives every other decision in the process to a large extent. Coupled with that is Cost Control, making sure you have what Harrison would say is right and tight management of all expenses and examining every process to ensure it is what you need and continually improving that; essentially implementing lean in a railway context. Service is equally important; reliability is a key driver in the performance metrics of a precision scheduled railway. This also implies a certain simplification of the operation to ensure better reproducibility going forward. Harrison argues that by doing these things in a repeatable and more consistent manner, you also then by its very nature drive up safety as well, so that a safe operation is a natural outcome of a more precise scheduled operation. And, of course, people are key because people are who actually produce this entire product. So, he very much focuses on recruiting and developing, and rewarding the right people, training them to understand the principles of a precision scheduled railway, and pushing delegation down to the individuals so that everybody is pulling together.
So, we could say there are a number of key attributes as a result in Precision Scheduled Railroading. One is what we call the Service Plan, which is the nuts and bolts, the pieces that come together to define how we will operate the railway. It is viewed as a sacred process that integrates all the processes. It is built out of pieces, but focuses on how they all interact with each other. Essentially, the concept is it's the network, that you need to focus on the total network and not the individual parts. So, if you can do that and you can execute consistently then you focus on moving towards management by exception rather than firefighting. And doing that, the guiding principles then drive value on leadership and initiative, trying to reduce bureaucracy, levels of decision-making, and rewarding teamwork, sharing knowledge, information, and credit for good deeds. All that comes together in the focus on cost control, again with the network nature of the railroad perspective. So, we will try to avoid local optimization and the unintended consequences, which often drive up the total network cost of the railway.
There are a lot of thinking and ideas and press and commentary and opinions and emotion around precision scheduled railway. I think a few need to be thought about, one is that PSR's aim is only on one-time cost cutting. Just looking at the 20 year performance history of CN shows this not to be the case. Another is that it is driven strictly by this goal to reduce the operating ratio (OR), or operating cost divided by operating revenues. The key performance metrics remain revenue and profit with a focus on reliable customer service; OR really becomes a yardstick in which to measure the lean railroad. Again, focusing on closing hump yards, or major yards. There is a very strong need to reduce the number of car handlings or classifications or switchings, which improves reliability and asset velocity, which will drive down the number of car handlings and in some cases you might end up reducing the role of a yard or closing the hump yard, not for the sake of closing the hump yard, but because you've successfully reduced the number of cars switched. The same is true of people. By pushing the right guidelines and moving decision-making downward you do drive a leaner organization. There are people consequences, we cannot deny that. We also often hear that going to a precision and railway just creates chaos. This is a function of how fast you are doing that. If you do it very quickly, you will perhaps end up with chaos. But if you do it in a more measured way, which a number of railways are currently trying to do, I think you will not see the same kind of chaos. It is not really about scheduled railway, it is really about the transformation process. Finally, this idea there's a focus on de-marketing business. They would term it more decluttering the railway. There are a lot of OD pairs or movements within the railway that don't really fit naturally to how the railway operates. The extent to which you can focus on those that do make sense, you can make the system more efficient. It is important to emphasize it is not a cookie-cutter approach and it will be different at every railway.
Just briefly, what does this look like on a railway? Here is Canadian Pacific's performance from the point that Hunter Harrison took over for the first two years. What you will see here is a steady driving down of expenses. What is interesting to note is revenue per gross ton mile at the bottom of the chart stays pretty flat. At the same time, expenses per gross ton mile go down and that is the goal of the precision scheduled railway process in terms of driving a leaner organization. There is a lot of complexity to how railways operate. There is a pickup process from the customer, there is switching of those cars into trains, movement on one train to another train, switching from one train to another and so on. Each of those has a potential for causing both expense, delay and reliability issues. And it's each of those things that precision scheduling tries to address. I will not explain all of them, but they do underlie the entire thinking of how PSR is structured.
So now I have to talk in more specific terms about the kinds of things that are done by railways in order to achieve the benefits of PSR. We can break that into three areas: Operating strategy, People and Organization, and Capital Investments. Primarily, I will talk about operating strategy today, which brings together the core principles and tries to drive the efficiency as we discussed through cost control and asset velocity. Other areas include capital investment, which applies to both the physical plant and the rolling stock. And there's a lot of controversy on this because often people perceive that there is underinvestment with PSR, particularly in physical plant. However, if we separate that from the rolling assets, it is clear that if we can improve asset velocity, you will need fewer cars, fewer locomotives, and that is a natural outcome that is not an adverse outcome, but simply a benefit of that improved velocity. People is the other big area, there are issues around reduction and what is expected of people in their roles. That is as much about philosophy and management as it is about PSR. It can be a controversial issue because we do see layoffs and staffing reductions and changes in the role of how people work. I will stay away from that as a somewhat politically sensitive area, but it is a factor that needs to be thought about. I would argue it is not an intrinsic element of PSR, but merely one element or approach to implementing PSR.
So, given that, I will talk about four areas in terms of operating strategy. One is local service, the second is increasing car velocity or asset velocity, the third is to minimize the number of trains you operate and the associated road locomotive and crew needs, and finally, the impact on customer service. I think it is important to note that PSR is often characterized as making its improvements by increasing point-to-point moves, and it is really the opposite, as the long train philosophy that is used by PSR actually causes more traffic to be treated as general manifest traffic where is has to be switched between one train and another. So, it is really a different focus; it is more a focus on being very good at running the trains and switching the cars more so than trying to be point-to-point service.
So, I am going to take you through each of these areas relatively quickly in terms of how they are changed in the PSR world. The first is local service. This is the action of taking cars to customers and picking cars up from customers, and we do that with local crews and local locomotives. And one of the first things we want to do is rethink how many local crews there are, where they go on duty, in order to maximize their workloads and consolidate duty stations so that you can have more than one local job at the same location, which then permits you to share power/locomotives on a 24/7 basis. So, trying to move to a 24/7 basis with balanced workloads ultimately will improve your asset utilization and your labor utilization. Another thing you try to do is ask more of the local crews, in particular doing more switching in advance, so that when cars are brought into the yard, they are grouped in such a way that they can move deeper into the network without more switching at intermediate yards, which ultimately improves car velocity. In the same way it (PSR) improves the quality of the work that they do to ensure they get a full day's work for a full day's pay.
We also ask more of customers. Again, just like we asked crews to do more preparation of cars, we do the same with customers, asking them to identify large groups that can be moved deeper into the network. We ask customers to move to a 24/7 basis, both to improve the efficiency of scheduling the local crews, but also ideally to move towards 7 day loading and unloading to reduce peaking, and to be able to turn cars faster. So, a lot of different impacts. I will highlight one in particular, which has always been observed, which is you use fewer locomotives. Typically, in a large Class 1, 500 or more locomotives will come out of the local service simply through the actions we just talked about.
Second, as I mentioned, is increasing car and asset velocity. Part of this is a continuous improvement organizational focus issue: build the operating plan around asset velocity, look for opportunities, monitor the network constantly to ensure that you can see opportunities and drive at them and change them, and have everybody throughout the organization empowered who makes those changes. Another simple example is to move towards at least two departures per day for every large group of cars so that the delay time, if you miss a connection, goes from 24 hours to 12 hours – another simple strategy for increasing car velocity. Very important to drive up connection performance and trip plan adherence. That earlier diagram I showed you showed how you are going to relay cars from a local train to a road train, may be from one road train to another, and so on. If you can drive plan adherence, so that the right cars are getting on the right trains at the right time, then you can improve car velocity by reducing error rates. And so, they have put in a lot of systems to do quality checks, to look at if right things are being done, and not just take note of them, but to take direct action to go out and call out the field people and say you did this wrong, you need to correct it, not in the future, but now. And in some cases, they actually say take the train apart and try again and get it right this time, and people quickly learn to do it right the first time. It is a big change in culture to get there. And then as I mentioned indirectly already, lots of strategies to reduce the amount of switching so that you have fewer connections between trains, fewer cars handled at yards, which both reduce your cost and improve your reliability of the network. So, you get all of these benefits that ultimately drive in terms of reduced errors in the network, less switching, improved service. It does drive more complex trains, which requires more discipline to execute, but it drives down capital and labor cost, improves reliability, and so on.
The third area they focus on is minimizing train miles and minimizing locomotive and crew needs. One-way they do that is simple, they try to run fewer, longer, more generic trains. The rail industry has gotten into a pattern where they run a lot of specialty trains for particular customers, particular commodities. One emphasis is to move away from that and to mix the traffic more on trains so you can provide more departures for traffic, because you have more different trains to use, but also drive up the size of the trains, which reduces the total number of trains operating. Another is to balance the network, ensure that you have the same number of trains going in one direction as the other direction, because that naturally then also balances your locomotives and crews, improving your crew utilization and your locomotive utilization in that process. And if you can run the trains on time, that also indirectly improves the ability to manage and plan those crews and locomotives, in addition to becoming a better producer of quality trip plans for your customer. One of the things they try to do in the process is use strategies of what we call dynamic blocking to look for opportunities to fill out trains and ensure those trains are filled, staging and monitoring traffic to do that process is done a way that again will drive down intermediate switching, increase train size, and increase velocity.
I pretty much covered the top part of this page, which is focus on crew and locomotive cycling. It comes from the idea of balancing the network, so you have a natural cycle in the process. Other elements of that is to get rid of what they call protection locomotives, putting an extra locomotive on a train just in case one might fail. If you do a better job of maintaining locomotives you won't have to do that, and you will need fewer locomotives. Squeeze out idle time; if locomotives are sitting around for a period of time waiting for a train, can you do something else with those locomotives? It might break some silos to do that, but move away from assignment silos and try to be more flexible to maximize the use of locomotives. And then finally, flatten peaks. Do anything you can to encourage customers to ship regularly on a seven-day basis, so we have less peaking in the railway and we have less imbalance in needs for locomotives, crews, and the movement of trains. What they found is that you can reduce the road fleet, this is in addition to the local fleet, by 500 to 1,000 (road) locomotives on a large Class 1, getting better capacity utilization in the process. They would claim that they also have seen improved velocity and reliability, less switching, reduced intermediate handling, ability to close yards, so on, and tighter ability to manage traffic by monitoring that. Overall, the goal then is you have better predictability and fewer deadheads and less idle time.
And then the last area of impact is customer services. And here the goal would be to eliminate boutique blocks in trains. By that I mean try to move away from one-off customer solutions. Often those one-off customer solutions were put in because the railroad was running unreliably, and those one-off solutions were trying to get around the poor performance of the network. But if you can simplify and improve the performance of network and make it more reliable, then you don't need these boutique solutions. In addition, boutique solutions, particularly things like one-off unit trains where you have to spend 3-5 days accumulating enough cars to run the train, end up destroying your asset velocity and also create massive imbalance in terms of train movements and crew movements. So, all of these things move towards eliminating these boutique solutions and trying to standardize the product offer. So, we try to then have the same services across all customers, with increased operating discipline and velocity, with fewer customer-specific services, tighten the use of unit trains, no deep dives for customers. If you can do it right the first time you don't have to rescue customer traffic. No single customer commitments; standardize movements and make sure everything fits within the product offerings you have. I think we covered pretty much the other impacts.
So, basically the idea is moving towards plan simplification, easier management, better velocity, better train capacity utilization, fewer trains, all of these things. You do have some possible negative service impacts on specific customers, but ideally those are countered by the improved velocity and reliability of the system, and you certainly have a simpler commitment process. That gives you an idea hopefully, very quickly, of some of the things the railways do in a specific manner to address and implement PSR. I think it is important to think about that PSR is not achieved through a single action or a small number of actions, but through a collection of actions that cumulatively result in a significant change in the operating practices and culture. You can't point to one thing; it is many things. Without doing all those things, you are not likely achieve the benefits you would like. No single PSR railroad does all of these changes, but most of them strive to do the majority of them. Finally, Hunter Harrison did publish two books on precision scheduled railways. They were published for internal use at Canadian National, but there are copies available often in the used market if you want to dive deeper, a good thing to try to read at some point. That concludes my talk and I will pass the baton back to Jennifer to introduce the next speaker.
Jennifer Symoun
Thank you, Carl. I see we do have a number of questions; we will get to those after all the presentations but continue typing them in please. Our next presentation will be given by Jeremiah Dirnberger, a Product Manager for Digital Yard Solutions at Wabtec Freight. Looking at the rail ecosystem, he works closely with customers to ensure the rail yard of the future unlocks value and propels the industry to new heights. Previously, Jeremiah worked for two Class I railroads (CSX and Canadian Pacific) supporting field operations by performing simulation, data analysis and capacity planning. He started his career at the University of Illinois writing his thesis on improving hump yard operations. He has been published in the Transportation Research Record and several conference proceedings. A member of two AREMA committees and the INFORMS Railway Applications Sections, he is a regular presenter and panelist at industry conferences.
Jeremiah Dirnberger
Thank you. Let's dive in. We will be looking at, or I will be talking almost at a more execution level of how some of these principles that Carl talked about are being implemented inside of railyards on the North American Network. We are going to look at three main areas. The first being that there is a unique challenge, Carl alluded to it, but we will talk a little bit more about that. And then we will talk specifically about PSR and yard operations, looking at an example I lived through recently working with a Class 1 with one of the systems that we are implementing. And then I will talk a little bit more about the future and the opportunity to use digital and data technologies to really transform the planning and operations across shortlines and Class 1s, because one of the interesting challenges that we face is that we are a network of networks.
Let's talk about that unique challenge first off. So, as Carl showed in one of his slides as the shipment moves from origin of destination on a single network, there are several handoffs. It goes from a local train service, the shipper takes the car, it is handed off at a local yard to another train that takes that to the next classification yard and so on through the network. What is interesting and a challenge that is fairly unique with rail transportation in North America is that for a typical rail shipment, a good portion of them, move from one network to another. So, say a short line railroad would pick up the originating traffic and take it to a Class 1, the Class 1 could take it to a regional railroad which then would hand it back to short line for delivery. You could go from Class 1 to Class 1, go short line back to short line. Several different combinations that occur as the origin and destination pairs are being served through these multiple networks. So, everywhere the handoff is happening, both intra-network and internetwork from one to another, you have got the opportunity for inefficiencies and service failures at these connection points. Those connection points are yards, terminals, or interchange. So that is why a lot of the focus in the PSR mindset is that eliminating those handoffs in the service plans, let's just skip them, let's avoid them, let's go ahead and move bigger blocks deeper into the networks so you don't have to have the handoff. But at some point you're going to have to have them; it is just a fundamental nature of railroading.
How do we balance the operating principles (It is the plan period, don't deviate from the plan) with there is always going to be variability? What PSR is trying to do is eliminate that at all costs, and there is different ways that that is done. What we are starting to see and what we are working with the industry to do is let's look at automation and using more automation as well as enhanced visibility to the data, plus really putting in decision support that will start to unlock that next level beyond PSR. PSR has really improved the bottom lines of several railroads, that is why we are excited about it, and the operations have changed, but at some point, there is only so much that people can do, so how do you move to the next level? So, what we have seen, let's focus on the yards, is that at a network level, as they have gone in and made changes to the service plan and done various things to implement these PSR principles, there have definitely been operational disruptions. Both good and bad depending on who you might talk with. And so, the industry is working through that. Some railroads are going with a softer, gentler approach, therefore less of a disruption. Some went very fast, more disruptions. But as far as in the yard space, technological disruptions have yet to gain a lot of momentum. Carl referred to the hump yard versus flat yard dynamic in railroading that PSR has highlighted, that there is a tendency because of the flexibility and the tenets of PSR, they tend to prefer flat yards over hump yards. And that is what we have seen in the industry is that the ratio has tended to grow and continue to grow more towards flat yards versus hump yards. So, that is where a lot of the focus has been, the change of going from turning a hump yard into a flat yard or just closing a yard altogether. But as far as the investments go in terms of moving a yard and the existing yards forward into the future, there is very little of that. And we see selected automation, a lot of decision support is very minimal, planning and execution; as much as they want to focus on the network at a yard level, it tends to be very localized. Although, there are things we will talk about here in a minute that are being done to try to take away some of that localization. You know, asset tracking is still not done in real time, and then the network considerations can be very ad hoc. And so, this is really the challenge that we are facing. So, how do we begin to move and take PSR to the next level?
And so, we will talk about a real-world example of how this is being implemented inside of the classification yards at a Class 1. We have been working with the Norfolk Southern Railway for a few years on a system called Hump Yard Planner; it is an end-to-end decision support tool. Last year, the pilot was deemed successful. In October and November, we were there on the front lines with the railroaders. And in that process, I was able to observe and learn and understand more about how they are implementing PSR on the front lines in their railyards. So, as we were building the system, it was done in "traditional railroading". That is something when you look at any of Hunter's books or work or talk to any of his disciples, there is the "traditional railroading", which is that there is the train, there is a schedule, but we are not worried about the on-time performance; if the train needs to be annulled or canceled for some reason, that might be good because of the power situation or whatever; you know, those kind of things tended to happen, but there are ripple effects in the network. Well, that is very different now where the plan is the plan; the train runs whether it has got 2 cars or 150. So, those kinds of principles now needed to be incorporated into the tool we had developed and so we are working on doing that right now. But from that, what we observed was that the core tenet, if you want to think of there being an underlying alternate goal of PSR, is it has reduced variability. Because if you can do that, you can reduce your buffers that exist whether it is infrastructure, whether it is power, whether it is people, and you can cut those back which allows you to be a leaner operation. And that is PSR at the core.
So, how this is being implemented in these large classification operations is that, initially, it is being done very non-technical, very nondigital. In fact, I was at CP (Canadian Pacific) very shortly after Hunter Harrison had announced he was leaving and the rumor was he was going to CSX, and the CIO there said I was told by my (soon to be former) boss, if it is more complicated than a spreadsheet, we do not need it. And so, that may have been a management philosophy that was there for a while, but that is going away quickly. This next generation of disciples of PSR, they know technology is the way to move to the next level. And you see that if you look at the curve of a company like CN that has had the longest exposure to it or CP, being close behind. So that what you see here is if I am trying to do this without technology in a yard, I do that by getting variability out.
So, one of the principles is process the oldest traffic first; first in, first over. It does not matter if there is a tighter connection that might need to be made, just always do the oldest traffic first. If you do that, you get the cars into the classification yard sooner. And then there is another piece to this where there is a start build plan, where you cannot start building a train until a certain point in time, and that is defined in a rigid playbook. There is a set time of day and the playbook says at 11 o'clock you can start building the train XYZ and you cannot start until that point. And if you do, there are consequences if you start before that point. So then if you have got both ends of the spectrum, cars get in as soon as they can and cars cannot start until the last possible minute, that is how the plan is designed, I will make as many connections as possible without having a system dynamically managing that. The other in-between piece is that, you can see in the picture there those tracks, that is the classification yard with 50 tracks at this particular yard, and they have moved toward a static assignment. So, if I am a car going to Birmingham, those cars always should be placed in track 10, for example. And they actually measure that, and if every car gets put there so that every time I go to do that work (right back to Carl and the repeatable process), I always go to track 10 when I need the Birmingham cars. So, the crews come to expect that. They start to understand there is a certain length of time it takes because I am always going the same place, so they can start to build their processes around that without a digital system managing that. That is really where all these things are being done, and this is how you put these into practice without a tool to manage it.
Another one that is a neat thing to see happening is that, as part of the asset utilization push, if I have got the available time in the hump, which is the part of the process that is the most automated, and my bottleneck is on the build or the pullback site, I can use the hump. And they are encouraged to use the engines doing that job to help out and do the work that is going to help the bottleneck out. And so that serves two purposes; you increase your bottleneck's capacity so you can get more throughput, but it helps out in asset utilization that you don't have things sitting around or crews not doing work. And number five, above all follow the plan, follow the plan, and follow the plan. So, that is what we saw and we are putting this into a digital tool that had been developed so we cannot only take these rigid principles, but also give it some dynamic features so that we can leverage and when variability does happen there is a set response and it is going to be able to be managed and done through Yard Planner.
So, now that is inside of a node, and we will take the rest of the time here and talk about going across the network, going into an area that PSR has not touched on directly. I think there have definitely been indirect impacts, and Peter is going to talk about that here next. But as we are looking at one network, there are things that I can do on intra-network and that is what PSR has focused on. You know, I'm a CSX or a UP or NS or any of the big Class 1s that is trying to put this into practice, they can control their network. And they definitely can do that, and they do everything they can to manage that. They are going to try to look across and leverage those assets and use them as much as possible. So, there is technology that we are working on to take that to another level, balance your workloads across terminals, be able to squeeze further dwell time reduction beyond what the plan has enabled you to do so you have more throughput and increase your asset productivity.
The next opportunity is, what about from network to network, we started to go from carrier to carrier, what can be done to streamline that and increase that asset velocity moving through one railroad to another railroad? A lot of that has to do with data integration and there is a large amount of work being done to seamlessly pass the data between carriers and then ultimately to the shippers and receivers so they can know where their shipments are and be updating them continually as things change. So that is a point to start down this path, but what else can be done as we think about going from short line to regional, to Class 1, to help drive increased supply chain visibility, to reduce dwell time, and ultimately deliver on the service reliability games that PSR has promised and is starting to deliver on, but maybe there is a limit because a lot of the shipments do involve more than one carrier.
Finally, the third opportunity that we have seen that working actively on is really going throughout the supply chain and really giving that rail pipeline a visibility and planning solution. In fact, we've had a successful pilot with this that is being rolled out now where rail customers are able to connect the data dots, so to speak, to be able to see the visibility in terms of where shipments and assets are through the network. But then ultimately, could this be done in a way that then provides the information back to the originating customer to then load those cars in a sequence will help take work out downstream. And that is actually the last point I will talk about here. As we think about upstream and downstream flows, we feel there is a lot of potential to reduce the work as we go from one carrier to the next or one yard to the next. Because, essentially, the plan is for that yard and the plan is what it is, but as things dynamically change, could we go ahead and provide information upstream that could take out work downstream and allow the assets to flow through the network faster? So, we did some work last year looking into the future and how we could turn this into reality, and we have got some working concepts. And one of the ideas here is a multi-carrier block optimization or downstream planning, where if I am a shipment or one railroad and I have my yard, and I am sending a train or a block of cars to the next railroad at their yard, is that things are being done dynamically depending on what is going on and the conditions of each network to set up the next yard at the next railroad for success and take work out at that next location. So there are a lot of moving pieces to make this work, but in order to have a similar look and feel to what people have come to expect in the day of Amazon and other home delivery services like that, the railroad industry is going to have to start moving more in that direction. PSR is a very good initial step towards that in some ways, but there are other things that need to happen to help move the industry to really deliver on service liability that, just by the nature of how it is set up, can be challenging to do.
So finally, I will finish on the things that I see as what we need to do as we work towards the future. Really, if we are going to continue to move and advanced the railroad, which would be the key to unlocking the service reliability that the industry needs. I remember reading an article years ago that said, "the battle for service is won or lost in the rail yards." We really have to have data visibility across partners. Class 1s aren't always willing to be open and share data freely with the short line and the regional railroads, and that has to change so that it is seamless and real-time. Asset tracking; definitely there needs to be a better way than AEI to actually know where an asset is. Not just location or every 20 miles or it is in the yard, I need to know what track it is and where it is without somebody having to have manually entered the data into a computer. Integrated planning tools; we are doing a lot of work with that. Really starting to make proactive decisions that consider the network, because in the end "it is about the network", and so that is what ultimately matters. So finally, as we start to do, and this has been done to an extent from the dispatching of these Class 1s as part of the PSR, they have sort of consolidated the dispatching back into a centralized location. Can we do that sort of thing with the yards and start controlling them more regionally and ultimately maybe at a higher level? And so, I will finish with this. Railroad automation, which is one of the things the industry is working towards. It is about people first and understanding how the people fit into it, designing your processes around that, and ultimately fitting in your digital products that will enable that. With that, I will pass it back to Jennifer for the next speaker.
Jennifer Symoun
Thank you, Jeremiah. Again, everybody please continue to type your questions and we will get to those at the end. Our final presentation will be given by Pete Swan, an Associate Professor of Logistics and Operations Management at Penn State Harrisburg. He has been a member of the Penn State faculty since 1998. Dr. Swan began his career in the railroad industry working for CSX in operations, marketing, and information systems. After ten years in the railroad industry he returned to college for his Ph.D. in Operations Management from the University of Michigan.
At the University of Michigan, Dr. Swan was active in the Trucking Industry Program, where he participated in both the Driver Survey and the LTL Case Study. Dr. Swan's research primarily involves transportation economics, transportation operations, supply chain modeling. He co-authored papers on carrier effects on total logistics cost, effects of road tolls on trucking traffic and crash costs, as well as a paper viewing transportation security through the lens of agency theory.
Pete Swan
Thank you. So, I don't need to talk much about myself now; we have heard all of that. I would like to say I think I add two things to this webinar. One is that I look at PSR from a view of operations management. There have been some operations management terms that have been thrown around, but I think I add something to what has already been said. The second thing is that, although I no longer work in the industry, I have been an interested observer for many decades. And I am kind of an outsider; that permits me to say things that other people that are not outsiders cannot say. So, what I'm going to talk about, I will start operations management. And then I'm going to talk about what I have seen as the effects of PSR after reading testimonies from the Surface Transportation Board on demurrage and talking with many customers and shortlines; I will relay what I've found. Then I want to talk about what I see as philosophical conflicts between PSR and what the railroad industry is maybe thought to be or supposed to be or wants to be, and then some questions for the future.
So, previously, railroads had a goal of maximizing revenue and therefore profits from a portfolio of assets and weren't as concerned with optimizing utilization of assets as maximizing the revenue from the assets they have had. That obviously is changing now. Railroading is also a batch production industry. Batch production continues to dominate, and that is important because we are going to talk about lean manufacturing. Lean manufacturing is most applicable to repetitive processes like assembly lines. It can be used in batch, and is, and can be used in service, but the main purpose of the is repetitive. On railroads, we also have physical bottlenecks. The previous speakers talked about the yard trim as being a bottleneck in the hump yard, you also have bottlenecks at junctions, bottlenecks at other places in the system, and those limit capacity. So, historically, railroads have tried to figure out ways to maximize the amount of traffic that could get through bottlenecks to maximize their profit.
Now, one thing that's very important to understand about the railroad industry, is that it is a service industry. Service industries are susceptible to having variable demand. And that is especially true for railroads because you could also describe the demand for rail freight as a dependent demand; it is dependent on the demand for the goods being transported. So, if the auto industry has a bad year and there aren't cars being sold, the railroad is not going to move cars. And additionally, as a service industry, the railroad has no ability to store product. So, what PSR attempts to do is take this service industry with variable dependent demand and change it to something more of a less variable demand and maybe one less dependent, maybe not. You know, it is interesting talking to people in the railroad industry, they will tell you they love it because every day is different and probably still is, but the goal of PSR in many ways is to change that and make every day the same, which I find interesting.
So, back to operations management. If you are going to operate to solve problems with bottlenecks, you choose what is called theory of constraints, and that focuses on bottlenecks to increase throughput. I will not go into great detail, but that is what railroads really used to try and do. Today, they are moving to what you would call lean manufacturing, a philosophy of waste reduction. And the focus then, especially for railroads, is on asset utilization; you have heard this already, it is very true. I would go a step further and say that one, what you maximize asset utilization with is with level loading; to keep them being used 24/7, to keep them in constant motion, and to have them moving at basically the same speed all the time. Again, think assembly line here. I did not hear it today, but in the previous presentations on PSR we heard about Six Sigma. So, I am going to skip that because nobody has mentioned it and I don't think I need to talk about it. One thing I will say is that there is a law of variability on operations management. And it states that variability in the manufacturing process or variability in the demand reduces the efficiency of that process. Hence, level loading of the railroad will increase efficiency, which is what many railroads are attempting to do and doing with some success.
So, what have we got with precision scheduled railroading? Well, ops, we've reduced cost by reducing reliance on hump yards and that has been gone over; I will not talk much more about that. General purpose trains have been talked about. What I will reinforce is that asset utilization under PSR has improved dramatically. That's been mentioned that the railroads need fewer locomotives, they also need fewer crews, fewer yards, and many shippers have seen reduced trip time average. I have seen some significant reductions. I have also seen some significant reductions in standard deviation of trip time, or variation. There are clearly some significant improvements for shippers and what precision scheduled railroading has done. Additionally, what other people have mentioned is the improvement in operating ratio. So, I don't think precision scheduled railroading in and of itself is all about operating ratio, but unfortunately for the railroads, Wall Street has made it all about operating ratios. So, I think to some extent, there is some pressure on the railroads going forward to improve that.
There have also been some bad results of precision scheduling railroading. Because they have reduced the number of yards that are handling trains, you see an increase in circuitous routing for some cars; not all, but some. And you are also seeing longer transit times and more variation in transit times for some shippers. So, with the transit time thing, there are clearly some winners and some losers here. The worst effect that people have seen is the first and last mile; spotting cars to customers and picking them up. This is particularly true in areas where there is a low density of rail traffic. What I have seen is that short lines and shippers that are working seven days a week and have significant business have seen significant improvement under PSR. But the ones that are maybe interchanging three a week or they are getting spotted three times a week, in general, they end up being the loser. And I think one of the reasons for this is, we mentioned the story of the two locals that operated during the day and under PSR they put one at night and one at day and they both share the same engines and now we need 50% of engines. That works fine where you have multiple locals. But, if you have a yard where there is a yard crew and a local that operates some of the time and some at other times, the tendency in PSR is to collapse that all into one crew and they will then do whatever is most important. It is hard to step out and go three days a week somewhere. In fact, there is one short line I talked to that used to get 50 cars interchanged three days a week that is now getting one interchange of 150 cars once a week. Unit trains on some of the customers, it is a gain for the railroad to put them in general purpose trains, but in some cases, if the traffic density on some of the lines isn't high, it ends up with longer transit times. I've seen shippers have to put more cars into a fleet when unit service train was ended. Railroads have also increased demurrage charges and have changed the demurrage rules. It used to be for railroad equipment you got two days to unload it. If you did it in one day, you got a credit to do it again for three days. Now there is one day to unload railroad equipment and if you are using private equipment, you have zero days of free time where you used to have one. As a result, the demurrage and accessorial charges at CSX were 1.8% of revenue in 2014, they are now 3.3% of revenue in 2018.
The other side to PSR that really is not PSR that a lot of people are having trouble coming to grips with is what has happened with labor and finance. Railroads under PSR have been shedding employees at a fast clip, let me put it that way. As an example, CSX from 2014-2018 dropped 27% of their employees; officials down 41%, professionals have been down 62%. You would expect to see drops in maintenance and equipment people with fewer locomotives, you would expect to see drops in maintenance of way with some routes deemphasized, but a lot of these job losses have nothing to do with that. A lot of them come in marketing and customer service. So, it is very difficult for customers to find relief from demurrage bills that are wrong. One of the classic comments I got was that it takes six months to resolve a complaint, and you have to take it to the vice president level before you can get it resolved, because the vice president is the only one with authority to do that. There is also a financial thing going on. So, the operating ratio is one thing, but many of the railroads have been replacing stock equity with debt. BNSF, which is private, doubled its payout to the parent Berkshire Hathaway in 2018 over 2016, while borrowing money for capital expenditures. And Norfolk Southern purchased $2.7 billion in stock in 2018, partially financed by $2 billion in debt. UP had an $8 billion stock buyback program financed partially almost by $3 billion in new debt. So that is the bad.
Now we get to the ugly. I have a picture for the ugly. This is out of a Trains magazine. The ugly part is that there is freight that is vanishing. So, one of the things I heard from many customers is there are certain locations that were demarketed or the service deteriorated to the point where they no longer ship rail. That was a theme I heard from a lot of people. I also heard that it was very difficult to secure rates on new business given that there are fewer marketing and sales employees, and the ones that are left typically have moved around and are not where they used to be and may not know everything they need to know about the market. Concentration of authority at higher levels in the organization makes it harder to get a rate. And additionally, the costing systems which worked okay under the old regime are no longer working so well under PSR. Return on assets has become more important than customer service, more important than profit even, and more important than growing traffic.
So, let me get to the philosophical conflicts and then I can end pretty quick here. Lean manufacturing is trying to change the railroad into a level loading facility, or an assembly line. That is the operations, but the reality is that railroad business is still a service business. So, can railroads change variable dependent demand into a less-variable, more-elastic demand? I question that and I question the attempt to try and do that to the extent that the railroads are doing that. Growth versus maximizing asset utilization. The people who say that the US railroads are going to be able to grow are going to be able to pivot the growth look to Canadian railroads, which has grown quite a bit in the last few years. But you look at where the growth has come. It has come in intermodal, in grain, in petrochemical, in increases in existing businesses. And, the networks are much simpler; they are linear versus the U.S. networks which are not so simple. I very much doubt that growth is going to happen in merchandise trade for the U.S. side, but we will see what it is in the future. Finally, railroads are supposed to be common carriers that have a responsibility to carry freight from anyone who needs something carried out at a reasonable rate. The demarketing of lanes of business, which has happened in intermodal in particular, seems to be a violation of this common carrier responsibility to me. But you would have to weigh that against the fiduciary responsibilities to stockholders to make sure that they get their money. How does that play out? I don't know, but it's up to the Surface Transportation Board to deal with that issue and I think they are starting to think about it, but we are a ways off from having it dealt with.
What does the future look like? Well the good side, dramatically improved asset utilization; locomotives, cars, crews at lowers costs; CN and CP will continue growth with good returns, I have no doubt with that; and U.S. stockholders and the other big four in the US are going to continue to have good returns as well. This is, by the way, prior to the coronavirus, so give me leeway there. The bad is bad implementation. Has been, still is, but it is short-lived. This is one of the things I find interesting, and that is that one of the big reasons the railroads hold PSR as this big thing is all the cost it is saving. But it is really not saving cost for the logistics system, almost all of the savings are getting sent to shareholders. Yes, some shippers have quicker and more reliable service, but I don't see the difference as being significant. The big cost savings are in labor and assets, and those are basically all being sent to shareholders. The ugly. Continued signs to point toward continued loss of carload traffic, the returns to other stakeholders other than stockholders, and a possible demise of common carrier nature of railroads going into the future.
So, that is the end of my comments and I will stand by. Thanks.
Jennifer Symoun
Thank you. We will now go ahead and start off the Q&A session. I know there have been some audio issues right now, not sure if that was coming from Pete's line or the phone lines in general, but hopefully we will just continue on through the end, and hopefully make it through and everybody can hear me clearly. Since we just finished Pete's presentation, I am going to go ahead and start with questions for you, Pete, and then we will go back through the other presentations.
The first question actually is a question for Pete and Carl. Is the prospect of road locomotive fleet reductions likely to enable railroads to modernize their fleets by retiring old, lower tier, less reliable locomotives rather than keeping them on hand as extras or relegating them to low volume locations? So, Pete, we will start with you.
Pete Swan
The answer is yes and no. I don't see any of the major railroads other than maybe CN buying many tier 4 locomotives, the newest. I see them rebuilding some of the older locomotives and basically shedding some of the fleet. Many of the locomotives that are dropped out from PSR will be sold to either scrap or leased to short lines. They will rebuild stuff which will be a reason to keep some of the locomotive shed as they're being rebuilt. But I don't see any new locomotives purchased.
Jennifer Symoun
Carl, any thoughts from you?
Carl Van Dyke
I think he pretty much covered it. You are bringing down the average age of the fleet by retiring the oldest and most unreliable units. You are modernizing, indirectly, the short line fleet, because they tend to be the buyers of the retired Class 1 fleet. And it has caused a near stop in new locomotive orders, which is a significant issue for the locomotive builders who are seeing their order books dry up in the short term. And it may be 5-10+ years before they work through the existing fleet and start to be ordering again in any meaningful way.
Jennifer Symoun
All right, thank you. Pete, another question for you. Under the ugly, has there been a shift to move stuff by truck instead of rail even for longer distances where rail is supposed to be more efficient? Did we lose Pete?
Let's see. It is a little choppy still.
Pete Swan
Okay, I'll stick to this then. So, many of the shippers who have dropped rail service either have gone to truck or have closed the facility. Where you are seeing the most traffic moving to truck is in intermodal, where railroads have dropped intermodal lanes and trucking has taken up the slack. And the railroads have been very reluctant to lower prices, either for carload or for intermodal space. So, they're losing a little bit of freight to trucks on the margin, because they want to see revenue margins maintained for intermodal.
Jennifer Symoun
All right, thank you. I think we have covered the ones I have seen come in for Pete. So, let's move on. We will start at the top with questions for Carl. The first question for you with regard to PSR introduction and implementation, what role have railroad labor unions played in the process, the good, the bad, and the ugly?
Carl Van Dyke
I am not sure I fully answer that. I think it is varied by every single railway. I think, in general, the unions have not been that happy with the process in the sense that it is both reducing their number of members and it is also tending to push up the amount of work asked of individual employees. And this has been an ongoing point of conflict. In a very narrow constructionist way, the railroads are not asking in general for employees to do that much which was not already allowed for within the regular contracts, but they are pushing them harder within the limits of those contracts. There have been some efforts, for example, Canadian National wanted to implement the same thing they did at Illinois Central where crews could swap halfway, and everybody would be home at night. And they have been unsuccessful in implementing that strategy due to reluctance by the unions. So, it is a very mixed bag, but the downsizing of the labor force is clearly generally viewed as a negative by the union.
Jennifer Symoun
Thank you. And, I should mention, for any of these questions, if any of the other presenters want to jump in and provide responses please feel free to. But I am going to move on. Carl, another question for you. I have heard there is a cooperative arrangement, at least in the U.S. system, where in Class 1 providers agree to share their rolling stock when available. How sensitive is PSR to participation or nonparticipation in these types of cooperative arrangements?
Carl Van Dyke
They do share rolling stock. There are multiple sub fleets and there are different rules for each sub fleet. It is quite a complex situation in terms of depending on what the line of business is and the car type. In general, I don't know that PSR has really changed that overall pattern in any material way. When you do have another railway's car on your property, you pay a demurrage charge or car hire charge for having it on your property. And certainly, there is always, particularly in times of recession, a tendency to try to shove the foreign rail cars off the railroad as quickly as possible to reduce those payments while maximizing the loading of your own equipment, even if that equipment is going onto the other railway. But that has been going on since time in memorial in the industry. There is some downsizing of the fleet coming out of PSR and that is definitely an impact. But in terms of the sharing process, I don't think there has been a material impact.
Pete Swan
Jennifer, this is Pete. There is one thing I would like to add on that. So, one of the reasons, I think, that may be attractive for PSR is that there is one sub fleet of boxcars where they are getting very old and large numbers of boxcars are going to drop out of being available for service in the near future. And I think one of the hopes of PSR is that the number of trips that a boxcar can make or the mileage it will do, will improve its utilization to the point where it will be cost-effective for the railroads or others to build new boxcars and rejuvenate the fleet at some point. Because paper companies in particular are kind of like looking at a cliff of when boxcars will become so rare it will be hard to find cars to load paper into.
Jennifer Symoun
Thank you both for the responses on that one. Carl, another question for you. Has the industry observed significant changes on the part of shippers to try to adjust work outputs to produce shipments on a more regular basis to reduce peaking and shipping needs at a regularized shipment? Are there any examples you are aware of?
Carl Van Dyke
I can't cite specific customers. I can speak to the generality, which is it really depends largely on the size of the shipper. In general, where the railroads are asking more for the larger shippers, it tends to be a negotiated process with those shippers. There are some that are very accepting and supportive of the process and there are others who are not in particular. They generally are operating on a five day basis. The railway tends to come in and try to persuade them to operate on a seven day basis, and that may or may not fit with their economic and business model, and that can become a point of contention. There is nothing that forces the shipper to absolutely operate seven days, so the railroads will also try to do various types of incentivization to push that. So, it can be a point of contention, and it has been, particularly during the early implementation process of PSR in individual railways, a major issue along with requests to get the shippers to make blocks that go deeper into the network. If the shippers can see the benefits of doing that in terms of improved car velocity and so on and shipment reliability, then more often than not, they do come around. But it is a sticky process. It also politically has been an issue, whether that is done on a high-handed thou must basis or on a more negotiated, more friendly basis. In the early days of PSR, it was on a fairly high-handed basis. What has been interesting to watch, if you think of CN as moving to version 2 of PSR – they have tried to move to a more customer centric and friendly approach and try to do this more through a cooperative negotiation. And I think there it's shown benefit there. So, I have hope that we will see a general pattern towards that approach going forward.
Pete Swan
Jennifer, this is Pete. Let me add two things on that as well. It used to be all the Class 1s, or most of them, would let customers order in specific cars into their plant. One of the things that has happened under PSR, basically railroads say, this is what we do, enjoy it or not, is that the railroads say if you order three cars, they will give you any three cars instead of one bleach, one boxcar, and one covered hopper. And that has been a problem for some shippers. Additionally, many companies are landlocked. So, if you use private cars and you have no free days, you have to take in all the cars that arrive when they arrive, which for some shippers means building more tracks. And if there is not room on the parcel they have to do that, it is problematic for both the shipper and the railroad. To the building cars to go deeper into the network. At least a couple of the short lines I talked to were approached by Class 1s to block cars to go deeper into the Class 1 network. And the response typically is okay, so you want me to perform more work to make your cost less, what is in it for me? And the typical answer is well, the cars will move faster and the response as well, I don't have the money to switch cars for you, so they will be what they are. Thank you.
Jennifer Symoun
Thank you. Carl, and of the question, does PSR consider any external elements such as interaction with at-grade crossings with roadways to avoid peak hour conflicts?
Carl Van Dyke
Probably no more so or less than it ever has in the past. It has the potential to do better on that. First, I would say most freight railroads do not spend a lot of time thinking about this issue except on a local basis where they know they are going to be stopping trains or switching activities as opposed to road trains. But if the trains are being run on a more disciplined basis, one would hope or think that they could do a better job of doing those kinds of things. I can't speak to what extent they actually are, but when they are running the trains on a more random basis, then the conflicts are probably more predictable. They will have them because they are not managing the issue at all in that sense. So, they have the potential to manage against those issues now. I can't speak to whether they really are or not.
Jennifer Symoun
All right, thank you. Another question for you, how is the implementation of PSR impacting relationships between Class 1 and short line railroads? For example, is the streamlining process as part of PSR resulting in opportunities for short lines to acquire divested branch lines?
Carl Van Dyke
I think Pete has spoken to the operational impacts, which are a definite point of tension as to how that is all going to work. There are certainly some cases, where people are probably most aware of CSX right now, where some additional short lines are being shed or created because of the initiatives related to PSR. I do not think of those really as PSR per se, but more in the drive to make the system leaner, which was going on in waves over the last several decades. But yes, there is a bit of a new wave at some railways related to their PSR initiatives.
Jennifer Symoun
All right, thank you. Another question, so Carl, this one was directed towards you, but I actually will pose it for Jeremiah or Pete as well. Do large-scale disruptions, such as the current coronavirus outbreak, impact the availability or demand of various products make it more difficult for railways to maintain the set schedule that is a feature of PSR?
Carl Van Dyke
So, I think we have to go back and look at the crash, the Great Recession. I mean, that certainly generated a period of time with rapidly declining traffic volumes and a need to restructure the plan very quickly to respond to that. The railways did fairly well in that process. Here again, they will respond, it is a plan. There is a group of people monitoring the traffic situation and readjusting the plan. And I believe that they will, unless we have an unbelievable drop in traffic, which I suppose is possible if this gets really bad, but assuming we are looking at 10 or 20% drop in traffic, I believe the railways within the context of PSR will be able to adjust their operating plans to reflect the adjusted volumes that are coming. Part of the problem with Corona versus prior events is it has a more unpredictable nature to it, as to the duration and recovery and so on, than a conventional recession would have, which tends to have a slow growth rate on the backend. I don't know. Very hard to predict what the patterns are going to be at this moment.
Pete Swan
So, this is Pete. I will add one thing. I think one likely effect of rejiggering the network, as Carl mentioned, because that will happen, is that if the traffic drops on certain segments below a certain level, you are going to see reductions in service. And so, there is probably going to be a new reality of what service is going to be for certain shippers. Thank you.
Jennifer Symoun
Jeremiah, anything you wanted to add?
Jeremiah Dirnberger
I think the definite potential. Probably what the person asking the question is seeing is, since your buffers are reduced you do not have the extra assets just sitting around ready to go. There is definitely a large-scale disruption; it could take more time to recover. I know CSX went through something like that pre-PSR, but it was more of just their network across the northern tier does not have the same number of large terminals as the NS does. So, when things happen, especially in Chicago, there is no other place to propagate that to the system to catch that and buffer that before it works its way all the way through the network. Interestingly enough right now, PSR has really been rolled out at the same time as volumes have declined over the last couple of years. So, all things being equal, if volumes stay the same, how would that look? It is still tough to tell. Is it because of PSR? Is it because volumes have gone down? Probably because of both that we are seeing through the operational situation that we are. And if traffic were to suddenly grow, how would that be absorbed? So, it is a little unclear at the moment because we have two things going on at the same time. We will have to see how things go as this situation progresses.
Carl Van Dyke
I will just make one other observation. The bigger threat to the rail industry right now would be if you had a widespread infection in the group of people who dispatched trains, for example, just like you would if you had it in the air traffic controllers, where then the railroad might lose its ability to really effectively operate. So, that is probably the bigger concern, at least in the near term on all this.
Jennifer Symoun
All right, thank you. Carl, another question for you. On slides 11 and 14, to what degree does the benefit reduce idle time and squeeze out idle time translate into reduced engine on time and thereby emissions?
Carl Van Dyke
Reduced yard air? I'm sorry. Say that last part again.
Jennifer Symoun
How does it translate into reduced engine on times and thereby emissions?
Carl Van Dyke
As in engine idling?
Jennifer Symoun
I believe so, although Abby, if you want to clarify, let's assume. Yes, Abby said yes.
Carl Van Dyke
Okay. I don't know. There are two different issues there. One is with improved locomotive utilization, and we are using a total number of locomotives has gone down on a big Class1 by 1000 or more locomotives, and so by definition you have reduced the total number of locomotives that are running either idle or active on a Class 1 to handle the traffic, which presumably would have some positive benefits from an ecological standpoint. In general, it should be less idle time for a locomotive, but that does not change the total amount of time the engine is on. Turning engines on and off has always been an issue of mechanical considerations and so on than anything else. I think the big impact is on having fewer locomotives, more so than any other real measure, if you looked at an individual locomotive.
Jennifer Symoun
All right. Another question for you about slide 14. Is the goal of flattening peaks run network as evenly as possible seven days per week possible on parts of the network where passenger and freight shared track?
Carl Van Dyke
I don't know that the two are really in conflict. So, we have long distance Amtrak and we have commuter service. In commuter service territory, in general, there is a negotiated relationship between the commuter agencies and the railway as to what each is going to have in terms of track access and hours of train operations and so on. And I don't see that those have changed dramatically. There are probably some issues in the Chicago area where there may be more potential for more conflict. But in general, those are managed, and they are relatively small geographic areas relative to the overall industry. And in many of the commuter territories, there is immaterial amount of freight traffic, so the conflicts are not that large. In terms of Amtrak identification, longer trains and so on, definitely will have an impact on how they interact with those trains, but those trains don't have that strong a time of day element to them to begin with, so it is a different set of issues than a peak/off-peak or 24-hour type of issue that is at play.
Jennifer Symoun
All right, thank you. Let's see, we have a question, I believe this one is for Jeremiah that came in during your presentation, but if anybody else wants to answer, feel free. Is PSR presented to railroads with any upper limit to what it could achieve? Is it given as "the sky is the limit" or do the people pushing the concept realize there are upper limits? For example, longer trains break apart more often and that steps up the entire subdivision what happens in interlocking. Will PSR operators cap train lengths to try to prevent operational issues or is that always blamed on the operating crews not being able to handle an impossible train? So, it's a long question.
Jeremiah Dirnberger
Yeah. I can give an interesting anecdote from when I was working in the Class 1 industry. I was at CSX before the PSR changes came into being. And one of the things that the company did seeing that they needed some changes, was they really just said we are going to run longer trains; like that is the ticket, that is answer. And it was one part of PSR, but it was not really thought of in a sense how Carl explained it, which is really the true PSR principle, which is general-purpose. All trains are more general-purpose; multiple networks, multiple traffic types operating on the same network, combine them to get that longer train. So, we are doing some very extreme stuff to run these two mile or even longer, 18,000 foot trains thinking that that was going to get that kind of asset utilization, but it caused a lot more problems if the network is not designed for that. I think now the advocates and those implementing PSR, the practitioners, they understand that, ultimately, your infrastructure on your network is going to dictate what you can do. Whether it is your passing siding lengths, that's why CN has been investing a lot across their lines to go ahead and trying to extend, whether they are double tracking or extending the sidings out, to 10,000 feet or greater. Or in your yards themselves, because you have got a certain type of infrastructure that dictates what your yard is going to do. So, I think that has been taken into account. I know some of the latest implementations that have been done, the softer, gentler PSR have been more thoughtful of that kind of frontline; let's go to the locations themselves and talk with the yardmasters, terminal superintendents, the front-line people, and figure out what is the operation like there, and then design a playbook that follows our headquarters' mandated principles of practices, but the playbook fits the area they are in. So, I guess in short, maybe Wall Street thinks the sky is the limit, but I think in reality, the railroaders realize that there are certain limits to what they can do and perhaps to go beyond to the next level technology and digital solutions will be the way to go.
Jennifer Symoun
Thank you.
Pete Swan
I will add a little bit to that. So, there has been a learning curve here as well. Early on, CSX I remember derailed a couple trains on the connection track between the C&OO and the B&O at Fostoria, OH. And what they did is they put a center beam flatcar that was empty very close to the engine and the engine basically pulled it off on the inside of the curve. NS has done this a couple times on a horseshoe curve and CSX did it once in Baltimore, too. They have learned to use distributed power, putting locomotives in the middle of the train, in the rear of the train to control the force is better, and also not to put center beam flat cars on the front of a train. And this has been a lesson they have had to learn from derailments. But, by and large, now those problems seem to have gone away. I will say that there are limitations on train lengths; passing siding is one. The other is building the train at origin and breaking it apart in destination. You know, the longer the train, the longer it takes you to put it together, the longer it takes you to do an air test. And if it is wintertime, you simply can't do an air test on a train that is too long, you will not have enough pressure on the rear end. So, there are physical limitations that the railroads are aware of, but they are also very adept at adjusting to what they need to adjust to in order to maximize the cars per train.
Jennifer Symoun
Thank you. We are running out of time, so I will skip to another question here for Jeremiah. On slide 8, does wait to start building a train until the time dictated by the plan allow for more time on yard air versus onboard air? Is it predicated at all on the proposed elongation of the Class 1 brake testing interval from 4 to 24 hours?
Jeremiah Dirnberger
I would say, in actuality, what it basically is doing is they are trying to maximize the amount of time you can be building that train. And the way it has been explained to me is, if you go past that point, mechanical is going to put additional resources there to make sure that trains to get there on time. So, it probably reduces the time on yard air or at least tries to keep it to a minimum. Whether or not it is dictated by that, I really don't think that is the case, but I think it is primarily driven by the idea of wait as long as possible so that as many cars have the highest probability of making their connection. And then, from that point on, everybody in the entire yard works to get the train out on time versus before, a lot of times, we had mechanical optimizing and doing their own thing, you had transportation focusing and doing their own thing. Whereas now, basically all of that reports up and everything, to get a train and turn the assets, flows up and one person is ultimately responsible for that work.
Carl Van Dyke
I would just say in the past, railroads had a bad habit of focusing on getting the train out on time where they would start creating the train early to make sure it was ready, but you do, then you are basically running away from some traffic that had not gone over the hump yet. And by not building it before the right time, you are maximizing that the inbound cars are going to get on the right outbound train. And that is the focus, not really about air. The cars are not on air anyway (on the classification tracks), only in the departure yards they might go on yard air.
Jennifer Symoun
Thank you. There is one last question I want to put for all presenters, and if you want to just give a quick response and we will go ahead and close out. I think we have addressed the majority; I know we have missed a few. But if you did not get a response to your question and you are interested in getting one, let me know and I get that from the presenters. So, one more question, moving forward, what do you think are the most significant impacts PSR will have on freight and passenger rail integration in urban areas? So, Carl, let's start with you.
Carl Van Dyke
I think that there is going to be a conflict there at all times, and that the focus of the railways is on their freight business. Their profits are driven by the freight business. PSR somewhat hardens them in terms of really focusing in on that reality, so that you are going to have to look and negotiate very hard when you are trying to have passengers on the same track as freight in terms of what you are going to have to do to the infrastructure and other things to ensure that both the freight and the passenger can operate in minimum conflict with each other. That conflict has always existed. I think PSR may bring it more sharply into focus.
Jennifer Symoun
All right. Jeremiah?
Jeremiah Dirnberger
Yeah. I would definitely agree with what Carl said. I mean, you have got a completely separate network that is wanting to share the same resources as the network that PSR is trying to standardize and streamline and combine traffic types into general-purpose trains. So, unless the passenger business is willing to let their cars move in those general-purpose trains, you are probably going to have a much more heightened stress between the two. Unless there are situations where a railroad determines there is a redundant line they don't need now, and that could get a commuter agency or some other entity that wants to do passenger would take that over, some of those opportunities may present themselves now. But in the situations where we are still sharing, you may have that. But that may be balanced by the fact that the freight operations should be more consistent and reliable. So, you may not have as many detrimental impacts to the passenger service because the freight trains are trying to be run much more reliably. So, we will see a little bit of both.
Jennifer Symoun
All right, thanks, and Pete?
Pete Swan
That is a tough question. I agree with everything that has been said. I will add a couple more comments. So, when you say urban freight, one of the things that comes to mind is actually delivering cars within the city as opposed to running trains through Chicago. I think under PSR, there will be a tendency to de-market that freight. To the extent that freight railroads do not need a line to get to somewhere else, they would probably be more than happy to spin it off to the urban agency or a short line or something, and run cars to a transload center or get it to a short line just to get out of it. The other thing I will mention is in the effort in Chicago for the Create program to basically separate freight and passenger wherever possible is probably the best answer for solving this age-old problem to give them separate right of ways. And I think there is going to be more and more reliance on that in the future, that whereby the passenger will have its line and the freight will have its, and they will not get in each other's way. Thank you.
Jennifer Symoun
Thank you. We are a little over time. I know there are a few questions we did not get to. If the presenters do have time to stay on and want to type a response, feel free to or I could just follow up with you after and get some responses if people are interested.
With that we are going to go ahead and close out for today. I do want to thank all three of our presenters and thank you everybody for attending today's seminar. I will be sending out a link of the recording along with the presentations by tomorrow. The April Talking Freight is not yet available for registration but once it is I will send notice through the Freight Planning LISTSERV. The Freight Planning LISTSERV is the primary means of sharing information about upcoming seminars. I also encourage you to join the LISTSERV if you have not already done so. With that, everybody, take care, be healthy, and enjoy the rest of your day.
