View the June 16, 2021 seminar recording
Jennifer Symoun
Good afternoon or good morning. Welcome to the Talking Freight Seminar Series. My name is Jennifer Symoun and I will moderate today's seminar. Today's topic is Using Freight Mobility Tools to Measure Truck Flow Efficiency.
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. You can log your credits after the webinar on the AICP CM web site. The event # is in the chat box.
Certificates of participation are also available for Talking Freight seminars. These certificates may be used for 1.5 professional development hours if accepted by your licensing agency. To receive a certificate, 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 Jeff Purdy, a Transportation Specialist in the
Federal Highway Administration Office of Freight Management and Operations. Jeff's responsibilities include freight performance measures, freight highway operations and bottlenecks, freight demand modeling and data, truck parking, emergency routing, bridge strikes, and connected/autonomous vehicles. Prior to moving to FHWA Headquarters, he served as the Technical Services Team Leader and Transportation Planner for the FHWA Wyoming Division, overseeing statewide and metropolitan transportation planning, air quality, non-motorized transportation, local public agency programs, right-of-way, and research.
Jeff Purdy
Okay, thank you, Jennifer. What I would like to do is get sort of an overview of this Freight Mobility Trends analysis tool that we have been working with recently here at Federal Highways with the MPM-RDS. Let me walk through a little bit of background of why we are looking at nationwide freight mobility trends. As you can see from these slides, these are some of the trends that we see from things such as the freight analysis framework and other projections being done. We expect to see a pretty substantial increase in freight movement, particularly truck movement on the National Highway System. Goods movement by truck represents the large majority of the total tonnage that's being moved around the country domestically in terms of freight transportation. That is projected to grow pretty substantially and so this truck traffic obviously is mixing with other traffic on the interstate system, on the NHS, in our urban areas, which is contributing to increased congestion on our National Highway System. So, we wanted to develop some tools that we could use at the national level to try to monitor freight trends to understand how things are being affected at a macro scale and where potential bottlenecks are in the systems and how the system is performing from year-to-year.
So, we have the National Performance Management Research Data Set, or the NPMRDS, which is a very powerful tool for Federal Highways and states to use to be able to look at mobility on the NHS and look at performance measures and performance-based transportation planning and programming. And there is a lot of detail with the NPMRDS, but we wanted to be able to take that data and to be able to analyze it on the national level. So, we developed a tool where we can process the NPMRDS data set at the national level, which is a very large data set, we are talking half a terabyte per year of data. We developed a system that we could process that data and develop a nationwide network using some mobility indicators for the National Highway System at a national scale. So, what we have done is gone through and combined a lot of the TMCs to cut the number of segments down to about a quarter of what you would have on the NPMRDS. That allows us to work with this at a large scale or at the national level. And what we have done is we have developed a series of mobility indicators that we use, in terms of truck hours of delay per mile, a travel time index that looks at peak period flow, planning time index that looks at 95th percentile travel time, a reliability index which is similar to the national performance measure for travel time reliability that compares 95th percentile to 50th percentile travel time, a buffer index, and also a cost of congestion. These are just a few of our major measures that we have that we are using to look at things at a national level, and we continue to refine and further develop this tool.
So, this is sort of a shot of what the tool looks like for the national level map. This particular map is showing delay per mile on the interstate system. And, as you can see, the states that are colored in orange are the states that have the higher delay per mile. But we can look at it for these different measures. We can also look at it at urban versus rural areas, we can look at it on the interstate, we can look at it on the entire NHS. For example, this right here is looking at the travel time index, which is a measure of the peak period traffic flow. We can also look at things such as this map right here which shows reliability. What we have done on this particular map is filtered it for just the rural NHS areas. So, this is showing the reliability on the rural sections of the NHS by state. Then we have this pop-up on the right hand side of the slide that shows the State of Maine. You can look at the data either annually or quarterly. In this particular instance, we are looking at it quarterly. You can see the seasonal fluctuations where you can see the reliability gets worse during the winter months and better during the summer months. So, it is an interesting tool to be able to look at a lot of these trends nationwide.
Some of the things that we are seeing. What we have done, we have now processed the 2017 through 2020 data. But we have done some analysis and reports rooking at the initial trends from 2017 to 2019. And we saw that things were somewhat stable, but the delay was worsening slightly between 2017 and 2019, but there seemed to be some improvement in performance in 2019. Looking at it from a trend standpoint, this shows the delay per mile on the NHS by functional classification with the orange being the interstate, the blue being freeways, and the gray being other NHS routes. So, we saw that the interstate system overall is where most of the delay is, which is not much of a surprise because our measure of delay is weighted by volume and that is where a lot of the truck volume is. But then, when we look at going off the interstate system, particularly when you go to NHS arterials, we are seeing that while there is less delay on the arterials because of lower volumes, the arterials often tend to be less reliable than the interstate system, particularly when you get to a local arterial roadway that maybe has a lot of intersections and driveways. But we can also look at this in terms of rule versus urban, where we see there is a lot of delay particularly on the urban interstate system and the higher level roadways in the urban areas, but we also see there is some decline in terms of performance year-over-year in rural areas, particularly when it comes to reliability.
We are also, in addition to looking at it on a state by state basis, we can look at urbanized areas and how they perform. We can look at MPO areas and how they perform. We also have another dashboard that is part of this tool that looks at bottlenecks. You can either look at just the interstate system, or you can look at the entire NHS. You can look at delay per mile which is one of the measures we use for bottlenecks, but you could also look at what locations around the country have the worst reliability. So, you can filter it by different measures. And we used this to develop a new list that we put out this past year of major interstate bottlenecks. This is the list from 2019; it is on our website if you want to go take a look at this. But we plan to track these locations from year-to-year and in addition to other locations using this system to be able to see how different locations where performance improves or gets worse due to congestion or where roadway improvements improve performance at certain bottleneck locations. So, with our bottleneck tool, this is sort of zooming in on a particular state. Here, this is the state of Maryland, and you can see individual interstate routes and the darker red areas are the locations that have the higher delay per mile. We can take this data and assemble it at a regional level and look at, for example, here we have the I-95 corridor that goes from Massachusetts down to Virginia, and this particular map shows all of the major bottleneck locations along that northeast corridor between Massachusetts and Virginia. So, you can use this for regional planning purposes to see how one bottleneck location in a particular state may fit in in terms of the amount of delay compared to other nearby states.
We also in this tool were able to track performance at different intermodal facilities such as freight airports, major rail intermodal facilities. We were also tracking the top 25 ports around the country in terms of tonnage, as well as delay at border crossings. And so, what we are able to do is with all these different locations is to look at what the long term trends are in terms of access to these intermodal facilities. We hope this will be a valuable tool for us to use here at U.S.D.O.T., but also for states and MPOs to use in terms of being able to review national trends in terms of performance of the NHS and the interstate system and to look at developing different types of strategies that can be used to try to address congestion at bottlenecks or improve reliability of the system to things such as transportation system management and operations. To be able to look at things at a larger more macro scale is really what our intent was with this particular tool. Here at the bottom of the slide the link to the web page where we posted the dashboard for this tool, and we continue to make updates and refinements to it and add additional data.
At this point what I would like to do is maybe share a quick demo if I can share my screen. Okay, you should be seeing a map of the United States right now.
Jennifer Symoun
Yes.
Jeff Purdy
This particular map is showing 2020 annual data delay per mile. We can also select (it takes a minute to process) and we can see what the planning time index is. This is for all areas. What we could do is say okay, what is the planning time index by state for urban NHS routes? Or we can filter by functional classification and say let's just look at planning time index for urban interstate routes so you can see how that is performing. We can also filter by the primary highway freight system or by the Strategic Highway Network. You can also select an individual state and then go into the individual state here. This is Pennsylvania and it is showing by urbanized area the delay per mile. You can also look at the total congestion cost, which obviously here Philadelphia is going to come out as the highest since it has the most mileage of NHS of the urban areas and also probably the most traffic, so the highest congestion cost, which is why we typically use some of these other measures which are normalized by the length of the system. And you can look at urban areas or you can go in and look at the MPO areas and again you can filter by do we want to look at the entire NHS, or do we want to just look at the interstate system? And then because we are looking at MPOs, we can filter for urban or rural areas of the MPOs that are outside the urbanized boundaries.
I also mentioned that we have another dashboard associated with this which is the bottleneck map here. What this is showing is delay per mile on the entire interstate. You can do the whole NHS or just the interstate or NHS routes that are off the interstate. You could, again, do the primary highway freight system, the STRAHNET. You can filter by time of day if you want to look at the entire day or if you just want to look at peak periods, you can do that. In addition to looking at delay per mile you can look at planning time index, buffer index, truck reliability, congestion costs, all of these to develop a list of bottlenecks to be able to see how locations are performing.
Then, in addition to looking at the national scale, (sometimes it takes a little while to process) here is my home state of Michigan. You can see here we have the interstate system, delay per mile. We can sort of zoom in here on southeastern Michigan. You can see the red locations are where we have the highest delay per mile. Then down here we have I-94 going through downtown Detroit as the number one location in terms of delay per mile in the state. I-94 is obviously a major freight corridor. To the west on I-94 if you go at least to Chicago, so between Detroit and Chicago we have a lot of freight movements on I-94, but also heading east takes you up here to Port Huron where we have the Blue Water Bridge crossing into Canada. You can see here we have some information in terms of the delay there. But you can see some other locations showing up with delays. We've got I-75 up in sort of the Troy area in Oakland County, I-696 over in Farmington Hills/Novi area. So, you can see these different locations and how they perform.
One thing to note with this information is it is bidirectional, so you can look at northbound or southbound, eastbound or westbound directions to be able to show the information for the different directions. In addition, the bottleneck tool allows you to look at the intermodal facilities that I mentioned and what we do is, in terms of the top amount of cargo movement for airports, freight airports, border areas, intermodal facilities and ports, you are able to go in and look at individual port locations. Like here you could go to the port of Oakland (we will give it a second here to process). Let me go back. Sometimes you have to reset the location. All right, we'll look at an airport. One thing that is interesting about this page here is that what it is showing is there are statistics on these top airports here at the top of the screen that you can rank this by the different performance measures. We have, I believe this is ranking on delay per mile with the area around the Cincinnati International Airport being the highest in this particular measure. But then you can also look at routes that are within five miles of these different airports and you can see how they rank on a national level. For example, we have New York; I-678 is a critical freight route that heads to JFK airport, so we have a lot of delays in that location. And then Alan, the next speaker from NYDOT, will maybe talk a little about that area. But another area with major delay is the Chicago area. I am sorry, it is not cooperating. But you are able to zoom in and be able to get information, when this is working properly, for that airport.
So, that is sort of an overview. Again, it is a tool that we recently developed, and we continue to make refinements to it. And, you know, I would be happy to answer any questions that anybody has at this point.
Jennifer Symoun
All right, well, thank you Jeff. Again, if you do have any question, please type them into the chat pod and we will take the questions after all three presentations. So, our next presentation will be given by Alan Warde, a career employee of the New York State Department of Transportation. Alan has been working with performance management and asset management issues for highway and transit for over 20 years. Prior to this he did oversight work for NYSDOT for its involvement in transit issues of the Metropolitan Transportation Authority, the provider of transit services in the NYC region. So, Alan, you can go ahead when you are ready.
Alan Warde
Okay, thank you, Jennifer. The presentation I am going to give today is going to be a brief rundown of the tool that we used and developed with a university partner of hours. We are working with the NPMRDS data, the National Performance Management Research Data Set, hence the NPMRDS acronym is better than that title. Anyway, to start off with a little background. FHWA began providing NPMRDS data back around 2014. This is data which includes travel time and speeds for highway segments in five minute intervals for an entire day, entire week. We recognized early on that the NPMRDS data was extremely valuable, but we also found that we did not have either the hardware capability or, quite frankly, the software capability and skills to adequately download and manipulate the data. We had an existing research contract with the State University of New York and their Visualization and Informatics Lab, which I will call AVAIL, and that provided an opportunity to develop a user-friendly NPMRDS tool kit. This is an open source tool kit. Let's skip to the next slide here.
The tool kit itself was developed with a lot of input from New York State's 14 MPOs. We held monthly meetings from the start with NYSMPO's modeling working group. NYSMPO is the MPO organization for all of New York. Both the MPOs and we at DOT have had some interesting uses for the tool. The larger MPOs have been using the tool to inform their congestion management plans. It was used by us at DOT to develop the freight bottleneck list for our freight plan. We have also used it on occasion to provide speed data needed for several highway design tasks.
The tool basically has several different tiers. The top level is what we refer to as the Macro tool and that allows snapshots of the County MPO jurisdiction and statewide geographies for the state. As you can see here, it has a very broad set of measures which can be displayed via a map. I will give a little demo of that later on. In addition to the Macro view, we also are able to do route-specific analyses. The tool allows us to do on-the-fly route creation, so we can pick the routes that make up a corridor or better applicable to a particular project that we are interested in. There is also a port generator that allows us to display the data in multiple graphic forms which I will demonstrate. It also gives us a broad range of metrics and allows us to do "what if" comparisons for the routes that we choose. We have also performed a conflation with the NPMRDS data and our own linear referencing system database so that we can integrate speeds and other data for analysis purposes. This opens up the NPMRDS to being joined with other data on traffic volumes, classification counts. We can even have data on the construction of each road segment that can be applied to the speed data.
All of the data is downloadable. The map data is downloadable as either CSV file or Shapefile. The graphic data, which I will demonstrate, is downloadable as either CSV file or as a PNG file so we can take the graphic itself and load it into a report of our choice. Our tool is also, I think, somewhat unique from the other NPMRDS analytic tools that are out there in that it provides us information on the quality and number of sample data that underlies the calculations and metrics that we are using for our analysis. This data is available in the report graphics. We have a special function in the back row tool which gives us a deep dive into an individual highway segment, and it is also available as a hover text in the Macro tool.
Our tool also is what we have been using to calculate the required federal PM3 reliability measures: the level of travel time reliability, truck travel time reliability, persons hours of excessive delay. We can do this both on the required statewide basis as well as we have disaggregated the calculations down to the MPO and even down to the County level. As far as near term things that we are interested in adding to the tool, we are looking very heavily at incorporating business location data and we are also looking at integrating freight commodity flow data such as from TransSearch as was done with the FHWA tool that Jeff demonstrated. We are also very open to other areas of possibly looking at the air pollution generated by the traffic indicated by the route segments.
At this point, I would like to do a quick demonstration. So, we will see here, I'll try sharing my screen and, okay. Jennifer, are you seeing a map here?
Jennifer Symoun
Yes, I am.
Alan Warde
Okay. What I am going to start with here is a quick demonstration of our route creation function and we will also go into a little bit of the report generation function. We can generate routes with a point and click method. And just get a little bit of background, what you are looking at here is the Mohawk River which is the border between the northern part of Schenectady County which is down here and Saratoga County which is up here. In 2017 we replaced a bridge across here that crosses at Rexford. And that bridge had capacity problems along with having state of repair issues. So, the new bridge was built parallel to the old one and then in addition there was a roundabout that was created, and work was done on the approaches also. So, to create a route for the analysis, what we do is we find where we want to start and then where we want to finish, and this gives us a list of the TMCs that were chosen, and we simply save the route. And by saving it as a group, it is available to all the registered users at the Department of Transportation. And you press the magic button, and we are done; we have saved this route.
Now, the next thing I am going to do is show you how we go about creating reports. What we have done here, I have a dropdown and I selected the route for the Rexford Bridge and in creating the report we have a number of things that can vary. In this case, I am going to look at the full time period from when we first started getting the data which was January of 2016 all the way up to the end of 2020. We can choose the time of day that we want to have the analysis focus on. Since we have five minute resolutions, if we wanted to, we could pick a single five minute period. The resolution as I mentioned, the basis is five minutes, but we have aggregated it all the way up to a year. For this particular analysis, this demonstration, I have picked just the daily average speed. What we see here is that at this point right here is when the new Rexford bridge was opened. And you can see that there is a fairly dramatic difference in the average daily speeds before the new bridge was opened. You can also see, as I mentioned, the bridge was built parallel to the older bridge with the attempt to try and minimize work zone issues. There were some, obviously, but for the most part the traffic flowed as it usually had at that point. It was subject to some very severe peak congestion. But, nevertheless, we see that since the new bridge was installed, the speeds have been dramatically faster than before.
To give you another idea of just what our reporting capabilities are, as Jeff mentioned, 678 which is more commonly known as the Van Wyck Expressway, connects the JFK airport in New York City to the Long Island Expressway, which is the only interstate that goes from New York City east to Long Island. And it also connects to the Clearview, which you can't really see on this map which is another east-west limited access highway that serves Queens and goes on into Long Island. This is a map which shows where the boundaries are of each of the segments that we have chosen. This first graphic is a speed profile for 2019 and it shows the averages for each five minute time period during the full 24-hours of the day. As you can see, the overnight period up until about 5:00 or 6:00 in the morning traffic is basically at free flow. However, a quarter after six in the morning those speeds drop down to, I think, as low as maybe even like 6 miles per hour in some cases. But certainly, well below the 50-mile per hour speed limit. Moving on, we can see that the impact of the COVID pandemic. Here we are doing the same calculation, only it is for 2020 rather than just for 2019. And these are truck-only speeds. The vehicle traffic would be slightly faster. But, again, we are seeing that there is an improvement in the daytime speeds for the bridge.
This graphic here shows the difference between those two graphs. So, we can see there is generally a 10 to 15 mile increase during the daylight hours for the Van Wyck. These three graphics here show our calculation of the truck delay that occurred. This is an annual calculation, but it is showing again for each five minute segment, just how much delay can be attributed for the entire year to that segment. You can see that it really gets quite congested in the peak areas. Again, with the pandemic, traffic volumes fell, and we have a much, much lower amount of congestion. Again, the difference graph shows us the extent that the 2020 delay was less than what we saw in the pre-pandemic. This next graphic here is basically a heat map. We have each TMC or highway segment is plotted along this axis here and we have the time of day down here. And in this case, we are showing where the delay improved along the course of the Van Wyck. As you can see in the map we have down here, when I highlight a highway segment in the heat map, it is also reflected in the more graphic map below.
So, that is basically our demonstration for the reporting tool. Lastly, what I want to demonstrate is our Macro tool. What I have pulled up here is a map of Westchester County which is just north of New York City; this little sliver down here is the Bronx. But this is showing us all of the roads that are within the NPMRDS. I should have mentioned this earlier, but we bought into an expansion pack for the NPMRDS data, so this is including roads other than the NHS in this. We have the ability to call on different types of routing, so we can highlight just the NHS if we so desire, or just the interstates. But in this case, we are looking at everything. What we are seeing is a plot of the truck travel time and reliability measure with just the a.m. peak. And the circles that are being shown are bottleneck locations. What that means is that these have the highest ranking TTTR score for the County in this case. What is somewhat interesting is you can see along the side here we have a listing of those bottlenecks, and we can see that the scores are generally quite high. You can also see here, as I mentioned before, we have a measure of the data quality that is behind each one of these areas. And what that is, it is the percentage of the time periods that are actually reporting data during the day. So, in effect, this number one indicated bottleneck is only being supported by 1% of the data. So, for a morning peak there would be 36 five minute periods. So, this is just a really tiny fraction of that. We have the capability of going in and setting a threshold for the percentage of time periods reporting. What I am doing here is saying that we want to exclude everything that has less than a 30% reporting percentage. And, as you can see, we get a different set of bottlenecks, but they all have much, much higher rates of reporting than what we were seeing before. So, this is the tool that we are increasingly using to make sure that the bottlenecks that we identify have a robust level of data informing them. So, with that, this is the end of our demonstration and I look forward to people's questions.
Jennifer Symoun
All right. Thank you, Alan. Our final presentation will be given by L.D. White, a research specialist in the Transportation Planning Program at the Texas A&M Transportation Institute (TTI). He has over 20 years of experience as a researcher at TTI, involving a wide range of topics from air quality and emissions modeling to transportation planning. Most recently, his primary focus has been on the development of interactive data visualizations and analysis tools to assist in freight, transportation planning, and data driven decision making. So, go ahead.
L.D. White
Thank you and thank you for the opportunity to present today and talk about the truck congestion and analysis tools that we at TTI have worked with the Texas Department of Transportation to build. It is strictly for the state of Texas, but right now we have other efforts going on that expand the tool to do other things, but this one here is strictly for Texas. It is totally an open source software coding which makes it no license, no registration, totally customizable. There are no license fees and at any time we can actually transmit this to a sponsor to be hosted on any infrastructure they would choose I would have it hosted on. Currently in our development, since we are under development and building here at TTI, we have it hosted in the cloud for greater access and distribution. What TCAT does is it provides user access from the underlying data of roadway efforts between TTI and TXDOT. That effort produces officially what is a database of Texas roadways greater with a class above 7. So, mainly no local roadways. It is based on the RHiNo traffic volumes, which is Texas' version of HPMS. So it's based on HPMS data that for the most recent year we have in TCAT now is over 200,000 road segments and over 100,000 centerline miles. The volume data from HPMS is combined with the INRIX speed data to get into the congestion estimation and the mobility performance measure estimation.
The primary tool, while that effort produces things all the way across vehicles and trucks, the primary focus on this tool is annual truck performance. So, everything that is in TCAT is annually based. And while we do have things that are for the all-vehicles, it is mainly for a reference point. This tool in itself is centered primarily on trucks. And throughout it you will see many of the traditional mobility measures that you have heard about already, delay, delay per mile, congestion cost, and others. Everything is stored at the individual HPMS road segment. So, there is very, very small detail available that can be aggregated all the way up to regional level, but it starts and is stored in the database at the smallest level we can get. And that provides us with the ability to be able to create custom corridors which can be done on the fly, which is one of the main key aspects we have with this. And with that, being able to look at that all the way down from the individual ACMS segment, being able to generate large region summaries really makes it very flexible. And because we have that relationship already connected with HPMS, anything that is connected to HPMS already that comes in with the HPMS data set or we have the RHiNo data set or we can connect to it in another way, so that means any other context type data set, any other geographic data sets we can attach to it, or we can show visibly at the same time, we can get those in to provide other visualizations and other context data.
So, one of the key things with TCAT is being able to create the user created corridors on the fly. So as with some of the other tools, you do have the ability to go in and create real time any area of interest, any corridor of interest that you have as you go. The road segment data is then aggregated across these multiple if you could extend RHiNo HPMS segments. And we also have for different years now of data the within TCAT to be able to look at trends, even on the custom corridors, visualizations, mapping visualizations and things are generated automatically so when you expand and look at the visualization you can automatically see the mapping trends as some of the measures.
So, one of the things because of the size of the state of Texas and the sheer number of road segments in a exist in Texas, one of the things we did is break it out into sort of a logical and already predetermined geographies. We wanted users to be able to access data quickly and we knew that we had some geographies and regions that were already of interest such as TXDOT districts, different counties, MPO regions, just to facilitate the building of summaries and isolation of corridors quicker and easier. Once you get into TCAT, one of the summaries has an option to these RHiNo-based segments have the option to create different levels including the highway system, function class, rural/urban just with a click of a button and really quick access to these summaries.
One of the other layers that we include in this is actually the Texas top 100 congestion roadways themselves. It is built as part of the other project. This is kind of just another reporting mechanism to have it in there where we can look at the top 100 truck congestion segments and look at trends and have ways to pull in the data to visualize alongside the larger data set. The monitored growth segments in the congestion roadways covers about 1,800 segments and a little over 10,000 centerline miles from all urban areas in Texas. We have different layers for each of the top 100 truck segments, the top overall congested road segments for overall vehicles for reference, and there are summary stats and we also have ranking trends going back to 2014, so you can pull these in and see how they have changed over time.
One of the other aspects that we focused on a lot with TCAT is getting some contacts and some other layers in to be able to provide some information as to why congestion might exist in an area and what we could possibly be doing about that congestion. What we have done is we have actually linked into the TxDOT using an API that is publicly available, called the TxDOT Data Portal, and they have within that access to their unified transportation plan. So, we are able to have the user specify which category of projects, whether they are beginning now, beginning later, five years, within four years, five to ten years, and then some more outward timeframes, to load in layers where you can see okay, here is our congestion stat laid on top of that with, okay, here is where we are having the highway roadway project. Getting those two at the same time in a downloadable map in a way where you can see them at the same time, visualize them, and be able to present them and download them.
In addition to the transportation projects, we also have different layers just for informational purposes. There is not a ton of data with these because we are just trying to get the idea of things that might exist with things such as truck trip generators, freight infrastructure, our energy sector in the middle of the Odessa area which I am hoping to do the demonstration here in a little bit about that area. There is a lot of oil activity, so being able to get things going like okay, here are things that could be going on in that area; things with ports, intermodal facilities, cargo airports, and different geographic locations of things that exist that could be contributing.
As part of some of the region summary aspect of TCAT what we have set up is a dashboard of annual truck congestion report cards. These are all pre-populated so that the extent of the user to be able to go ahead and download these and visualize these and see these is merely selecting a region or area. These are all pre-generated, pre-populated, and can be downloaded as PDFs for printing purposes, as PNGs to be inserted into reports. Just for this one here, this is the H-G area report card providing you a quick snapshot of over time what is happening and how is the area doing in terms of the mobility measures and congestion measures. They are available for every region and geography that's included in TCAT, and we have some others where we have identified areas and corridors such as interstates of interest. As we identify more, we are able to add these in because we have the vast databases behind this to do these regions at a greater level. These are handy because you can go ahead and with one click they are there and can be printed out and taken with you for wherever you may be for meetings or to give to others.
So, in the future, this is intended to be updated annually, the data behind it. So, every year we hope to have the yearly updates for the data behind it so there is the actual TCAT itself is updated and the report cards. Future things we discussed about possibly doing is adding some crash and safety information at some point, looking at things with some truck origin and destination information on the top 100 truck congested bottlenecks, and time of day and average trip length to try to provide some extra additional information as to what is going on on those bottlenecks that we know are problematic.
And with that I think if I can try to do a quick share my screen and see if I can make this work and not. Is it showing a map by any chance?
Jennifer Symoun
Yes, we are seeing the TCAT tool.
L.D. White
Excellent. When TCAT first launches it comes up with the Texas 100 monitored roadways, this is not the Texas 100 project, this is not the actual full RHiNo data set behind it. This is actually the top 100 bottlenecks and roadways monitored because of it. We have a box on the right hand side where you can get your information just from highlighting over a roadway. As you can see this gets pretty detailed around the Dallas area. Again, this is just the Texas 100, this is not the full database itself. But the user can go in and change the focus. Right now, the focus is showing the roadways, there is no focus on the roadways. The user has the option to go in and change the focus, so we can change the highlighting and pull up the top 100 ranking which you can go in and now I am looking at focus to the annual person-hours of trip delay. When you zoom back out you can find where Houston, Austin, we now have much more of a picture of where the higher congestion areas are. From there you can add that to the road segment table. The road segment table is where on these layers you get the individual roadway segment information and on the other layers that we have that uses the full data set, you have the option to create the custom corridors. On these, these are just the top 100 bottlenecks; once you get inside of these, you click on add segment and get to the bottom, and you can go in and ask it to expand the road and when you expand the road it automatically pops out with the ranking.
And this actually has both the truck ranking and overall ranking. You can turn these on or off. They are fully downloadable and removable. You have the option of going in and adding in different fields from different measures. And this is where we can get on this other tab, we can get onto the other map layers where this is the context and the project information, this is where these come in. So, we can go in and toggle this layer on and it will go in and it works with the TxDOT Open Data portal to pull these things in and all of those red squares are transportation projects occurring under way now or beginning soon in that area. Then as you scroll through and start adding more and more pop-ups and you can see there is more information on these. When these pull up, we can now access all kinds of information about what that project is.
So, that is for the 100 monitored roadways. Once we get into the others, we can get to the other database that has the HTMS RHiNo database that is much more extensive. We have different layers, one for each of the TxDOT districts. I will turn these layers off. So, there is actually one for each of the 25 TxDOT districts and we will just pull up Abilene, because I am thinking of pulling that one up. This is for just one TxDOT district. All of these different geographies and regions, one for an area where you can pull up stuff for individual County. Some of these counties are really small, some are really fat, some of them aren't, some are a little bigger. One for each of the MPOs, so you can go in and actually go and load in, there is Abilene MPO, and start the whole process similar to what we were with the others. The one thing I like to end the demo with though is with one of the regions. We have regions, corridors. Actually, before I do that, one other function we have in this is you can go in and search for individual roadways. If I am just interested in one of the interstates, you can go in and select, for example I-10 and it will show me the information for just I-10 as it goes across. And much like the other stats, we have different years. You can select four different years. We have the same ability to change the focus of the roadways.
For demonstrating the summaries and the route selections though, I am going to go to the Permian Basin region. The Permian Basin region has a lot of oil and gas drilling activity and has been since I grew up there. But over the last few years it has really become a hot bed of activity. So, what we can do by starting off in that region, right now you are going to get by hovering over things the information on each roadway segment as you highlight other it. This is actually a little over 27,000 roadway segments. I had these numbers a second ago. It is around 20,000 Centerline miles and between 42 and 43,000 lane miles of highway that are covered in this one area. So, if you go in and change the focus instead of being all the roadways, once you do that it is going to start to give you an idea of where the highest value for the annual truck delay. And what becomes surprising when you start to do this is this over here in the right hand corner is actually a metropolitan area. Midland and Odessa City areas are urban areas you will expect that. But this red mark right here is actually fairly rural. And that is where it is actually a considerable amount of oil and gas drilling activity going on in that area and it has been a focus of many efforts to look into this. So, by changing that focus, it automatically pops out and you can see it.
So, if we wanted to look at something along that roadway, the user can go in and select that segment and it will come up with a pop-up window and it will give you two different options, to add a segment itself or to create a corridor. Once you collect the segments for that corridor and highlight things in red where you continue to build your corridor, once you click two here, it will come in with another section of the road segment table similar to the other. The difference of it being is it is not building graphics in terms of ranking, it is building graphics in terms of what is going on in the last four years in that area. You can turn on all the different metrics and get different graphics that you can download, and you can see that there is definitely something going on there and it is a lot due to a truck activity. Much with the other road segment table you can go in and toggle off many of the different metrics.
The summary table on this one, though, has many more different functions instead of just providing route I.D. The summary you can go in and build summaries for the entire area, not just the corridor. Right now we have different classifications you can go in and redistribute things. Everything is aggregated and you can download this, you can go down and you get a nice little pop-up to change whatever metrics you want the look at. We also have a custom summary that is a little more advanced and operates more like an excel pivot table so you can build things how you would like. So, for this one since we have that area, we can go in and change it to County and a map with all of the counties instead of just by County total, we want to look at it by functional system. Right now, this is just number of links. We can change it to rural and urban and number of road segments, which really doesn't tell you a whole lot, but if you change the sum, we can go in and we can go something like summing the first hour of delay, and then you have that broken out by a large area. And much with the other, this tab stays fairly consistent. It doesn't change regardless of what layer you are on, so you can start clicking and see what transportation projects are going on in that general area and you can see as you go along, you can see what is going on and that we have different transportation projects going on in that area.
The other thing I wanted to show you really quickly is the truck congestion report cards. It is just a basic delivery method of getting these out there and as you go through, you can change the way they exist and all the data updates. For example, I-10, we have access to all the top 100 road segments. And, as these update, we can go in to generate a PDF, download the TTTN and save say them for the future. I believe that is it. Very quickly, that is it.
Jennifer Symoun
All right. Well, thank you, L.D. I know we have a number of questions that have come in, so we will go ahead and start off the Q&A session with those questions. And then if we have time remaining at the end, we can open up the phones, but I will go ahead and start by reading the questions that were typed in. Since you just ended L.D., I am going to go ahead and start with you. The first question for you is how does TCAT differ from the NPMRDS database problem overseen by Rutgers University? It seems like it covers a lot of the same performance measures.
L.D. White
I am not 100% confident in the Rutgers University. I am afraid to answer that question, honestly, because I don't know exactly what that tool looks like and what it does. But some of the ones that I have seen are mostly speed-based and we have that detailed volume worked into the calculations. That may be where the main difference is, but I honestly don't know, and I would hate to take too much of a leap into that question without knowing the answer to that.
Jennifer Symoun
Understandable. If there is anybody who happens to be on who might know more about the difference, feel free to type that in, as well. Another question for you. How will the O-D information be collected and compiled?
L.D. White
That would most likely be based on what we have right now and what we are thinking and can do would be INRIX information based off of their trajectory information. That is the way I have access to it at the moment. We have done work with those data sets prior, so that would be enriched later.
Jennifer Symoun
Alright, thank you. If you think of any more questions for L.D., feel free to type them in. I will go up to questions for Jeff now. Jeff, the first question for you is how current is the data in the FHWA dashboard? Is data from or for 2020 going to be treated differently?
Jeff Purdy
Well, right now what we have published online goes through 2020 and so we do see a major decline in delay and congestion, obviously, in 2020. In fact, I think I ran the numbers and nationwide on the interstate system the delay between 2019 and 2020 went down 20%, and for the top 100 bottlenecks, the delay went down about 45% between 2019 and 2020. So, we see obviously a sharp decline in delay. I mean, basically I think there might be some interesting insights that come out of the data once we complete our analysis of 2020 data in terms of, you know, if you have situations where you have a high level of people teleworking, the impact that that can have on the transportation system in terms of the reduction in delay, so I think we will be looking at those numbers and I think they will really provide some useful information in terms of the effects of commuting patterns and how drastic changes to commuting patterns can affect travel conditions for freight on the interstate.
Jennifer Symoun
Alright, thank you. Another question for you. How are you collecting the delay data? Is it mostly based on LDS such as INRIX and specifically international border crossing delay data?
Jeff Purdy
Okay. Well, in general in terms of delay, obviously we are using the travel time data from the NPMRDS which comes from the INRIX probe data for trucks. With our tool we are only looking at trucks. Basically, what we are doing is we are looking at the difference between actual travel time and reference speed. And our reference speed is based upon 85th percentile travel time during off-peak and overnight periods. We are looking at a typical truck volume distribution throughout the day, so we are able to look at 15 minute increments and then we are able to look at the differences between the travel time during that 15 minutes and the reference speed, and then multiply that by the truck volume to get the total truck hours of delay, and then we total that up for the entire year. And then to normalize nationwide we divide by the segment length to get a delay per mile for each segment.
With regard to the borders, the only data we have is the INRIX data which provides speed for the segment approaching the border. We do not have as part of this system the actual time that it takes for a truck to be processed and cross the border between the U.S. and Canada or the U.S. and Mexico. What we have is basically the truck speeds leading up to the border.
Jennifer Symoun
Alright, thank you. Somebody asked for the link to the FHWA freight dashboard. Is that something you can share in the chat pod?
Jeff Purdy
Yes. I will put that in the chat pod. The last slide on my PowerPoint presentation had the link to our web page. If you go to the web page there are links to the national dashboard, the bottleneck dashboard, and the corridor dashboard on that particular page, but I will put a direct link for the national dashboard into the chat box.
Jennifer Symoun
Alright, thank you. And then one more question for you. Can O-D information for only trucks be collected from NPMRDS?
Jeff Purdy
No, not from the NPMRDS per se, because basically what we have is we have truck speeds along each TMC segment along the roadway. Basically, all we know is how fast during any particular 5 or 15 minute time period of the day, how fast are trucks moving along a section of road. We don't have data on where those trucks came from or where they are going to, we just know how fast they are going on that section of road. And then from HPMS, we have the volume of trucks from that particular segment, but we don't have origins or destinations.
Jennifer Symoun
Alright, thank you. Alan, we will move on to some questions for you. Does the business location data identify truck types such as MD or HD or by class and/or also to business types such warehouse, logistics or retail, etc.?
Alan Warde
The business data sources that we have looked at so far do not identify truck type. At best they give you the location and an estimate of employees and you often have to take that employee estimate with grain of salt. So, no, it doesn't provide us with the truck types. I think as far as the second part of your question, the data does provide you with the NIX code and I think another system. So, you can filter out those codes that are representing industries that you are not specifically interested in. It is a challenge I think to do what you are asking to specifically find warehouses. It takes a fair amount of work to come up with a specific location because of the organization of the data sets that we have seen so far.
Audio cut off at this point.
