June 20, 2012
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 Pavement Damage Analysis Tool (PaveDAT) for Overweight Truck Permit Calculation.
Before I go any further, I do want to let those of you who are calling into the teleconference for the audio know that you need to mute your computer speakers or else you will be hearing your audio over the computer as well.
Before we get started on today's seminar I wanted to make a few announcements from the FHWA Office of Freight Management and Operations.
This week the FHWA Office of Freight Management and Operations released a RFA for Off Hours Freight Delivery Pilot Project Grants. All level of governments are eligible to apply. FHWA will award up to $450,000 in research grants to explore the benefits of delivering goods during "off hours" in small- and medium-sized urban areas with growing congestion problems. Working with the Environmental Protection Agency, the pilot will look at how truck deliveries made outside of peak and rush hours - when there is less traffic on the highways - can save time and money for freight carriers, improve air quality and create more sustainable and livable cities. The FHWA anticipates the funding will be used to help businesses retool their operations to accommodate shipments during off hours and distributors reconfigure routes and supply chains. While transportation agencies are eligible to apply for the grants, FHWA encourages them to partner with businesses involved in freight movement. The solicitation is posted at www.grants.gov . Details can be found on the link on the slide on your screen and also in the chat box.
Today we'll have three presenters - Roger Mingo of R.D. Mingo and Associates, Mark Berndt, of CDM Smith, and Eulois Cleckley of the DC Department of Transportation.
Roger Mingo, P.E., worked for FHWA as a Highway Engineer for 17 years before becoming an independent transportation policy consultant. While at FHWA, he worked in the areas of highway design, traffic operations, community impacts, and highway cost allocation. He has been involved in pavement cost modeling for more than 30 years, starting while serving as Chief Engineer on the 1982 Highway Cost Allocation study, and working on modeling improvements nearly continuously ever since.
Mark Berndt is an Associate with CDM Smith and is the discipline leader for Freight Planning. Mark has 30+ years of public and private sector professional experience related to freight planning and truck policy. After 20 years with the State of Minnesota, Mark joined the consulting community in 2001and has managed freight studies for a wide range of clients at the local, state and national level. Mark is currently the Secretary of the TRB Agricultural Transportation Committee and is the former chair of the TRB Truck Size and Weight Committee.
Eulois Cleckley currently serves as the Manager of State and Metropolitan Planning for the District Department of Transportation (DDOT). Mr. Cleckley also oversees the Motor Carrier Division which was established to develop policies and manage initiatives that ensure the efficient and safe mobility of trucks and buses in the District of Columbia while mitigating community impacts and preserving transportation infrastructure. Mr. Cleckley has private and public sector experience related to freight transportation, commercial vehicle planning and supply chain management. He also serves on the National Academy of Science/Transportation Research Board, Urban Freight and Vehicle Size and Weight Committees, and serves as the District of Columbia representative on the Metropolitan Washington Council of Governments Transportation Planning Board.
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. If we run out of time and are unable to address all questions we will attempt to get written responses from the presenters to the unanswered questions.
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 notify all attendees once these materials are posted online.
One final note: 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. I have included more detailed instructions in the file share box on how to obtain your credits after the seminar. Please also download the evaluation form from the file share box and submit this form to me after you have filled it out.
We're now going to go ahead and get started. Today's topic, for those of you who just joined us, is Pavement Damage Analysis Tool (PaveDAT) for Overweight Truck Permit Calculation. As a reminder, if you have questions during the presentation please type them into the chat box and they will be answered in the last 30 minutes of the seminar. I'm going to turn it over to Roger Mingo.
John Nicholas
Thank you, Jennifer. I'm John and I work in the Office of Freight Management. Just wanted to say a few words of introduction about the Federal Highway's involvement with pavement assessment tool project. The tool grew out of a joint research project out of Federal Highway's Office of Policy and Freight Management and Operations. The Office of Policy contracted to have research done on a simplified version of the national pavement cost model. At the same time, freight office identified the need to create an easy to use pavement damage assessment tool to in help determining the damage cost of heavy permanent loads. That's based on the MEPDG model as opposed to the ESAL alternative. They've agreed to launch the tool. Washington D.C. agreed in that effort. This past year we've trained five additional states in the use of the PavDAT tool: Texas, Pennsylvania, New York, Virginia and Washington State. With that, I would like to thank Rodger for agreeing to give a brief introduction of the tool and agreeing to share experiences.
Roger Mingo
This national pavement cost model, or NAPCOM, has been widely used for almost 20 years now for highway cost allocation and pavement cost analysis. It's been under continuous improvement pretty much the whole time. The full-scale NAPCOM is very complicated and it's intended to be applied to the HPMS inventory of all the sample highway sections of the country. When we were redoing NAPCOM and improving the models, it was a great opportunity to come up with a simplified model to be used for one particular vehicle at a time. We owe a lot to our predecessors in this. ARA designed MEPDG and Auburn University helped with the application of MEPDG for this purpose. We built upon that work and used a state of the art pavement model, MEPDG, applying it thousands of times in a variety of climate areas. I think it was almost 20,000 times that we ran the model in order to derive a new set of independently derived pavement damage equivalence factors.
You are probably familiar with load equivalent factors of some sort. ESALs (Equivalent Single Axle Loads) are one type of load equivalence factor (LEF). ESALs are based on a 50 year old study that was done in Illinois. It had limited application in that it really only applied to a very small number of pavements and one climate region. The pavements were made thin so they could be tested to failure by different axle loads in a reasonable period of time. They were not real pavement sections. We have been relying upon the ESALs derived in that study ever since. We have never had the modeling capability with models that have been verified with empirical data to come up with better answers. As far back as the 1982 cost allocation study, the trucking industry was against using ESALs, which have a fourth power equivalent. If you are twice as heavy as another vehicle, it does 16 times as much damage. This was maybe not far off for some kinds of damage, but it was very far off for others. As long ago as 1982, NAPCOM and FHWA have not relied upon the assumptions made for ESALs. Beginning in the 1982 cost allocation study, FHWA has tried to derive new LEFs for highway cost allocation purposes.
With this PaveDAT tool, we've been able to use a model that has been empirically verified. What does PaveDAT do? It estimates the pavement cost for a specific vehicle on a specific type of trip. That is, you have to know exactly how much your vehicle weighs, what its axle loads are, and what kind of highways it's going to travel over. Very quickly you can calculate what the pavement cost associated with that movement of that vehicle is. It's kind of oriented toward truck permits or overweight truck operations. It comes with a lot of default data. It relies on the same base data that NAPCOM relies on, which was used by FHWA for the last cost allocation study which has not been published yet (it has been in DOT for more than a year now). That data included about 200 million weight data observations for all the states, and a full set of estimates of travel by truck type. So with that data we can supply a lot of default data. The only problem is that the default data is now three or four years old, and it's time to update it. It's probably not horrible, but if a state really wants to use this and make it the model they're going to use for wide application, they should look at the possibility of updating their state's data. You can customize it as much as you want. As it is, it comes with data that is reasonably representative. So given that, all you really have to do is supply some trip information.
You need the axle weight for each group on the truck. You treat groups of two or three as one axle group because axles have a different effect on pavements depending on how they are grouped. You cannot just look at individual parts of the axle. You have to look at the group too. We simplified the kinds of highways upon which the truck travels. The National Pavement Cost Model has 12 different highway functional classes. We did not think that was reasonable for PaveDAT. There are broad groups of highway classes that have similar kinds of pavements, and there really is not much difference between urban and rural pavements. We grouped the highway systems into interstate, state highway, and local roads. The boundary between these highway groups is somewhat up to the state. Once you put those numbers in for your particular truck you get results instantly in cents/mile and total cost in dollars, and it shows the cost for each highway type as well as the whole trip.
This is the main user page. There are pages behind this with detailed calculations, but all your inputs and results show up on this page when you're using PaveDAT. All you have to do is supply the numbers that are blue and the results come down below in black. This is a little small to read. I am going to break this screen up. This is just an overview of what the screen looks like. As one example, let us look at what happens if you have a 97,000-pound six-axle truck. It is a tractor semi-trailer with a standard tandem drive axle and trailer with a tridem axle. Here are the results. This simple trip shows a breakdown of a small amount of local travel and moderate amount of state highway and most on the interstate system for a total of 100 miles. You have to supply the axles in thousands of pounds or kips, in this case. This slide shows a five-axle, 80,000 vehicle with a 12,000-pound steering axle and two 34,000-pound tandem axels. The weight of each tridem axle includes the total of the two single parts of the axle together. To analyze the 97,000-pound six-axle truck, you would put 57 here for the third axle weight and three for the number of axles. When you do that, you get these results. It shows the cents/mile traveled by each highway system, the total cost for that particular trip and that particular highway system, and then the total cost for all hundred miles. So, this weighted average of this particular example shows a 49 cent per mile average pavement cost. That compares with an 80,000-pound truck that has about 39 or 40-cents per mile.
Here is another example, let's consider an average loaded truck. If you look at all the trucks in all of our WIM data, five axle trucks average about 71,000 pounds. They don't always travel at 80,000 lbs. This average truck has approximately 11,000 points on the steering axle and 30,000 pounds on each of its tandem axles. This truck produces about 25-cents a mile of pavement damage in our example trip. The question is, if you are going to try to charge the 97,000-pound truck for a permit, do you want to charge the full amount of 49-cents a mile? Or do you want to charge it the difference between the 97,000-mile and the general legal weight of 80,000 pounds? Or do you want to charge the difference between the 97,000 pounds and the 71,000 pound truck? You get a different answer. If you compare it to the 71,000-pound truck, it is 25-cents a mile. If you look at the amount that an average over-the-road truck pays in user fees per mile, you get a difference of 40-cents a mile. It is a policy decision on the state's part to decide which number to charge, whether it is the difference between the user fees paid or the difference between the average truck, or the difference between the maximum legally loaded truck. There is a wide range of answers. That is not the model's job, that is the job of the state to decide what fee will be.
Now, in order to help you with that, the model lets you specify a vehicle. You don't have to look at these calculations. You don't have to look at what a 71,000-pound or 80,000-pound truck does. On another page of the spreadsheetl, you can supply the base vehicle. In this case I picked the 71,000-pound truck, typical truck, as a base vehicle. Once you do that, the results you get back show the depth. This shows now the difference between the 97,000-pound special vehicle and the 71,000-pound base vehicle. That 25-cents per gallon that we got manually before, comes up automatically on the user page. An overweight five-axle 3S2 with 10% more weight, 88,000 pounds total, does a lot more damage than a 97,000-pound six axle truck. For the 97,000-pound truck, it does about 25 cents damage. For the 88,000-pound truck with five axles, it does 38-cents worth of damage than the base vehicle. In that application, if you would like to look at what the effect of overweight trucks would be instead of permitted trucks, this shows graphically that you get a much higher number if you overweight them than if you don't overweight them.
The load equivalent factors are the guts of that model. They're based on many runs of the pavement model and you get a different load equivalence factor for each type of pavement distress. You really need the pavement engineers in a state involved in trying to help you weight how important each of the distresses are. Unlike ESALs, which apply to general pavement damage, these LEF's apply differently to different distresses. You have to have some sense of how important rutting is for flexible pavement compared with cracking, and you have to have some sense of whether your pavement mix design results in more rutting in the base layer or in the surface layer because each makes some difference in the result.
I'm going to skip over exactly how we derived the load equivalence factors. In the power point presentation you can download, it shows in detail the LEF calculations. This slide shows a set of runs for one pavement section we used, andit shows how we derived the target number of axles for each weight. It's kind of self-explanatory. I'm going to go over here because it's getting into more detail than we have time for. It shows we derive the number of daily axles to reach each target distress level. We used the 34,000-pound tandem axle as a base, and this slide shows the axle loads needed to create target valuse for rutting and cracking. There's also IRI which is the roughness. These are the distresses for flexible pavements that you have to weight based on how important each of those distresses are. This shows the example of how much work it takes, how many more axles it takes to create the cracking depending on what the axle weight is and type of axle.
The WIM data included in PaveDAT came from everything reported to the Federal Highway Administration during the years 2007 and 2008, which is a small subset of data that the states collect. Even though we had 200 million from 45 states, any given state usually collects that much in a year. You can get probably get better weight data if you want to update it for a particular state.
One of the other parts of the model is that we looked at the total amount of pavement expenditures in all the states, by all levels of government,to calculate the national current pavement expenditures per truck mile. We took the load related share of the pavement cost based on recent state cost allocation studies. Most states have not done a cost allocation study, so we used the few we had the results for to come up with what portion of pavement cost were load related by highway system. That's the basis for how we translated the loaded equivalent factors to cents per mile. A state can probably update that if they've done recent work with cost allocation or recent analysis on how much of the pavement costs are load related. The problem is in any given state in any given year they might not be spending enough money or might be spending more than they need and playing catch up. You have to look at how much over the last few years a state has spent, and then look at whether they're keeping up or not keeping up with pavement damage. That's why we use national data for this portion. We felt all the states' expenditures together over a three year period were probably fairly close to the pavement costs if they were keeping up.
When you try to use the data at the state level, you can customize it pretty well. Your state can modify the pavement distress importance, you can put in more detailed cost information, you can update WIM data, or you can combine default data from adjacent states. There are a couple states we had to do this for. There were six states we did not have data for and we used WIM data from adjacent states and the pavement type distributions in adjacent states. As John mentioned, we completed this model last year. We made presentations to five pilot states. So far, no state has adopted it for permit cost, but it's been used for special studies and applied in some form in Oregon, D.C., Wisconsin, and I might add Vermont because there's a special study where we used PaveDAT for the pavement analysis. Texas is in the middle of adapting the model for its own use. They want to have five distinct types of highway systems, and they're developing the data. They have the University of Texas working on improving the data in a special implementation project. It's been used in some places. With that, I'll turn it back over the Jennifer.
Jennifer Symoun
Thank you, Roger. Our next presentation will be given by the D.C. Department of Transportation. You can go ahead when you are ready.
Eulois Cleckley
Thank you, Jennifer. I would like to thank and I appreciate Federal Highway as well as Roger and the CDM for assisting the District with developing the District wide truck enforcement plan and using the PaveDAT model to help the District ascertain and understand what the impact of overweight vehicles are on the city infrastructure. We couldn't address the impact of the commercial vehicles traveling on the interstates. The whole study was conducted to try to really come up with a base number that we can present to not only our director from the District Department of Transportation, but our elected officials as well as the general public so they can have a good sense of what the impact of oversized vehicles is on the city infrastructure. I will provide an introduction and a background and then Mark will be going through the study process and findings and also the observations and suggestions that we learned from the PaveDAT model. I'll follow up with a conclusion.
For those of you who are not aware, Washington D.C. is really a changing and very dynamic city. It is approximately 70 square miles in size and we have had an increase in new residents coming since 2000. We're a little bit above 600,000 residents within the District. We have the highest personal income per capita with a record number from 2006. On the flip side, we have about 19% of our residents that area below poverty level. The city has really seen a boom in economic development which directly relates to how freight is moved in and throughout the District. We see a lot of oversized vehicles related to that economic development that spread throughout the District.
When it comes to how we have a transportation network that supports the economic development and something that is sustainable, understanding the data behind what the impact of these vehicles are to our infrastructure is extremely important to our director. The city recently has embarked upon another element in its sustainability position that the mayor just released about a month and a half ago. It is really focused on an economic development where we're trying to as a city accommodate an additional 250,000 residents over the next 20 years and have all trips in the District, in fact, 75% of the trips in the District, for residents or for commuters. 75% of those trips will be done outside of a personal vehicle either using walking or transit. This is all in relation to how we handle freight and freight movement within a District.
Washington D.C. is a city state, so we get federal funding from our gas tax just like any other state. From a freight movement standpoint, we are a destination point. We don't have a real bus freight industry. Our trading partners are in Indiana, Maryland and Virginia. We are very much impacted by the regional freight movement and the growth and regional freight movement here in the District. The District is buttressed by Maryland and Virginia, and the proposed different types of changes will have an impact on the District. So it's something we want to make sure we prepare for here and make sure our transportation infrastructure is addressing that accordingly. For a freight that is destined for the District is expected to have an increase of about 425%, again, by the year 2035 with pretty much to same trading partners.
With regards to the amount of commercial vehicles that are traveling and our infrastructure, we are experiencing quiet an abundance of commercial vehicles that are actually traveling on our system as opposed to our interstates. It might be flipped for other larger states. But here in the District, the majority of the absorption for some of these larger vehicles occurs in our arterial route system. We have about 45,000 daily commercial vehicles, single or two axel vehicles, coming to the District. That relates to the amount of construction vehicles coming to the District or small package deliveries that occur within the District, so we're not experiencing the high number of combination vehicles coming into the city.
So, with that as a baseline, one of the items that I've been trying to address here in my three and a half years in the District is to try to ensure that DDOT has a more robust system. The baseline to trying to understand this is getting a clear understanding of where we exist with regards to our policy structure. We knew before the study that we had a lot of dump truck vehicles and a lot of oversized vehicles on some of our major thoroughfares, but we didn't have the background data to clearly justify any additional enforcement needs and/or changing in policy. So, over the past several years, we embarked on the study, but also we really wanted to have the study be used as a catalyst to make changes and have a more robust program.
The study first took a looked at identifying truck safety concerns. The study was able to identify where we had high crisis areas within the District. The main thing the study focused on was assessing monetary impacts of overweight trucks. We wanted to come up with that value, so we can right size our fees and/or make sure we have the proper resources allocated for enforcement purposes. With the plan, we also wanted to provide a comprehensive approach to enforce truck operations throughout the city. Ultimately, we wanted to provide a safe environment that balances the needs of all road users in the District. We've used the PaveDAT model to move forward and come up with results which Mark will be explaining the approach from the utilization standpoint of the model and the results we came up with.
Mark Berndt
Thank you. I'll start off by saying essentially the study had three broad tasks. First was to look at truck safety hot spots in the District and evaluate specific safety issues. We also did an analysis that compared truck crashes in the District to surrounding jurisdictions to get a feel for where things might be different in the District. The second broad task was to quantify the effects of overweight vehicles on both pavement and bridge networks. That was a primary task that CDM Smith handled. Finally, we wanted to come up with conclusions and look at what the District should do in terms of future investments and future goals in terms of both truck safety and limiting the amount of damage from overweight vehicles.
So, I'm sort of building on what Eulois had said, but there was no formal truck management program in the District. There had been one previous examination of overweight fines conducted by Howard University. Initially, that study intended to look at a broader scope above legal and illegal truck operations, but due to data limitations and other things, it really focused on the fines that the District was charging in relation to surrounding jurisdictions. It basically came to the conclusion that the District's fines were probably not on par with some of the surrounding areas.
Eulois also had a number of other initiatives going. One of those was to adopt an official truck route network for the District. He was doing a lot of outreach at the time. Essentially, in taking on those three broad tasks that I mentioned, we were faced with a couple of challenges. One is a somewhat limited budget given the amount of work tasks required. Each task, especially the safety and pavement of bridge analysis, in and of themselves require a lot of detailed data and can become quite an extensive analysis. The other thing was the director of DDOT was looking for answers within about a six month time period. As we were wrestling with exactly how to approach the infrastructure piece of this, we had some discussions with FHWA, and they brought up this new spreadsheet tool they had called PaveDAT.
So, our bridge and pavement staff and I visited with Roger, and he provided us with an overview, and we decided to use it for this study. Again, it required some manipulation as Roger had mentioned, but there were a couple of real advantages. One was the model was preloaded with WIM data. It was several year old, but it still would give us a lot of time savings. We did do some manipulation of the existing information to make it more D.C. specific. We also looked at information from some of the surrounding jurisdictions. It was used it to manipulate some of the weight-in-motion data for the District in terms of what we ultimately used as the data set. We also used historical cost information for the District in terms of the annual expenditures. We used the model pretty closely for coming up with the pavement cost estimates. For the bridge analysis, we basically use the weigh-in-motion data set. Roger didn't talk about it, but the model is loaded essentially with both the raw data as well as an edited set of data. Roger has some various tools that, over the years in working with WIM data, have come up with some methods for going through that data and modifying it a little bit so that it is more reasonable. We were able to look at both data sets and make judgments on which worked best.
So to start out we also had vehicle class counts from the District, and given the fact that the District is a dense, urban area, we see that the single unit vehicles have the most mileage in terms of the weigh in motion site. So single unit two axle trucks by far had the most travel followed by three axle, and then it's almost an even split between the commercial buses and five axle tractor semitrailer combinations. This chart shows the axle counts from the single unit two axle trucks were by far the highest counts across the WIM stations. Because the three axle single unit trucks tended to show more overweight axles, they actually had the highest number of load equivalency factors. As Rodger mentioned, the load equivalency factors for MEPDG within the model would have required some additional data inputs from District engineers, and we talked about that. Given the timelines, we decided that we were going to have to use a somewhat more standard ESAL approach. We did manipulate the standard power of four model based on some more recent research and also used the ESAL factors for the different pavement types based on the pavement design standards for the different classes of highways in the District.
We also used the PaveDAT model on another study we were doing about the same time in Wisconsin, and so I thought it would be interesting to show you the results there in contrast to D.C. which was a dense, urban where the single unit two and three axle trucks had the largest load equivalency factors. When we look at Wisconsin, this was a statewide study, but it looked at a specific network which was their long truck network in Wisconsin. We see that the combination five axles by far outweigh any other vehicle class. This gives an example of how the different vehicle classes come into play.
Going back to D.C., this shows a graph by vehicle type, and you see across the bottom are the commercial vehicle classes and the FHWA classification system. In terms of overweight axle damage, the largest share would be attributed to the single unit three axle vehicles. This is only based on the WIM data and hasn't been expanded out to using vehicle miles of travel for the entire District.
This was actually supposed to be a slide comparing mileage vs. pavement consumption. Just before we went to press, I noticed the pavement damage was off by one class on the chart. So, I took them out. Remember that the single unit two axle vehicles have by far the greatest VMT. So, what we come up with is even though the single unit three axel vehicles showed a higher percentage of overweight axles, the total damage is greatest among the single unit two axle category followed by the single unit three axle. It's kind of a split between the semitrailers and commercial buses in the District. It's not uncommon for commercial buses to often have overweight axles, and so that does show up here in the data.
I'm not going to spend a lot of time talking about the bridge analysis. We did use some of the information from the PaveDAT spreadsheet to do the cost allocations for the bridges. Given the limitations and challenges, we had to make a few assumptions. One of the things we did was went back to some of the resent cost allocations as well as the last federal cost allocation. Those other studies suggested that overall, 59 or about 60% of bridge costs are allocated to passenger vehicles, and then the remaining 41% is allocated across all commercial vehicles, trucks, and buses. Then based on the axle loading information, we then further divided that out into a cost allocation for legal trucks and buses and then about 18% of the cost for bridges was allocated to overweight commercial vehicles. Then within that, we allocated across the different vehicle classes and again used annual average bridge cost to come up with some estimates. The total annual bridge cost attributed to overweight vehicles was about $10.5 million per year.
This next slide shows how the overweight pavement cost and bridge cost attributed to each of the vehicle classes. The total damage including buses attributed to overweight vehicles was about $20 million per year, or $16 million/ year just commercial freight. But again, a lot of that gets allocated to the single unit classes like dump trucks and trash compactors. Those types of vehicles are the single unit two and three axel trucks. So, what we found is that PaveDAT really worked well in terms of providing a readily available, low cost means to do an analysis that is often very resource intensive and data intensive. Just the fact that the weigh in motion data was there and readily available for all the jurisdictions was extremely helpful. We did talk with the pavement engineers in both the District as well as when we worked in Wisconsin. We ended up using the empirical method or the ESAL method, but relied on factors that had been updated a little bit since the original studies in the 1950s. It left open the option of talking with the pavement engineers about an MEPD approach. Again, this type of study is not what was intended when PaveDAT was developed, but I think it worked pretty well for what we wanted to use it for. It is helpful to look at the gross vehicle weights as well as the axle vehicle weights.
One of the issues they had was that there was no way to separate illegal trucks from trucks that might be operating over legal limits under permits. Typically, permits are issued to vehicles that have enough axles to support a higher gross weight, and you don't typically issue permits for higher axle weights although that is the case in some jurisdictions. In some of the multi-axle groups, the District allows higher axle group loading. That was a bit of an issue in terms of not being able to tell which were vehicles that were illegal verses which were vehicles operating under permit.
We came up with a couple of suggestions on how PaveDAT might be improved in the future for this sort of study. We implemented the study in 2010. The WIM data set that was pre-loaded was from 2005, and so they were a little bit dated. One of the things that we felt would be very helpful was that there is WIM data available through the Vehicle Travel Information System (VTIS) that you can get through the FHWA website. That data usually, I think, has about a one year lag. So, if there was a way of having the spreadsheet set up so you could download the most current data, it would allow you to get more timely data sets. But, it would require standardization in terms of format. I'm not an expert, so maybe that standardization is already there. If there's a way to use state collected data through things like portable WIMs and other things, it would be another useful way of being able to keep the data sets current. The cost data was a challenge that was more of a District issue, but we've encountered this in a number of jurisdictions. Looking at the cost data is not always uniform, and it is sometimes difficult to assemble over multiple years. The bridge engineers would have liked to have seen more details. Those are my observations. I'm going to turn it back over to Eulois to finish it up here.
Eulois Cleckley
Thanks, Mark. What we believe here is that PaveDAT tool was very useful. They mentioned the number we were looking for to be able to share with the community and share with our high political officials here, being the mayor or the council member or director, was $20 million of overweight infrastructure cost due to overweight vehicles in the District. It's something we have used to be able to justify and provide a reason for why we need to really take a look at adjusting our currencies for permitting. Also, we used it for taking a look at what some of our fines are with regards to those vehicles that are not abiding by our current regulatory structure. When we take a look at the remedies that we assume from the International Registration Plan, which is focused on vehicles over 26,000 pounds be it truck or bus, we take a look at the permitting fees that we actually receive from the industry as well as any fines that are associated with oversized vehicles. When we look at the 20 million-dollar number, we have really a short fall of about 12 to $14 million that the District is absorbing when it comes to the damage we have to fix on the infrastructure because we do not have the proper regulatory frame work or a robust safety program. The value of the damage was very important, and it was definitely a positive result and a benefit of the PaveDAT tool.
What we've done so far is we are working on making the adjustments on regulatory framework to make sure it's appropriate and in line with other states, but also from a permit fee standpoint, we can now take a look at adjusting our framework to make sure it's in line with what the damage would be on our infrastructure. One of the main elements we have been focusing on is improving our enforcement techniques throughout the District. We've begun to come up with a productive way of enforcing overweight vehicles on at least the interstate system. We have been taking a look at operations, but also trying to take a look at restructuring our enforcement activities and our enforcement program to not rely so much on the police department which is the only unit that can actually enforce oversize or overweight vehicles. We want the District to take a look at how we can have the proper resources allocated here in DDOT to have those resources be used for more enhanced enforcement techniques.
We also want to take a look at how we want to integrate additional forms of proactive enforcement outside of having individuals on the field. Since we are an urban area and really do not have a lot of land to develop and construct a full blown weigh station, perhaps there will be a way we can take a look at a relative evidence law for the District in cooperation with other carriers. We also want to take a look deploying technology to have some type of citation process and procedure to be able to enforce more proactively with the idea of bringing compliance so it can kind of minimize the damage that we would absorb here in the District.
At the end of 2010 and the beginning of 2011, we installed our first static-flat scale operation off I-295 in the south bound lane. It's about a half mile or so from the Maryland border. It really provides direct access to a lot of key parts within the District downtown area and a lot of other areas where there's a lot of economic development. So, based off of the data, what we see is that the trucks may not be overloaded coming into the District but definitely going out, moving aggregate, stone, and the like. We started the flat-scale operations as an attempt to create a more robust weight program here in the District. We're appreciative and excited for the Federal Highway allowing us to use the tool. With that being said, that's the study and presentation. I'm going to turn it back over to you.
Jennifer Symoun
Thank you. We're now going to go ahead and move on to the question and answer session with the questions posted into the chat box, and feel free to continue asking questions and we'll try to get to all of them. I'm going to go back up to the top to a question for Roger. The first question is Washington does not require a carrier to provide weight by axle. Would it be of any value to enter the excess growth weight for the permit? For example, permit application is worth 120,000 and the legal weight based on the expanded federal bridge formula is 100,000.
Roger Mingo
Okay. Jim, the answer to that is no. You really need to make some assumption about what the axle weights are. For example, that 120,000-pound vehicle, if it's on eight axles it is very different than if it's on seven. If you notice in the examples I used, I kind of used idealized axle loads for a trucks. I can assure you that of the many truck weights I looked at, I don't think there were any that were exactly that. That's an idealized load. You have to look at the number of axles and make some assumption about how evenly spread the loads are. If you make an even load assumption, it benefits the permitted vehicle the most. That is a conservative assumption. But, you have to somehow decide how to allocate that 20,000 extra pounds and how many axles there are.
Jennifer Symoun
The next question is did this study come up with a ratio of the damage done versus passenger vehicles? I have heard one truck equals about 4000 cars.
Roger Mingo
Well, that's one that, yeah, you could use it to derive that. It depends on your state's condition. The stereotype is one truck does 9600 times as much damage as one car, and that's assuming very simplistic axle loads. When the Federal Highway Administration completed the cost allocation study in 1982 when they first re-did the numbers, they tried to steer people away from thinking of it that way. In essence, cars do almost no damage. I'll come up with almost any number you want depending on how much you to assume. 4000 is probably more. Since a lot of the damage is not related to loads and cars get a share of that, you come up with a much lower number. If you look at only the load related damage you come up with a higher number. In other words, I'm going to punt on your question.
Jennifer Symoun
Alright. How was the cost per axle derived from the LEF? Are they different for flexible, rigid, and composite pavements?
Roger Mingo
That's a real good question. We don't have any special models for composites. We treat them like rigid. There are two models in PaveDAT. There's a rigid model and flexible model. The costs are allocated between the two of them for each state based on the proportions of miles we have from Highway Statistics, the number of lane miles that are rigid pavement, and the number of miles that are flexible for each highway system in each state. Those numbers are included in the model, and if you want to make a different assumption for your state or national average, you simply have to change the weighting factor for rigid verses flexible. The default data is based on miles. If you know what the costs are, if the state knows they spent X number of dollars on concrete and Y amount on flexible, then you can use those two cost figures instead of the miles. It's just a two number ratio on the second page of the model.
Jennifer Symoun
Where can we get the software?
Roger Mingo
I'll have to defer to John on that one if he's still around. It is property of the Federal Highway Administration. They made it available to several pilot states. When and to whom they release the software is totally up to them.
John Nicholas
We are still making it available to pilot states, and we are still contemplating about anything further.
Jennifer Symoun
Thank you, John. Are all state specific pavement types in the PaveDAT database? This person has concerns of extreme weather variances and other conditions.
Roger Mingo
Yes and no. We used a variety of climates. Those are state specific, and they deal with the response of different pavements to climates. We used a variety of pavement thicknesses, and there was surprisingly little difference for any given little distress for thickness. What does change a lot is what the prevalence of the distresses is as you go to different pavement types or different details of pavement design. That's where the state pavement engineer knowledge can help modify the default data that are in there. They use a much different pavement mix design in Texas than in Michigan for obvious reasons. It's much hotter there and they more of a rutting and cracking problem. That changes the distribution of the stresses you observe and that has a large effect on how much what one trucks axle has verses another. The opportunity is there to let a state make it custom.
Jennifer Symoun
What about axle spacing, is that a factor?
Roger Mingo
It is and that's why we use axle groups. The definition of exactly what is a tandem and what is a tridem goes beyond my knowledge. A fairly common tandem spacing is four feet. A fairly common tridem spacing is eight or nine feet. We use that as the default. For example, if you have a tandem spread that is eight or ten feet, then you're probably better off considering the group as two single axles rather than one tandem. Just in the same way that the pavement model considers spacing as kind of a typical for each of the axle types, we did that. We didn't do all the extra work required to look at what if you had a six foot spacing instead of four for tandems and what if you a ten foot spacing instead of eight on tridems. We assumed the same typical spacing for all tandems and tridems.
Jennifer Symoun
Also related to axels, what is the maximum number of axles that can be input?
Roger Mingo
Yes and no. If you have special tires and quadrems and the kinds of things used in fracking operations, it's beyond the scope of what PaveDAT looked at. It can handle up to eight axle groups. When I showed you the input screen, I deleted some of the lines to make it fit on the page better, but you can have eight axle groups with singles, tandems, or tridems. Depending on your configuration, it can handle up to 23 individual axles if all the axles except the steering axle are tridems.
Jennifer Symoun
What is the maximum weight of the truck that can be evaluated?
Roger Mingo
Well, we use the same axle weight inputs that the MEPDG used because that's becoming the standard, and the maximum weight is about 40,000 pounds for a single axle and roughly double that for tandem. You can in essence model axles up to that weight. If you have 23 axles at 40,000 pounds, that's a lot of weight. If you have axles more than 40,000 pounds per single axle, it goes beyond what the model can handle.
Jennifer Symoun
Okay. I think you pretty much answered this, but can non-standard truck configurations such as super loads with up to 16 tires per axle be analyzed using the software?
Roger Mingo
No.
Jennifer Symoun
Can pavement structure be input into the software?
Roger Mingo
This is a couple of questions coming later too. The model has a separate set of distress weightings for each of the main types of highways: interstate, state and local. Since you typically have much different pavements on a local road system than an interstate, you're going to observe different distresses. You can change the weighting of the distresses of each of those highway systems to account for mostly what you observed. This is a function of what kinds of pavement you put on. You can't change from six inch to eight inch slabs. That is beyond the scope of what can be done.
Jennifer Symoun
Okay. I'm going to ask one more. How many weight entries will be considered a valid sampling?
Roger Mingo
You're talking about WIM data. I wouldn't bother updating because we have at least that many for each of the states we used, and I think you're better off using slightly older data. The problem is that you get lots of five axle trucks in almost every state, maybe not D.C., and you get lots of three axle trucks. For some of the rarer truck types, if you don't have enough truck weighifngs, you're not going to have a valid weight distribution. There's no hard and fast answer. I guess you should look at it and see if it has adequate representation. I wouldn't bother for anything less than a million.
Jennifer Symoun
Alright. Thank you. We'll move on to other questions. Based on the placement of these no truck signs, was it ever considered that trucks may be forced into indirect routes between point A and point B or is it possible that some destination pairs were some inaccessible because of this strategy?
Eulois Cleckley
In the District, we spent quite a bit of time doing quite a bit of outreach to communities to come up with an appropriate routing system. We took about 16-18 different data points to try to figure out where trucks needed to go whether they be on an arterial route system or a collector system. Any truck restrictions on any other route outside of that identified truck route system is going to be consistent with our local street network and some of our collector road network. So really the issues with regards to having an impact on the routing really doesn't pertain to that overall effort because a lot of the local roads are roads that a lot of trucks don't need to be on to begin with. So really the assessment was to come up with a system because we did not have a system to make sure that we're sharing this with the industry to make it official in our code of regulations. It's something we can enforce effectively. Making sure we have a system that provides connectivity in and throughout the city was a part of it.
Jennifer Symoun
The next question, aren't commercial buses exempt from the weight limit laws?
Mark Berndt
Typically they are, but again, that's why we produce results that showed impacts with and without. But in most states, buses are not required to pull into weigh stations and they're not subject to overweight axle laws. I think that's the case in the District as well.
Eulois Cleckley
That's correct. I think for the District and other states that have high tourist attractions, this is a severe issue in the city. We just recently completed a study in conjunction with our MPO to try to come up with the number and volume of commercial buses by bus type that come into the city. On an average high tourist day, we're looking at totals over 1800 buses per day coming to the District. Relating that traffic to the amount of damage that a lot of commercial buses have on our city infrastructure is important. You may not have as many commercial buses coming into an urbanized area like us or throughout the state for the District. We're absorbing these costs, and it is something that we also share with our industry partners as well, so they have a clear understanding of what we're dealing with when it comes to commercial buses.
Jennifer Symoun
Okay. Thank you. Is there any reference for developing the cost data you mentioned in your presentation?
Mark Berndt
Well, it was just records that the District turned over to us from their accounting system on annual outlays for pavement rehabilitation and annual outlays for bridges. So it was just like internal documents really that we went through and came up with an annual average cost. We didn't have a specific reference for that. I know Roger said that there was annual cost based on the national average, but we didn't have a specific reference for the cost.
Jennifer Symoun
Roger, was there anything you wanted to add as far as the national data?
Roger Mingo
This is the cost data? In the cost data, we developed it from highway statistics reports which are not necessarily the best way to get it. It's very difficult. In my experience, it's difficult to get the detail you need to make some of the decisions you have to make about what's load related and not. That's often a major part of the highway cost allocation study. If you haven't done that, the best you can do is to look at overall expenditures and look at someone else's experience and what they found as a portion of load related cost. That's what we did for Federal Highway. We did the latest three years reported of different pavement expenditure types. That is one difficult part of the job, yes.
Jennifer Symoun
Often weigh-in motion stations mistakenly include personal occupancy vehicles as trucks. Did you try to confirm or verify in any other way that your astronomical counts were not due to miss-collaboration?
Mark Berndt
Well, really we looked at two different sources. There were vehicle class counts and that's where you show the high number of single unit two axle vehicles. The weigh-in motion senators picked up on that as being the highest count type vehicle. If you recall, the SU-3 had the highest percent of overweight axles in a percentage basis. We looked at the two different data sources including the ones the District collected as well as the three weigh-in motion stations, and we looked at surrounding jurisdictions and looked at the bell curve of the vehicle class counts from surrounding jurisdictions in Virginia and Maryland. I guess we came to the conclusion that the counts were reasonable based on what we were seeing.
Jennifer Symoun
When talking about legal weight verses overweight, what is being used? How much overweight are the vehicles being looked at?
Mark Berndt
Well, again, it varies. What you get from the weigh in motion data is a bell curve of axles by class. We use the commercial classes out of the 13 FHWA vehicle classes, and we use the District's axle weight limits as the limiters in what was considered legal and illegal. Those are the basic parameters.
Jennifer Symoun
Can you give more information about assembling the detailed pavement rehab cost? Did you look at primary verses secondary designation?
Mark Berndt
Again, I'm not sure about primary versus secondary. The allocations were not done for every street in the District. Again, it was limited to what was being defined as the commercial truck network. So it tended to be higher classes of roadways, and what we did was assemble investment or pavement cost data based on the roadway types in terms of what the District had invested in the past five years. We then came up with an average. It came out of a number of different annual reports that the District produces, and I'm not sure what the primary and secondary alludes to. The challenge was we had a six month time line and limited budget. I would agree that often times there's certainly a lot more research and a lot more analysis that could have taken place to come up with maybe a more exact answer. I think the idea was to give you an overall sort of feel for what the costs are based on the total cost outlaid and what the weight data being collected suggests in terms of the allocation of legal weight versus overweight.
Jennifer Symoun
Thank you. Back to Rodger, we have a few questions here and I think we've covered these but I'm going to combine them together. Most of the questions have to do with the difference between pavement types and the thickness of the pavement and grades used.
Roger Mingo
The only way to do that is by distinguishing among the stress prevalence on the different road types. This is what the main driver of the costs are.
Jennifer Symoun
Thank you. John, this might be a bit more for you. When talking about changing overweight vehicle enforcement policies is consideration being given to moving from a single weight approach to an ESAL based approach?
John Nicholas
That might be more for Rodger.
Roger Mingo
PaveDAT still has ESALs in it. As Mark described in the district, even though in many quarters, ESALs are not well thought of, in some they are. PaveDAT allows the option of weighing all of the stresses as 100% ESAL and not weighing the other LEF's. You can make that direct comparison for each vehicle and see what difference it makes. I haven't done that for all of them but I think it seems like the heaviest axles would get charged more under ESALs where the middle weight axles would get charged more under the LEFs.
Jennifer Symoun
I'll put this out to all of you. What feedback have you gotten from the trucking industry? How do they feel about paying for pavement damage based off of assumptions?
Mark Berndt
I made some presentations on this work in a few places. First of all, the trucking industry is on record for a number of years now as saying that we are underinvesting in the highway system. So the idea of paying more is not something that the industry overall feels as unwarranted. I don't think there's anyone I've spoken to that feels that the illegal operations should not pay their fair share. No one wants to see illegal trucks. It's not only a pavement damage issue, but if you're operating illegally verses someone whose operating legally, there's an economic incentive there too if there is no enforcement going on. The idea was to not necessarily directly charge based on these results but to make the case to decision makers in the District that it would be worthwhile to put more effort and resources into enforcement. The potential payback is much greater. I think the overall idea here was to raise concern about the need for more enforcement to actually catch the violators in the act. There was previous work that suggested the District might be behind in terms of the fines it was charging as well.
Eulois Cleckley
Just to piggyback on that, first we're concerned with compliance. We have to make sure we enforce the current regulation. The study kind of highlighted the need to do that because we were able to come up with that hard number. In conjunction with that, some of the things we've been looking at are the fact that our structure when it comes to larger vehicles and more specifically overweight and oversized vehicles, it's not in line with other states. It's not a direct correlation between the cost that we've identified as part of the study to making those permanent adjustments per say. It's just making sure we're in line with other states and more specifically to the states in the region since we're not a major generator of these passenger vehicles that are coming through the District. We need to be in concert with out surrounding states when it comes to fees and fines.
You can hall a large vehicle and only pay $16, and that's something that's been a flat rate coming to the city. You could travel anywhere in the city as long as you obtain that permit. That actual permit fee has not been adjusted in over 35 years. This is something as a city as we move toward having additional changes throughout the District that we have to address from a policy standpoint. So really, the study and the model helped highlight the need for us to have additional resources for enforcement but take a look at our regulatory framework.
Jennifer Symoun
Well, I think we've gotten through all the questions. We tried to combine all the ones related. We'll go ahead and close out for today. I want to thank all three presenters or actually four including John for presenting at today's seminar and thank you for everybody in attendance. The recording and the presentations will be posted online in the next two weeks, and I'll send out an e-mail when they are available. As a reminder, if you're an AICT member and would like to receive 1.5 certification maintenance credits for attending today, make sure you were signed in with first and last name or type your name into a chat box. The seminar is not yet available on the calendar, but I will send out an e-mail once it becomes available. I encourage everyone to download the evaluation form and send it to me after you have completed it. The next seminar will be held on July 18th. The seminar is not yet available for registration. It should be within the next few days. I'll send a notice out once it is available for registration. With that we'll go ahead and end. Thank you everybody and enjoy the rest of your day.
