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 Monitoring and Measuring Freight Mobility. Please be advised that today's seminar is being recorded.
Today we'll have two presenters: Crystal Jones of the Federal Highway Administration Office of Freight Management and Operations and Dan Murray of the American Transportation Research Institute. I'd like to thank Dan for joining us at the last moment, as our originally scheduled presenter, Michelle Teel, of the Missouri DOT had a last minute emergency and is unable to join us today.
Crystal Jones is a Transportation Specialist in FHWA's Office of Freight Management and Operations. She joined the Office of Freight Management and Operations in October 2003. Within the Office of Freight Management and Operations Crystal is responsible for developing and implementing programs, policy and initiatives aimed at improving the efficiency of goods movement on the Nation's freight transportation system. She has responsibility for managing FHWA's Freight Performance Measurement Program, an effort that utilizes anonymous commercial vehicle location information to derive measures of freight mobility. Crystal holds a Bachelor's Degree in Industrial Technology with and concentration in Computer Science, and a Masters of Science in Administration.
Dan Murray is Vice President of Research for the American Transportation Research Institute. He is responsible for developing and directing ATRI's substantial portfolio of trucking and transportation-related research and training initiatives. Mr. Murray works closely with academia, industry representatives, other research entities and all levels of government to identify and address the pressing needs of transportation stakeholders -- with a particular emphasis on:
Mr. Murray received his Bachelor of Arts from Gustavus Adolphus College in St. Peter, Minnesota and his Master of Science degree from Northwestern University in Evanston, Illinois.
I'd now like to go over a few logistical details prior to starting the seminar. 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 smaller text box underneath the chat area on the lower right side of your screen. Please make sure you are typing in the thin text box and not the large white area. Please also 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. Once we get through all of the questions that have been typed in, the Operator will give you instructions on how to ask a question over the phone. If you think of a question after the seminar, you can send it to the presenters directly, or I encourage you to use the Freight Planning LISTSERV. The LISTSERV is an email list and is a great forum for the distribution of information and a place where you can post questions to find out what other subscribers have learned in the area of Freight Planning. If you have not already joined the LISTSERV, the web address at which you can register is provided on the slide on your screen.
Finally, I would like to remind you that this session is being recorded. A file containing the audio and the visual portion of this seminar will be posted to the Talking Freight Web site within the next week. We encourage you to direct others in your office that may have not been able to attend this seminar to access the recorded seminar.
The PowerPoint presentations used during the seminar will also be available within the next week. I will notify all attendees of the availability of the PowerPoints, the recording, and a transcript of this seminar.
We're now going to go ahead and get started. Today's topic, for those of you who just joined us, is the Monitoring and Measuring Freight Mobility. Our first presentation will be given by Dan Murray of the American Transportation Research Institute. 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.
Dan Murray:
I'm going to zip through some slides quickly. The interesting stuff is in what is in Crystal Jones's presentation. That's the meat of the discussion. We have been working with Crystal and Federal Highway Administration on a number of performance measures and initiatives, because the trucking industry is one of the biggest stakeholders in the country when it comes to performance measures. I'm going to fly through slides and give you a basis of rationale for why the trucking industry is so interested in FPMs.
As background, ATRI is a charitable not-for-profit group. We're not allowed to lobby or advocate. We have a board of directors. The most important underpinning of our work is we have a research committee to develop the research agenda for the trucking industry. That group is composed of academia, state and federal government, carriers and industry, et cetera, a very broad group. Once a year they identify topics that are hot. Here's a snapshot of the board of directors.
Key realities, this really is important. I've worked for an MPO prior to my position here. I have worked in economic development. As many of you realize there's a large disconnect between industry and government. Our goal is to develop partnerships. If you would go back and say what does government want to or need to know about industry? Government needs to better understand the realities of the freight industry. Several key realities: class one railroads are at capacity, and are experiencing huge delays. All modes have their issues; water and maritime ports are locked up, a key source of congestion. Domestic air cargo shipments are in dramatic decline. If you draw a compass on a map ten hours out, a truck will probably get it there faster, cheaper. All of the modes have their issues.
I'm going to focus on trucking industry, obviously. We've been deregulated twice, but some would say we're still one of the most heavily regulated industries out there. It's extremely competitive. You can get a CDL, lease a truck and you're up and running, it's so easy. In the last three years we've had over 100,000 new firms enter the marketplace, that creates extreme competition and low profit margins. It makes it difficult out there for the carriers who do exist to plan beyond a 12 to 24 month period. And business failure rates are high. That's because margins are low. This is no surprise in terms of fuel costs, volatility, when you have margins at 3.6% it becomes challenging to maintain an operation that stays in the black.
Now if we switch hats, there are a lot of people involved. Some for infrastructure management, transportation efficiency, safety and hazmat management. All of these parties are involved in some aspect of freight transportation planning. One of the other big disconnects we face is the transportation industry doesn't understand the government planning processes or products. If you look at this list the shortest window of all of these is unified work plan of two years. Some these plans are 20 years, typically. You juxtapose that with industry and ask them about 24 months and they're clueless. Nobody saw a trucking recession or skyrocketing fuel prices a year ago. So it's very difficult from a planning stand point to keep up with the long-term planning activities of government. But through Crystal's program we're making inroads in that regard.
This is probably one of the most controversial slides. I do it to shake up the audience. One of the statements, I say to government partners is if you're not tackling a "top industry issue" you're probably not going to get the trucking industry's attention. Now ATRI has polled executives to say what are your top issues? We do that every year to follow trends. If you look at congestion, in 2005 it was number 8. In 2006 it was number 5. I'll go through a couple of these too. Fuel costs don't seem to disappear. In 2007 congestion is number 4. I was surprised, I thought it would hit number 2 or 3. But it's going in the wrong direction. Why do we care? This is one of these win-win opportunities that we have between industry and government to target congestion. Clearly some of the initiatives out there may not have high favor with the industry. Certainly all of them ought to be looked at and evaluated. When that is done I think some of our best, longest range opportunities for partnership exist in tackling these actives and lists.
That said what are options and solutions? Congestion, if that's the partnership we attack, which is a win-win is an important issue for the trucking industry. Last year we spent, in a 12 month period, an additional $10 billion in fuel costs, primarily because of increased costs of fuel. But indirectly the research shows idling and air emission requirements which increase weights on trucks, all of these secondary effects reduce fuel efficiency. The more we demand the higher the price goes. So I'm on record saying if we could target one topic here, congestion would probably address, including safety of course, seven out of ten items on the top ten list. This seems like a great opportunity, Crystal will talk shortly about options. But clearly congestion is not going away. I point out for example the new hours of service, which is between the court system and FMCSA right now -- we're hoping to get rule making by mid December. There's no congestion credit. I'm going to drive for ten or 11 hours depending on what they say. But at the end of the day 20 minutes lost in congestion is taking money out of the pocketbook of the driver or company.
What are the options? You look at this and you thing we might have a congestion problem out there. This is one map of the freight analysis framework. We all use this to get a high level view of what is going on now and a look into the future. When you look at this clearly it looks like we have congestion on the interstates and the national high system. What we can't do today is drill in and see what we ought to care about. The project that Crystal will talk about lets us get down in there. You can see that Chicago has some congestion problems. It may not be granular enough to look at a FAF map at this level. So we drill in a little bit more. Now we can start to see networks of alternative routes, areas of focus, whether that's bottle neck removals, but it's still from a planning stand point may not populate a TIP. We want to get closer.
What we now discover is what may be have been a system on major throughput really is a percentage of truck trips off the major system. Now we're getting the attention of municipalities. Does congestion pricing on an interstate make an impact or raise costs? Now we're looking at economic development, planning and zoning research to determine where we can make a difference on the transportation system.
Lastly you go into downtown Chicago using real truck data, you start to say this may not be a national, or region level issue. We may have to start working with downtown business councils on street directions and truck parking. For the first timing I believe, we have enough granular data to start to build this larger partnership, identify the stakeholders and parties who are not necessarily been at the table, and bring them in. Hopefully start to make some serious inroads on at least congestion, if not many of the other topics.
With that said, I wanted to throw out two suggestions from our Research Advisory committee. Just in August we identified our top ten research priorities. Two of them are interesting. One is the role of shipper contracts and requirements on safety and freight transportation. The vast majority of people think that trucking companies are charge of their destinies. In many cases contracts dictate schedules and travel times. McDonald's will not allow you to deliver during their rush periods. We're going to look at the role of shipper contracts. That's valuable information to determine who you are targeting and what are the impacts. Another is the safety impacts of congestion. This is of great interest. Whether or not moving trucks out of congestion would improve truck safety in general. To be determined, but an exciting project that has just begun.
The good stuff is coming from Crystal in terms of how things can be developed and monitored. Feel free to contact me at my email if you have additional interest. So, I would be happy to answer questions when Crystal is done.
J. Symoun:
Thank you, Dan. I see some questions posted. We will get to those after Crystal's presentation. I encourage you to keep posting questions. We'll now move on to Crystal Jones of the Federal Highway Administration Office of Freight Management and Operations. Crystal, when you are ready.
Crystal Jones:
Good afternoon. Thanks for joining us on the eve of Thanksgiving. It's a big turn out. I'm from the Office of Freight Management and Operations within FHWA. The purpose of this presentation is to provide information on FHWA's effort to measure freight mobility. The desired outcomes are to familiarize with the audience with the effort and to provide an opportunity for input and feedback on future research.
I'll go over our objectives. I'll provide some background on the project and will talk about the data applications, challenges and opportunities, future research areas and I'll conclude with where we're going in terms of our current research agenda.
Listed here are objectives of the Office of Freight Management and Operations. The first two bullets are really where the freight performance measurement initiative fits. The first, understanding the magnitude and geography of freight movement, gaining a better understanding of how freight is moving and the impact of congestion is a key element of the effort. We also want to use the effort to develop freight data products and tools for our partners, state DOTs, MPOs, to use for their planning and other purposes.
Within Federal Highway we have an objective called global connectivity. As stated here the objective is to facilitate a more efficient domestic and global transportation system that enables growth and development. The desired outcomes and goals of the objective are listed. When this objective was developed it was done because it was important, there were no preconceived notions about what would be the most appropriate outcome measures to monitor progress in the objective. Through research, internal coordination and coordination with stakeholder it was determined that travel time and reliability on freight significant corridors and border crossing and delay time were appropriate outcome measures.
The selection and development of these measures were supported by FHWA sponsored research that identified point to point travel times on freight-significant highways and crossing times at international borders as key indicators of freight performance. We coupled that research with some input from industry. We concluded that speed and reliability was the indicators to focus on. Listed here are examples that highlight that speed and reliability are indeed of interest to the industry and to the public sector transportation agencies.
In terms of the freight performance measurement program this slides provides information on the what, why, how and where. We're using trucks as probes to monitor congestion and delay on the system. The why, to support FHWA performance initiative. We started the project in about 2004. We did it through a partnership with the American Transportation Research Institute. ATRI is the third party managing this program on behalf of FHWA. We're routinely analyzing data for 25 interstates. Right now the border component focuses on five border crossings of the United States and Canada.
We took a crawl, walk, run approach. At the onset, we focused on five freight significant corridors. The five were I-5, I-70, I-10, I-45, and I-65. We collected data from trucks. These were the corridors included in our study efforts at the beginning. These are the current 25. Listed in red are the original. In blue are the ones added in 2005. The brown is the rest of the interstate. As Dan alluded to, we have access to that data for interstates but on a routine basis we focus on the 25 shown in red and blue on the map.
To some brief information on the methodology, I added this slide. I'm not going to go into detail. I wanted to show this illustrate that we're not getting all of the data from a carrier or vehicle. We're getting these five elements, ID, date and time, location and highway. The highway indicator is for now an interstate but it could be any particular highway. Again these are the basic the elements that we get from the commercial vehicles.
This slide goes a little bit into the processing method. FHWA doesn't take possession of the data. Early only we analyze the best business management plan for the data and we made a decision based on several issues not take possession of the raw data. ATRI is the third party that manages the data. This slide goes through the processing methods used to transition the raw data to data products into products and tools that FHWA uses.
The measures that we're deriving from the data are average operating speeds and travel time reliability. These measures can be derived for a particular segment or the entire corridor. I'll talk a little bit more about travel time reliability and the measures that we're deriving from the data. For the borders we're deriving crossing time and crossing time reliability. We picked reliability, again because the research showed that it's important to the industry. It's something that public transportation agencies focus on. A point I want to make here is that travel time reliability is a delayed measure that focuses non-recurring congestion and unexpected delay. We're using averages, but because we understand that averages don't tell the whole story we're also focuses on other measures such as reliability. The measure we use is similar to the reliability measures developed for urban areas, we're using a buffer index. It's the extra travel time needed to go from a particular highway segment to ensure on time arrival. For our effort we're using a 95th percentile travel time.
We have a commitment within our program to look at methods and data to make sure that we produce measures that are appropriate for various audiences. Last year we had the Texas Transportation Institute do a study. This shows limitations of the measures we currently derive from the data. Right now the way we process data does include pickup and deliver, hazmat and routing. A current program focus is to determine if there's a way to extract parts of the data set that may not be relevant to a particular audience. As an example, things like delay from a weigh station, hazmat or detours may not need to be included if a state is trying to determine if their infrastructure is adequate to meet demand. Again a key element of our current research is to determine how to extract from the data to derive measures that are useful for different audiences.
The effort to date has been very nationally focused. We try hard to support customers that have a request from the data. This is an example of how we used the data to support a Department of Defense request. They were having carrier performance issues on a trade route between a depot in Pennsylvania and an air hub in Charleston. The carriers were stated that congestion and weather were part of the reason that they were not able to meet their schedules. We used the data from FPM to analyze the route and provide feedback delay and reliability in their particular area of interest. This is just a snapshot for a particular customer.
I mentioned that our research is proposing to look at other measures, recognizing limitations based on the data elements we receive from the vehicles. These are potential measures we are considering. A measure we are likely to develop is city to city travel times. We also have an interest in deriving measures from this data that are specific to incident-based delay, weather-based delay, etc.
For the border component, since May of 2006 we've collected data from five crossings, they're listed here. A primary criterion for selection was truck entries into the US. The five crossings currently included represent approximately 50% of the total entries into the U.S. The point that I want to make here is that the effort is not set up to look at how well customs is processing vehicles. It's to look at the transportation system to supports a crossing.
Here's an example, we have data for both sides. We can go up to 100 miles on either side of the border. It looks at the border as a system. We didn't have the luxury of directly porting over our methodology from the corridors component. This is a graphically depiction of how we set up our processing method. What we did, we used equal distance zones on each side of the border. For any particular zone shown here, for instance CA2 on the Canadian side to U.S. 3 we can calculate travel time and travel time reliability. It gives us flexibility in determining what areas will be measured at a particular area of a crossing.
This is an example from on of the five crossing. Again we collect data on both sides. This represents one year of data and shows average crossing times by month.. This data, I believe, is on a two mile segment around the border.
Talking a little bit about potential data applications. Our core customers are state DOT, MPOs and academia. Missouri DOT has a Tracker Program. They were an early adopter of using our data. They have a goal related to commercial vehicle efficiency. For that goal they use average travel speeds for trucks on selected roadways as an indicator of how well they're meeting that particular goal. We provide Missouri DOT their segment of I-70 and data for all of I-70. As you can see they translate the data into products for their Tracker Program. In their case what they do, for their tracker process they drill into the data. They find out why is it between miles 60 and 80 you see a drop in average speed for their segment of I-70. They also compare their section of I-70 to the rest of I-70. In this graph it shows data from January to December of 2006. April just happened to be a month where Missouri started a highway improvement effort, their smooth highways project. They had work zones and constructions. As you can see, the initiative likely impacted travel speeds in April.
As Dan said, we have a significant amount of data for a large amount of highways on our network. We do see a lot of value in using this data. This particular slide is a new application that we're exploring. It looks at how far a particular commercial vehicle or trip is able to go in a particular timeframe. We certainly see that this data can be used for measuring particular areas. This focuses in on the L.A. area.
We want to look at how, how nonrecurring events affect the highway. This is a weather event from the Denver/Kansas area. This shows the progression of a storm that went through the area. As you can see if you look at the two slides, this area around Denver has some delay. You can see the impact that this weather event has on the system as it moves to Kansas. This shows the day that Kansas DOT shut down the roadway in that area.
Truck parking is another application that we're using at using the data for. New York provided us information on their truck parking areas. What we did, our data processing method includes a procedure that takes out trucks that we believe are stopped. We looked at where they were stopped. As you can see, the areas indicated by red are known truck parking facilities. There's a significant number of trucks in those areas. In addition, there are areas along the highway where trucks are stopping. We're looking at whether this can be used to identify where trucks are stopping where there not be adequate truck parking facilities.
Another area we are examining is whether the data might be able to support travel demand analysis. This slides that follow illustrate what times of day we are receiving data from the trucks. As you can see there are significantly more trucks from 12:00 p.m. to 4:00 p.m. when compared with 8am to 12pm.
Here are the benefits of using probe vehicles. One of the unique facets of our effort is that it separates trucks from passengers and enables the derivation of freight specific. That's one of the benefits of the methodology that we use. It covers a wide area without additional instrumentation, giving us the capability to move beyond examining congestion in urban areas. With benefits also come issues. First, there is significant amount of data to be managed and processed, just for the 25 interstates and it will be even more data with expansion beyond the interstate. FHWA has been working really hard with ATRI to develop methods that will allow us to pass on and share the benefits of this data, but a challenge is the need to ensure the data is not used inappropriately (e.g. for enforcement). Our methods and data sharing agreements have to be managed in a way that we mitigate risks of abandonment.
I stole this slide from Sean Turner who sat on a panel with me recently in California. He was talking about some data quality and rules one should follow when determining what quality is good enough. One of the rules he used that I particularly liked was that perfect data doesn't exist. Along those lines, while we're committed to moving forward and continuous improvement we do believe from some applications the data is good enough no. He used to maps to demonstrate evolving quality. This is what was good enough in 1674. This is 1788. And 1849. And then 2007. Similarly for our project we have maps to demonstration the program's evolution. In 2003 this is where we were, a small number of corridors and a limited amount of data. In 2005 we have evolved, primarily be improving our processing methods and expanding analysis technique, we also moved to a model where we directly partnered with providers instead of individual companies. . In 2006 we expanded our coverage significantly and continued improving processing methods. In 2007, as Dan spoke on, we do believe we're at a point now that we have more granular data and we will be able to support for applications beyond the state and national level and maybe even down to the local level.
In terms of next steps, I said that Federal Highway views the states and MPOs as our core customers and accordingly we want to make sure the data is accessible to them. We're developing a web-based tool that takes into consideration privacy and other issues associated with the data sharing. The purpose of the tool is to make sure that data is available to our partners. We're expanding beyond the interstates. We're also enhancing our data by adding additional vendors and fleets and partnering with agencies and universities to apply the data. And expanding data collection on the border by up to ten more crossings with the U.S. and Canada. And we also want to start receiving data for the U.S. and Mexico border. We have two efforts, one in San Diego and one in El Paso to look coupling this methodology with other data collection to derive travel time information.
Here's a slide on travel time reliability. I thought I put a link in there to the report but I will have to get that to Jennifer Symoun. I think that's all that I have.
J. Symoun:
Well, thank you. I hope that everybody enjoys the presentations. We have a few questions that we'll go through. Once we get through the questions we can open up the phone lines as well. I'm going to start from the top with the questions that were typed in for Dan.
There are two similar questions, what did the dots represent in downtown Chicago and what are the red dots on the "granular" level maps?
D. Murray:
Essentially the system allows us to manipulate the data in numerous ways. Those maps, we took a 48 hour time period and looked at specific truck positions over 48 hours. There's some duplication there. Our database is maybe, maybe close to 10% of all trucks in the U.S. transportation system. That was a snapshot of specific truck positions in 48 hours. What Crystal is also doing is taking an migrating to speeds and times. You see then continuums.
J. Symoun:
Thank you. The next question is have you defined delay on urban arterials and how do you measure it?
D. Murray:
Crystal, you can have that one.
C. Jones:
The answer today is that our effort thus far is at the national level. Most of our effort has been consumed by deriving national measures. Our goal, through the web tool, is to provide that data to those folks that better understand their highway network. We haven't had the resources to focus on deriving a methodology for urban arterials, our current analysis focuses only on interstates. But again, our goal is to get the data those that want do their own analysis below the national level. There are limitations on what formats we can share the data, but the web tool will address those issues. In FY08 we do plan to do analysis on up 5000 miles of freight significant highway segments that are not on interstates.
D. Murray:
I will say that ATRI is releasing a beta model at the TRB Annual Meeting that allows government and industry to play around. In that model we allow the user to define delay either as something under design capacity, or something slower than posted speed limits. People can play around with it.
J. Symoun:
That will be at the TRB meeting in January?
D. Murray:
Yes. ATRI will of course have a booth there and stacks of the CDs will be available.
J. Symoun:
The next question. This is for Crystal, Dan if you want to comment too. Was there any criteria based on which these 5 corridors were selected?
C. Jones:
Based on who our volunteers were at the time and where their fleets operated. We used that, we used qualitative input from the industry, and the Freight Analysis Framework (FAF). It wasn't to say these are the most significant. The initial effort was a beta test and the five corridors evolved into the natural selection for expansion.
J. Symoun:
The next question, have you considered including interstate corridors into major marine ports -- such as Norfolk VA, Port of NYNJ or Port of Los Angeles CA - in addition to land border crossings in you assessments of the transportation infrastructure performance?
C. Jones:
That slide that looked at truck traffic coming out of the L.A./Long Beach area is an example of how this can be done. Our effort thus far has been aimed at deriving national measures for the interstate highway system. But in this phase of research targeting bottlenecks, key infrastructure, like port areas is definitely on our agenda.
J. Symoun:
Okay. Does this study also look into U.S. Mexico border crossing issues through Texas?
C. Jones:
The issue that we identified when we went to the border crossing areas, was that the nature of the operations on the U.S. Mexico border might limit how much data we would be able to get. Initially we thought we wouldn't have enough to do what we were doing on the U.S. Canada border. We have two projects, one in the El Paso area. We will use RFID to measure crossing time. We hope to supplement it with GPS data. And in the San Diego area we have a project that we'll be starting. That will be using license plate readers or GPS. Those are the two efforts currently to expand to the US/MX border. The ideal is at the end of these projects we'll emerge with templates that could be used by other crossing to set up a continuous measurement system.
J. Symoun:
Okay. The next question for you Crystal, can an MPO obtain the disaggrate data for a specific urban area?
C. Jones:
Maybe Dan wants to handle that one. He's more attuned with the issues of data sharing. Our goal is to share the data and go down to a level that is pertinent for a user. Dan might be the better person to answer that.
D. Murray:
And the two words we don't and can't use would be disaggregated and raw data. In our world that means uncleansed. But once the data is cleansed and aggregated across provides and carriers we're working with federal highway to develop sort of terms of use agreements to distribute unprocessed data. That means that our primary products these days are dashboards, maps, charts, et cetera. If somebody wants a file for a specific area our view is we'll get that to them in some fashion, disaggregated and raw, we'll never break it down by specific truck or carrier. After that everything is fair game.
J. Symoun:
Okay. Another question here, can you show us an actual buffer time index graph?
C. Jones:
If you could type in clarification. I could up load a file quickly. A graph that looks at the buffer for a particular corridor or crossing over time. It depends on whether or not on how do you derive the index? Or showing it for a particular facility or border crossing.
J. Symoun:
Maybe while we're waiting I will move on to the next question. Are you monitoring only long-haul trucks, or others like drayage, package delivery, and other LTL trucking?
D. Murray:
The answer is we have all of those sectors in our databases. The next question is how representative are those? It appears we're probably in the vicinity of 25% overrepresented in truckload. And an equivalent amount in LTL. But that's, that's local hub and spoke LTL. This is all painful acronyms for a lot of people. But the most powerful representation of our data is as Crystal said, city pairs and urban pass through. If somebody is looking for local pickup and delivery by Dan Murray quick drop, we'll not have much of that in there. It's usually large to medium companies, longer haul and city pairs are where we're strongest, but we have data from all over, it's just how robust is it?
J. Symoun:
Back to the other question. It's about how you derive the index.
C. Jones:
Okay, I think there's a slide there with the formula.
J. Symoun:
Let me bring back your presentation.
C. Jones:
What it does, it takes the worse 5% of your trips and groups them and comes up with a 95 percentile
J. Symoun:
I think that's the slide right now. We'll open up the phone lines in a minute. If you want to get on and clarify.
C. Jones:
That link that I gave on the last presentation is a report that was done for the Office of Operations, it actually talks about travel time and travel time reliability. The last link is a document that one could go to. This slide now is what we use to derive the buffer index for a particular interstate or segment that is included in our study effort. It's expressed in terms of a percentage. You basically apply that to the travel time to come up with the extra needed to guarantee on time arrival.
J. Symoun:
A question for Dan. Did you say a data file for a specific region is available and if it is, how can MPOs obtain that?
D. Murray:
That's a better question for Crystal. The short answer is almost as we speak we're formalizing a process, Federal Highway is formalizing a process where requests come into federal highway. Depending on the type of request and a series of other things it's prioritized and processed.
C. Jones:
I should have highlighted managing requests for information as a program challenge. In the most ideal situation we want to meet every request from every MPO and every DOT. We do have limited resources, unfortunately. We determine that the best approach getting to the point where DOTs and MPOs have maximum access is to develop the web tool. However, because we see responding to requests for information as a way to get targeted feedback from MPOs and DOTs, we do try and honor as many as possible. If it's a small request that can be supported using existing processes it will probably more likely to be supported than a more extensive requests such as "providing specialized data reports every month for a year". The model we're using is that requests are considered and prioritized based on how aligned they are with what we view as core program element (e.g. to support transportation planning). We will respond to all requests and provide the timeframe to turn around the product that you request. If it's beyond what we have resources to support, we do encourage that you could contact ATRI to see if there's a small scale project that you might be willing to set up an arrangement with ATRI. That was a long answer to a short question, but I hope it –is sufficient.
J. Symoun:
Thank you. Another question. Do you have enough observations to show truck speeds by time of day? One hour, four hour, eight hour blocks?
D. Murray:
We can do that. I think, Crystal, we've looked at those time increments and day of week. We all feel comfortable that two hour blocks, particularly peak hour is where we focus our time. 2 and 4 hour blocks, is that accurate?
C. Jones:
We don't tell the carriers when we want to be communicated with. If a particular fleet has it set up where they communicate with their vehicles every hour that's how we often we get the data. The data shows that the one hour mark is where we get the predominant amount of pings from the vehicles. We haven't tested it down to 15 minutes or a half hour.
D. Murray:
To take advantage of the question, you would want to separate satellite communication from terrestrial. We describe this as research. When you get into something that is more robust and more real-time you're moving into operations. We've all said we're not interested in real-time traffic operations, processing at this point. We're sort of looking at, um, planning opportunities, rather than traffic management opportunities.
J. Symoun:
How often does each truck report its position?
D. Murray:
The answer to that is, Crystal provided it, it could be four times an hour. It could be hundreds of times, nearly thousands of times an hour. It depends on a, on the technology and the characteristics. We literally have had to scale our data collection back.
C. Jones:
I guess just to add, our processing methods doesn't use anything less than 15 minutes or greater than 2 hours for the reliability measures. Doesn't mean that we don't have the capability. Thus far our method has analyzed using those parameters.
D. Murray:
The truck parking was sort of an epiphany. We reversed the algorithm. Now we have this, truck parking and mapping tool. After you reverse what Crystal said. If it's moving get rid of it, because it's not a parked truck. That's how that differs.
J. Symoun:
Okay. I thing we got through everything that is typed in. We will open up the phone lines.
Operator:
Thank you. At this time if you would like to ask a question over the phone, please press star and one on your phone. Please make sure to record your name so that you may be introduced. That's star and one on your touch-tone phone. Record your name so you may be introduced. Please hold a moment for the first question. At this time I'm showing no questions.
J. Symoun:
Okay. We have another question that was typed in. That is how accurate is the lat/long data (e.g. can you differentiate between freeway vs. frontage road?)
D. Murray:
The answer is we can now differentiate between highways and frontage roads. We're down to the sent meters with GPS. Early on Crystal, we were debating where the positions were, because of the frontage road issue. Now we use newer data to calibrate the older data. It's gotten better. As Crystal said, the data process improves on almost a monthly basis.
C. Jones:
I think right now, we use 1/4 of a mile from we define as a shape file for a roadway. We throw out anything that doesn't fall within a 1/4 of a mile. For the most part we think that our methods give us a high degree of assurance that these vehicles are on the interstate that we're working with. But -- that question has come up before. Again, as Dan said, we're evolving it. We try hard to include in our processing methodologies that reduce that possibility of getting data that is not on a particular interstate.
J. Symoun:
Are there any data about converting travel time and time reliability to cost measures, e.g., dollars?
C. Jones:
I think there's a project right now. They're getting GPS from tracker devices in Canada. They're converting speed and reliability into costs and dollars. The slide that showed our measures. That's what we want it look at. How can you use the average cost of delay into dollar figures? That's on our agenda.
D. Murray:
The biggest challenge we face there is coming up with the costing side. The industry is too large and complex. The per hour cost of trucks is between $24 and $150. We've been working on that with the government and the industry. We think in February we'll be releasing a report that stratifies, coming up with general averages, by hour and mile. At that point I think we would be more comfortable to multiply that by hours of delay.
J. Symoun:
Has anyone started to evaluate how those of us in Operations can start to provide any performance measures on freight movement for our Transportation Commissions?
If anyone wants to type in an answer, I encourage that as well.
C. Jones:
Certainly from the national perspective that is our goal. To evolve this to a point where we have a high enough degree of confidence that the data is of high enough quality for decision making.. To evolve to a data source that can inform decisions. I don't know if that answers your question. Again our goal is a creditable data source that can be used for decision making and inform the budget and investment process. And probably I would say, we are, best case, I would hope 6 to 12 months from being able to say that we endorse it for those purposes.
J. Symoun:
There was a clarification. Florida has a Transportation Commission appointed by the Governor and Cabinet and they keep asking for measures for CVO and I don't have any.
C. Jones:
We have a request we are working on for Florida. I'm not sure if that's the purpose. But we see state DOTs as our core customers. If there's a specific request that Mike or someone from Florida wants to send in I can consider it. It depends on what type of measure. We're most comfortable using speed and reliability. Something that fits into the framework of what we have established and what we have the highest degree of confidence in. Speed and reliability would be the limitations of what we would be able to provide and provide for that type of request. If there's a specific request that can be formulated we would entertain it. If it extends beyond what we have then we could work offline to see if ATRI would be able is to work with Florida.
J. Symoun:
We'll go ahead and close out for today, so everyone can head out and go home for the holiday. I want to thank you for attending. Thank you to Crystal and Dan for presenting. I was pleasantly surprised by the number of people we had on today's conference.
The recorded version will be available in the next two weeks. I will send an email when it is available. If you didn't register for the seminar in advance, send me an email if you'd like to receive the follow up information. The next seminar will be held on December 12 and is titled "Accounting for Freight Benefits in Transportation Projects." Please note that this seminar will be held on the second Wednesday of the month instead of the usual third Wednesday. If you haven't done so already, I encourage you to visit the Talking Freight Web Site and sign up for this seminar. The address is up on the slide on your screen. I also encourage you to join the Freight Planning LISTSERV if you have not already done so. Enjoy the rest of your day and have a happy Thanksgiving!
