View the December 2018 seminar recording
Jennifer Symoun
Good afternoon or good morning to those of you to the West. Welcome to the Talking Freight Seminar Series. My name is Jennifer Symoun and I will moderate today's seminar. Today's topic is Using Freight Data in the Proper Geographic Contexts: Challenges and Opportunities.
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.
Today's seminar will last 90 minutes, with 60 minutes allocated for the speakers, and the final 30 minutes for audience Question and Answer. If during the presentations you think of a question, you can type it into the chat area. Please make sure you send your question to "Everyone" and indicate which presenter your question is for. Presenters will be unable to answer your questions during their presentations, but I will start off the question and answer session with the questions typed into the chat box. We will also take questions over the phone if time allows and I will provide instructions on how to do so once we get to that point.
The PowerPoint presentations used during the seminar are available for download from the file download box in the lower right corner of your screen. The presentations will also be available online within the next few weeks, along with a recording and a transcript. I will notify all attendees once these materials are posted online.
Talking Freight seminars are eligible for 1.5 certification maintenance credits for AICP members. In order to obtain credit for today's seminar, you must have logged in with your first and last name or if you are attending with a group of people you must type your first and last name into the chat box.
PDH certificates are also available for Talking Freight seminars. To receive 1.5 PDH credits, you will need to fill out a form. Please see the link in the chat box. Certificates will be emailed one week after the seminar. A seminar agenda has been included in the file download box for those who need to submit an agenda to their licensing agency.
Finally, I encourage everyone to download the evaluation form from the file share box please also and submit this form to me after you have filled it out.
Today we'll have several presentations given by:
Our first presentation will be given by Bruce Lambert. Mr. Lambert has almost 30 years of active involvement in the freight industry, focusing on freight data and models, planning and policy. He has researched many different modes but has mostly focused on maritime and intermodal transportation. He previously worked as the Executive Director of the Institute for Trade and Transportation Studies; as a Senior Economist at the U.S. Army Corps of Engineers, Institute of Water Resources; for the Federal Highway Administration where he managed the Freight Analysis Framework and developed performance measures for trucking movements on the interstate system; as a Senior Economist at Standard and Poor's DRI, where he examined global maritime shipments and international trade forecasts; and at the Port of Long Beach as the Port's Trade Analyst.
Bruce Lambert
Thank you. When I approached Chip about doing a talking freight on the question about anytime you are doing a freight analysis, there is a question of geography. If you look at the first two graphs for the state of Mississippi, in one graph, it is showing their higher functional class, what they identified as their key state freight corridors. This is traditionally a corridor-based freight planning perspective. The second graph on the slide shows the net exports from the state of Mississippi. We talk about global markets, we talk about these type of traffic patterns, what markets are high, how much cargo goes to the various markets. When it comes time to doing projects, next slide, this is what we face. We see a street, a street that has trucks and people walking on it and things of that nature. And yet, if you go to the next slide, we are really talking about projects. If you notice between the two pictures, there were no trucks, there was no UPS parked on the curb, no railroads, no nothing. And yet, from a freight perspective, all of these levels of geography matter. International trade has one element. Key state freight corridors have one element, and even local geography. And all of those have different data needs and requirements. However, a lot of times when we think about freight, it does not have geography neatly wrapped up for analysis. For instance, if you go to the next slide, we see a comparison of two people; we see a guy drinking out of a fire hose, and then we see a woman sipping a cup of tea. In both cases, they are drinking water. In one of them, we can argue, that the water is flavored with tea, but we do not know what the fire hydrant was flavored with. But basically, we are trying to assess data and sometimes it feels like this; I am getting data out of a hose or am I getting it in the right portion. There is always, if I knew the right amount of water it would help my job, make my cup of coffee taste better, make my cup of tea taste better, and help my job get easier. In that context, the final slide is, I have put together some people that I think are going to help show us, you know, get a dialog started about freight data and freight geography. And, in any context, I think that when you have people like this and everyone is struggling with geography at different levels, all of them are talking about freight and different considerations, and it is really up to you to ask them and engage them and get as much as you can from the presentation. Thank you, Jennifer, I will turn it back over to you.
Jennifer Symoun
Thank you, Bruce. Our next presentations will be given by three people from the Florida Department of Transportation – Joel Worrell, Thomas Hill, and Holly Cohen.
Joel Worrell is the Transportation Data Inventory Manager with the Florida Department of Transportation, in his current responsibility he manages the state data programs that support federal and state data reporting requirements including the Roadway Characteristics Inventory program, the Highway Performance Monitoring System program, the Transportation Data Quality Management program, and the Multimodal Data System program. Joel has over 11 years of experience working with data, GIS, and technology in the public and private arenas that include government, military, and transportation sectors.
Thomas Hill is the State Modeling Manager for the Florida Department of Transportation. During his eighteen-year career he has worked in travel demand modeling from many perspectives including: local government, private industry, MPO and State level positions. Additionally, he is a guest lecturer, contributor and presenter at workshops, conferences and universities. Currently, Thomas is involved with updating the Florida Statewide Passenger and Freight Model, evaluating and contrasting evacuation throughput estimates for hard shoulder use verses contraflow lanes, implementing a statewide cordon line to measure travel speeds using Bluetooth technology, and developing a micro-simulation model for the Port Everglades Seaport.
Holly Cohen, AICP is the Freight & Rail Planning Administrator for the Florida Department of Transportation (FDOT). She has worked at as both an FDOT employee and a consultant for over nine years, contributing to FDOT's freight and modal initiatives including the Florida Freight Mobility and Trade Plan and Strategic Intermodal System (SIS). Holly also manages eight rail safety inspectors, as part of the Federal Railroad Administration (FRA) State Rail Safety Participation Program.
Joel Worrell
Thank you, Jennifer. Thank you, Bruce, for providing the intro. It goes right into this presentation of Florida's truck empty backhaul analysis. We have provided State DOT perspective on a specific truck and freight specific issue, and what we see is this is directly tied into geography. Florida is uniquely set to have a unique supply chain and freight demands being that it is a peninsula. We have about 20.5 million people, 112 million visitors, and we have a wide amount of multi-modal facilities between commercial airports, seaports, and 2 spaceports. What we provide is a set list of data and statistics that we utilize for plans, studies, and such. We incorporate all of this into all of our multimodal plans, our transportation plan, and we make sure that the data stands up for our decision-making.
We invest into maintaining the system through our multimodal data system program. We have all the plans and all the data that helps us produce these plans. We integrated that with our core platforms, RCI, the risk inventory, our highway performance monitoring system, our traffic database, as well as our linear reference system. We standardize to approach the data governance framework. And what we tried to strive for in this office is to make a customer-driven approach to our public and our department, and to really integrate our operations and planning offices who are dealing with freight issues. We make sure that the data and the policy is working together to make sure that we can make good decision-making with data.
For our project, we wanted to look at truck empty backhaul, which was an issue that we identified in our board of transportation freight plan and our Motor Carrier System plans. We couldn't estimate accurately what that number was until we actually started looking at the weigh-in motion data. Anecdotally, we put about 75% as an estimate of trucks leaving the state that are empty. We wanted to quantify that and make that a number that we can drive down and communicate to our trade coordinators, our stakeholders, and the people who are dealing with these freight issues on the ground.
To objectively do this, we had to define a methodology. We had to define truck empty backhaul and a definition that suits us, and we had to develop a methodology that supports that. We had to analyze the data, identified some influencing factors, and summarized some results that we could potentially acquire. We worked with our consultants to develop a really good analysis and a straight forward analysis just looking at raw data.
Our methodology, part of looking at all the weigh motion records, we classify those through FHWA scheme F codes, which focused on class IX as our primary class type that we wanted to focus on for the study. We set some parameters and the average weight established here as 40,000 and below were empty and the 60,000 or greater were full. These class IX trucks were floating between 40 and 60,000 and further assumptions were developed in the analysis to identify if the trucks could be cubed out or even partially emptied.
In this graphic, we analyze the varying axel weights of the 2nd to the 5th axles. What you see is weight distribution listed in percentages based on the overall axel weight loads. If weights were equally distributed throughout the axles and not over 60,000, we would assume that there was cargo that would have to be dispersed. If it was not distributed equally throughout the axles, and the percentage of axle weight of two and three were larger than compared to four and five, we would realize that it will be partially empty. So, you can see the two trucks, and each axle percentage that we focused on. Again, reiterating our methodology, we wanted to provide an overall flowchart for people to understand the analysis, we outlined the decision-making for these trucks.
So, as we prepared the data, we looked at over 100 million records over a course of a couple of years. We filtered out for Class V and above. We used our processing software to analyze the trucks that are being identified through the weigh-in motion site. You can see that we have positions across the state border, and to other sites, about 30 overall. As we do that analysis, the first thing that you will see are the full trucks. I'm sorry, the empty trucks. So, the empty trucks here -- and I want you to focus on the three sites that are in red -- and what these percentages are, are of the Class IX that are empty, we factored that as the average for all trucks and that category. So, for all empty trucks and all trucks at that location, we got the percentages. For the empty trucks, we estimated about 50-30% for the main three quarters, on I-10, I-75 and I-95. Approximately 15-20% are entering Florida empty. So, for full trucks, we looked at the percentages again to see what Class 9's are in that category that we identified, and we can see here that on I-10, I-75, and I-95 there is a validation of this trade imbalance. So, anything in these percentages is kind of leaning toward the point that there is a large amount of trade coming into the state and there is not a lot going out.
So additionally, cubed out trucks were analyzed as well. So, the class 9's that fell between 40-60,000 lbs. that had distributed weight. Around 21-22% of class 9's coming into the state were seen to be cubed out, and about 14-16% of class 9's leaving the state were seen to be cubed out. And finally, this other category we analyzed was the partially empty. The team classified again within those parameters of 40-60,000 lbs. with uneven weight distribution. Those numbers in trucks were considerably less, at about 10%. So, we see that could potentially be the less -- private trucks returning with empty pallets and so forth. So, some of the factors that we come back to, as I mentioned earlier, it is the 3rd most populated state in the nation. Our geography is not a regional hub, it is a peninsula, and we deal with a lot of visitors coming into the state and we expect a lot more for the Christmas season. Some of the solutions that our consultants identified may help correct the number of imbalances we are experiencing. If we had more freight coming in to our ports and we were able to assist with getting that freight out more efficiently, we could attract more businesses and we can track more movement.
And finally, the analysis recommendations that we looked to work on internally is that we know we need to analyze more data. We want to make sure we're working with other data sets to really drill down to what commodities we can focus on that is just not available immediately to the department. We want to bring that public data and that private data together. In our role the public is essentially at the mercy of the freighting logistics business. Further analysis of these commodities is really of interest, and what ship supply chain can be supported. Delving into class information is what we want to do next. We want to look at some statistics and some truck types. And, essentially, we could do our own commodity surveys.
Thomas Hill
Thomas Hill, Florida Department of Transportation. There is a lot of information I am going to go over with the next two projects. And we're just going to skim over the top of these. I just want to note, though, that both of these projects are listed online on our modeling website, FSUTMSOnline.net. I'll give you the address at the last page of this presentation. So, we'll delve on into this.
So, today's conversation. The first project we have is measuring our economic impact of freight transportation based on FreightSIM. What we wanted to do was look at our investments into freight infrastructure and quantify the economic impact. Typically what we do when we measure improvement into infrastructure we look at VHT VMT type of variables, and we wanted to convert this into something more tangible in terms of economic outputs. I'm going to walk you through an example with the Port of Jacksonville. And then our second project discussion is the Freight Data Fusion project where we merged FAF data and Transearch data. And really this is to look at future year FAF data sets and be able to break this data set down into more finite resolutions.
So, our freight economic impact tool. We use this to perform a regional economic impact analysis. We have a statewide freight model, which we call FreightSIM, and we use the output from the post-processing tool to derive our economic analysis. The core of this is the freight demand combined with the multi-factoral economic model. In this case we used implant to give us economic indicators. And so, we use of the outputs in the model to convert those two monetary values. And then use the economic tool to distribute the economic values to different economic sectors within the state, specific to each area of the state. So, in the Port of Jacksonville, we conducted a scenario in which we looked at increasing capacity by 30% for the Port. Currently, this port serves 23.3 million tons of cargo annually, adding 30% we were estimating 33 million tons. For the topic of this conversation, we assume that the divergence of the freight trucks were diverted to the port, which makes this conversation a little easier. And then conducted analysis for the state and county level areas for 2017 and for 2035. So, looking here at the table, we can look at our VHT savings from a port with no change compared to a port with an increased capacity. And then measure our vehicle hours changed. That's the difference between our base and our improved scenarios.
So, the next thing we would do, we know in Florida, with the value of time is for freight movement. In the Jacksonville area, it is $23 an hour based off the stated preference survey conducted across the state and annualizing the value of time by 365 and the times VHT savings. We did this at the statewide level and at the county level. And in order to get a monetary value for future years, we also compounded a net present value so that we can grow that economic increase over time. So, the economic impact analysis. Again, we went to IMPLAN to get a description of what the economy looks like, but all regions in the state of Florida. And then we can apply the increases in VHT and conversion to the monetary value to each one of the classification systems and the ICS at the two-digit level. With freight and output plus the input output of the economic model that would drive from IMPLAN. In the boxes down here from the statewide savings for example, we know that we can estimate that we have an increase of 364 workers, a resulting $16 million impact increase over a 20-year period. We also did that at the countywide level. So then the next step in the tool is to break this down and say, "What does that mean to the local economy?" For the local economy, we can look at each of our sectors within the state and say from agricultural purpose, or from a mining perspective, we can derive the sectoral output or the benefit by employment category. Who is going to benefit the most by the economic investment into a freight modification or in this case, a freight port? This is at the statewide level, and we also derive to this at a countywide level, so two different perspectives at looking at the same improvement.
For our next project, this was combining FAF data and Transearch data fusion. And this was to look at how do we move our FreightSIM freight model data forward whenever there are new advances, new releases of FAF data. This comes from a large aggregate data that we have, five economic regions of the state of Florida. It is hard to apply that to the statewide model on a very fine resolution. Typical processes to look at density of the population and use those densities of population to disaggregate fact, we really did not think that that was the right approach. We thought that we could get a better look at it, particularly because we have Transearch data in-house. So how can we better use that to define future year FAF values?
Part of working this is understanding the differences between FAF and Transearch. FAF is a derivative of the Commodity Flow Survey that is conducted nationally. And the data then is provided to us and 43 state commodity types, standardized by the Standard Classification of Transportation Goods. It comes to us at a very coarse spatial resolution, and includes our forecast for 2015 and 2045, at five-year intervals. Compared then to Transearch. Transearch is a proprietary product, they are deriving their information from public sources, but really, from private sources; from people that contract with IHS Global and provided their direct shipping information and how they compile the data is a something that we don't know. They would not give us, we did ask the question, but they laughed at us. The flow from Transearch are provided in 500 different categories by the STCC, the Standard Transportation Commodity Code, but was really nice fine spatial resolution. They give it to us at the county level. We worked with IHS Global and incidentally, HIS Global makes both FAF and Transearch and data. But really Transearch is very expensive to acquire. It is hard to keep going upstairs and saying I need a lot of money for data. If we do have data for like 2040. So, FAF and Transearch both provider us annual commodity flows in the U.S. But the variability in the collection mechanisms are different and that relies again looking at the data from other sources, private sources and public sources. The variability is in the representation of the commodity flows. FAF represents actual flows on the network, while Transearch looks at the flow of commodity between one region and another region. And then you compare the tons. FAF reports two times more tons of commodity than Transearch. Even though, at the end of the day, they are kind of turning the same thing, the issue is that the intermediate stop in the logistics chain, which is the difference between FAF and Transearch. So how do you normalize all of this is the idea.
Both flows provide production consumption trends at a county level. So if we could find the path for these production consumption flows on the network then we can generate something similar to FAF flow. Something similar to Transearch. We can do this at the county level. But really, aggregating the data then, we can generate the FAF and send it directly to FAF flows. That is where the algorithm comes in, as the algorithm is trying to normalize the differences between FAF and Transearch. Let me see if I can pull that up. The challenge is how to link the two data together. Looking at Transearch, if I am going to pass all of the stuff, you will end up with a large number of paths because of the classification codes. So, we only considered direct path in our analyses, and one being intermediate County stop along the way. So, this is really where a lot of math happens. What this boils down to is in the bottom right-hand corner; this is the gold standard. This is essentially looking at the log likelihood of commodities from Transearch reports moving between one I and one J plus the same log likelihood of FAF, then normalized out by all I's and J's. This is work that was produced by Dr. Levine and a very talented team of mathematicians and engineers at the University of Central Florida and the University of South Florida. If over your holiday break you are bored with the family and you want something exciting to read, this is probably not it, but I would still recommend reading it. Absolutely brilliant work. So, what does that really mean, when converting this then to English?
The commodity types reported are based on two different classification systems. In order to normalize all this, we need to combine everything into similar commodity types. We were able to identify 13 commodity types and then classify Transearch flows by millions of tons. Or millions of tons within the state of Florida and external to the state of Florida. So, we could pull out two; agricultural products and food products. Take a look at the statistics, the flows of county to county flows, and compare that to what our estimated county level flows are. We had this for food products, and, to say what is statistically viable. This is much easier to look at graphically. So, looking at the next couple of slides, Transearch, which is what we are comparing to on the left, and on the right is our estimated flow. This is representing flows originating from Florida to all other areas. So, Transearch will show this to the left and our estimated flows on the right. These are the flows for agricultural products. And, looking again for food products originating from Florida to all other destinations, we are showing some good representation between what we are able to do compared back to the Transearch flows. Again, then, this is comparing an origin from within Florida to all other Counties. So, kind of the same thing looking at a different comparison. And do this comparison between for all 13 different commodity types, again, showing pretty good representation comparison transfers to our flows. And, further looking at the data, this is flows coming into Florida, having an origin outside of Florida and arriving in Florida for agricultural products. And for food products as well. Again, showing fairly good representation particularly in our most populated areas in the state of Florida.
And then, looking at pathing. We did not want to just look at what were the origins of the destinations, what is also the pathing. This is looking at the port of Miami and the distribution of product from the port of Miami to all areas within the state for agricultural products. And again, we know that based on some of our surveys and our food data information, about 80% of products that come out of the port of Miami have a destination within the South Florida regions. Transearch shows up that clearly. Our fuse data is doing a good job of representing and replicating that data. And, looking at food products, pretty good representation. A bit of an anomaly for Northwest Florida, but statistically it is still viable.
Again, if you want to know more about these projects, both of these reports are listed on FSUTMSOnline.net, and we also have several presentations provided there as well if you wanted to condense all of that into one thing. I think that that is me.
Holly Cohen
Continuing on where they left off, good afternoon, I am Holly Cohen, Freight and Rail Planning Administrator at FDOT. And to close out FDOT'S presentation today I want to highlight how our freight and multimodal presentation that it supports not only data-driven planning and decision making but outreach efforts as well. And that there are unique challenges associated with each area.
So, the strategic focus of the freight and multimodal operations office involves removing institutional infrastructure and funding bottlenecks to build a well-connected reliable and safe multimodal network. Joel went over the program goals and how data integrates with offices throughout the department and the state. This ties into our efforts as many of the plans and studies he referenced are led by our office. That includes the Freight and Multimodal Trade Plans, the Modal Care System Plan, and the Rail System Plan. And our consistent analysis and reporting of multimodal data is critical to each of these efforts and leads to continuity. Thomas also gave an example of analysis applications of freight data which are equally important to our planning and outreach efforts. Not being a quantifiable measure of freight transportation may help us better understand how investments may shape us in the future. And efforts to create an analysis by work are best in the Transearch data fusion allows us to estimate commodity flow trends with greater geographic specificity than with the grafted alone. Essentially, by working hand-in-hand with other FDOT offices and our partner state agencies and associations, we can better tell the freight story of how goods movement matters, and that is what our office champions.
The county freight overview is an example of how FDOT has used freight data for outreach. A few years ago, we teamed up with partner state agencies to create a set of brochures on Florida's freight infrastructure and trends. Essentially, by combining information on infrastructure from FDOT, top import-export trading partners, and employers from enterprise Florida and etc., there was a concise resource for telling the freight story specific to each of Florida's counties. And the overviews were developed with graphics and maps to be easily understood by the industry, city and country leadership, and the general public.
This is an overview of the objectives on the slides, and a quick snapshot of one of the pages of some of the transportation infrastructure, and import-export information included. These concise overviews were 8.5x11 pages with statistics and interesting facts about the county and state level, Guide level information on growing industries, largest employers and multimodal price facilities to help counties understand their local assets. Additional information on county level top commodity types for imports and exports, and which counties were the top trading partners for both imports and exports. Both counties understand how freight moves in their backyards. The contact information for each agency further showcased this partnership and make sure that the local teams were aware of each of these agencies and the functions. There is another snapshot of one of the pages with an overview of some of the additional information from specifically Hernando County. Each county freight overview also included a two-page spread map of multimodal freight infrastructure and the largest employers. It gives you an idea, visually, of what information is available in the area. For example, this map of Duval County had the Jax airport, the system highways, rail freight terminals, rail lines, and your military installations in the gold.
How are these resources received? Essentially, we received feedback from locals that the overviews were really helpful and that economic development groups and chambers, especially, were using them to showcase local freight assets. We received a ton of requests for updates and we did a lot of spot changes into airports and businesses within changes, so that told us the brochures continued to be used. As these products became dated, we considered if a full update was worthwhile. And essentially, we cannot justify the value an additional large purchase of the freight data such as Transearch. As Thomas mentioned previously, it is expensive, and it does not cover the same geography. Much of the trade partner data and top employment options can be updated with similar source lists, but we cannot do a full update. More recently, we look at revisiting the resources to showcase new insights in the commodity flows. By changing the content to highlight these new data sources and approaches, we can fund pioneering over the last few years and have even more search and analysis efforts. If we can identify the needs using the empty backhaul research, and use the data for commodity haul, we can update these types of brochures more often allowing us to provide valuable resources for the NPO's, Counties, EDO's that don't have the capability to create their own, and to better identify changes, trends, and successes.
In summary, FDOT sees the need to collect and analyze freight data to make decisions and to share insights on the regional level. The data is not always available or it has challenges associated. To meet these needs, we are pioneering new methods, measuring truck empty backhaul to performing regional economic impact analysis, to combine data resources that estimate commodity flows at the local level, and engage in support of the local level to better understand how freight moves. With that, I will turn it back over to Jennifer.
Jennifer Symoun
Thank you, Joel, Thomas, and Holly. Our next presentation will be given by Erik Johnson, the Virginia Department of Transportation's Freight Planning Specialist. Erik has served in that capacity for most of his twenty-two years with VDOT. His responsibilities include managing and/or providing support to freight-related statewide studies and represents VDOT for the freight planning functions of Virginia's multimodal planning efforts as well as several multistate corridor coalition groups. He is a graduate of I-95 Corridor Coalition's inaugural Freight Academy.
Erik Johnson
Thank you, Jennifer. Good afternoon. For all of the subject matter in freight experts, my presentation is only going to scratch the surface on a technical level. More because what I will be sharing is not technically complicated, but very simple. From the state D.O.T planning perspective, there are two avenues of thought in play. 1) Data tells a story from many perspectives you want to hear providing context. They make for a good narrative in terms of providing a backdrop for validating a transportation plan and the tool for persuasion. And 2) You can also measure performance requirements or targets highlighting specific needs, identify or solve problems and answer questions. There is more to say about what data we are using and how we at the Commonwealth of Virginia are using analytical freight data that I will discuss today. I will talk about our most important data, some of the purposes we use them for highlighting the ongoing effort of our three-year-old prioritized multimodal transportation program called Smart Scale, and our most recent truck parking study. And then touch on some of the challenges we encounter; and opportunities for creative problem-solving.
Commodity flow data and economic data are a significant part of our statewide freight program, particularly our long-range plan, VTrans. We use them for providing the geographic and economic context for the plan, identifying the characteristics of freight movement in the state, how freight moves on our Corridors of Statewide Significance, describe the employment or commodity sectors that are freight bearing in order to frame the economic significance of freight to make connections for the public so they can see how freight transportation and the planned improvements serve their personal and commercial interest. Commodity flow data is also important for weighing public investments in rail improvements to make the most impact in reducing highway congestion. This is done by analyzing the commodity flow data through a specific geographic lens, filtering through specific commodities and distances. If I had more time, I would have recruited our consultants to share insights and methodologies on one of our several I-81 corridor truck-rail diversion studies. As Florida mentioned, FAF and Transearch provide different views of the commodity flow survey, which I will not go into. Transearch has provided a useful view on many levels and the cost is commensurate. The project managers for the corridor EIS, or Environmental Impact Statement projects, had come to me regularly to acquire Transearch to fill out the freight component of their projects. It should not go without saying that our traffic count program is essential to track traffic volume trends. And, finally, we have been developing geographic information on freight generators. The current target is manufacturing warehouses and distribution center locations. In the future I would like to expand to the retail sector. This layer is like some of the proprietary business information sources which, unfortunately, can provide lat/longs on office buildings and homes. Our foundational source is aerial maps: ours and other online maps. I started this project in order to increase my awareness of freight generating industries; to correct what I was seeing on the other sources; to go beyond the highlight maps we have been providing in earlier VTrans efforts. It evolved through brainstorming; as a way to identify projects that acutely address freight needs by proximity--a concept that has been incorporated into our Smart Scale Program, which I will describe in a minute. Finally, we concluded a few years ago our keystone Truck Parking Study. And though the effort was significant and hugely helpful in elevating the ever-increasing need for safety rest parking for our over-the-road motor carriers, the effort of identifying the best locations for new parking areas goes on.
Let me move on to the most significant part of my presentation. Smart Scale is the name given to a data-driven transportation program that was birthed through legislation in our General Assembly a few years ago. House Bill 2, or HB2, until North Carolina passed a house bill 2 that surpassed us in fame. Though, Smart Scale is actually very fitting. I will just touch on a small portion of the program for this webinar. If you would like to look further into the rest of the program, there is a lot of information on our website shown on the bottom of the slide.
To give you a quick overview, there are six factor areas, each with more than one measure and a score weighting: safety, congestion mitigation, accessibility, environmental quality, economic development, and land-use. Then the scores are weighted again for regional context in our four factor categories beginning with the densest, most populated areas of the state and ending with the most rural. Now I'm going to discuss my area of the Smart Scale: Economic Development, Intermodal Access and Efficiency.
The measure, Economic Development–Intermodal Access and Efficiency, is designed to reward products that serve access to freight generating operations and ports: in essence, the last mile; projects that are on the STAA truck network; and are scaled by the tons estimated for the project facility. A maximum score of six is available to the project–times the tons factor. For example, projects that provide direct access to freight generators within a mile from the project, a score of 2 is tallied. If the freight generator is more than one mile but less than three miles, they received one point. If no facility lies within a three-mile margin, no points are awarded for this measure. Similarly, for if an STAA facility is being approved, a score of 2 is awarded. If the project route accesses the STAA, then it awarded 1 point. Projects providing access to one of Virginia's ports or select commercial airports are awarded points like the freight generators. Two points for being within the mile, one point within being three, and no points if over three miles. Finally, Smart Scale provides an innovative way to utilize Transearch. Given that the model is not entirely reliable on the project level, we divided the segment totals by the provided unit counts to estimate segment tons per truck, or truck density; and multiply that by the truck volume reported from our count program from the most recent year. The final mathematical scaling is designed to provide a number that fits in the context of all six factor areas.
Just to give you a picture of what it looks like. These are a large portion of this year's project locations; and this is Round 3 of Smart Scale. As you can see, project applications are coming from all over the state with a concentration of projects in the densest metropolitan areas of Hampton Roads, Richmond, and Northern Virginia. These are the freight generators: the warehouse, distribution and manufacturers across the commonwealth of Virginia. These are the ports of Virginia and the select commercial airports. This is our national truck network, the STAA: federal, state, and approved access routes. This is what it looks like together.
Just to give you a picture as a reference, the project is highlighted at the center of the circles. The small orange square just outside the red circle is a distribution facility. This is on I-81 in the north at the border of West Virginia. The red circle is the one-mile radius, and the orange circle is the three-mile radius. An example of how the scoring works is illustrated here; the project is in orange at the bottom. The red outlined area is a group of freight generators. The purple outlined area is the Norfolk international Airport. Both the airport and the freight generators are in the one-mile radius. The project is part of the STAA network. This project hits all six points. Note, we are not measuring to building or a driveway, we are measuring to the property line or the boundary of the smallest census block. The initial score is determined by an arc GIS tool and checked by a human manual with an arc GIS. That was a quick overview.
On the truck parking, I do not have a lot of time, so I will just brush on the truck parking study we completed in 2014 and published in 2015. By the way, there is a hot link here, and still available at VDOT.virginia.gov under our projects and studies [page]. We set out to investigate what the practice of truck drivers was by surveying state troopers, rest area staff, truck stops, VDOT residency staff, and the truck drivers themselves. Our surveys were designed to provide geographic results. Providing maps like this heat map. I'm sorry, let me go back. Providing maps like this heat map, highlighted apparent densities of trucks congregated at our interchange ramps, rest area ramps, and other noted locations. Armed with the heat maps and a mathematical model shared by Pennsylvania D.O.T, we came up with this, a detailed map that highlights significant truck parking gaps all over the Commonwealth. Short of getting an expanded truck parking safety program, we have set the floodlights on the subject and brought funds to truck parking information at VDOT.
I would like to highlight a few challenges associated with these projects and a few opportunities we have begun to work on which may provide improvements. Number one, commodity flow data models become unreliable at the project-level resolution. I realize assembling a national freight model is an enormous undertaking and I need to do my part to enhance that part that is for my state. Second, the networks used by these models need regular updating and cleaning. As we add to the network on the ground, we need to ensure that the model network is also updated. I have known for years that the network that we were using from Transearch is from 1996. I realize that I will need to be involved in getting the network updated for my state. We have already started to do that and plan to review the routing structure in the near future. The effort will hopefully provide lasting benefits to the accuracy of our project scoring program along with the planning reporting. Economic and business data are often attributed to headquarter offices, including homes, instead of the actual generator of freight traffic. These make beautiful maps. However, you need to understand what you are getting. This is primarily why I have taken all the effort of inspecting the aerial maps to create a more accurate layer of freight generators. On the truck parking area, the GIS location data, like StreetLight data, may provide a significant insight into truck parking needs. This could provide more accurate view of location and frequency that could increase confidence and we are finding better solutions. That is the conclusion of my presentation.
Jennifer Symoun
Thank you, Erik. Our final presentation will be given jointly by Michael Brown and Chandler Duncan of Metro Analytics.
Mike Brown is a Utah-based traffic engineer and certified planner who specializes in strategies to reduce congestion in ways that are good for business, people, and deliveries. Mike is the founder and president of Metro Analytics - a firm that focuses on the business case for infrastructure through designs that promote land use stability. Metro Analytics has an impressive array of freight and economic-focused professionals, and Mike is pleased to present some of their ongoing work in Cargo Oriented Development.
Chandler Duncan is Vice President of Metro Analytics, and a freight and economics specialist based in North Carolina. Chandler has conducted hundreds of economic studies for clients in nearly every state to determine the return on investment and societal benefits of various options for infrastructure spending. Chandler is currently the Principal Investigator for a TRB effort to define "Right-Sizing" across all modes and economic situations, and to provide analysis tools for right-sizing decisions. Chandler is an expert in using data to locate both nationally and regionally significant economic clusters so that the focus of infrastructure spending is well-targeted to support our most valuable locations.
Chandler Duncan
Thank you, this is Chandler Duncan. For our final presentation, we look through a lot of a big models and big data sources and things that people have been working on for quite some time. And one of the things that Mike and I wanted to make part of the presentation today was, when you get to that level of urban geography and you get into town or even the sub area, how do you use this information? How is all of this information boiled down to actually developing infrastructure and developing policies that will be conducive to making the best possible business environment? When you get into the urban environment, the entire planning context changes in a lot of ways and you start moving from data to actual design. And what does the data really tell you? That is the focus of today's presentation. How to use freight data to create an urban environment that is prosperous and sustainable?
To do that, we work with our friends at the Center for Neighborhood Technology, who I do not think are here. But they have, in the last century, developed a concept called cargo-oriented development which actually complements the idea of transit-oriented development that a lot of communities have embraced. And what does that mean for freight? Can you have urban design and infrastructure that is actually friendly to freight and industrial needs, and that balances that with concerns of livability? It considers the full range of urban performance areas in both economic and transportation, and even environmental areas like land-use, safety, and environmental measures. It also addresses issues that often are overlooked in freight planning, or that livability often does not look at freight. Like, when you look at complete streets and all the livability literature you see, you see beautiful pictures where people are eating biscotti and drinking lattes and it is beautiful. But where did that biscotti come from? How did get it into the shop? Because on the whole sidewalk you do not see a truck. The question gets to, how could you make a livable city? How do you balance industrial household and personal needs for both state and infrastructure? This a triangle, created by the Neighborhood for technology, looks at the system performance as a baseline on which you are developing in an urban environment, leading up to then supporting the cost-benefit effectiveness and the efficiency of how you are using infrastructure, leading up to the opportunity for regional economic development and ultimately livability at the highest possible standard of living. This is what we try to accomplish when we go into the urban environment and we are looking at freight.
The cargo-oriented development metrics include typical metrics of safety and environmental air quality, a lot of regular transportation metrics. But you also might have local economic development metrics that you get from developers or development organizations, freight system efficiency metrics that you might be able to get from shippers and carriers both operating in the environment, and other types of data about land-use that you might get from a city assessor or a city planning and zoning department. You tend to be using an all sources approach when you get into the urban level of geography. Urban geography, as we said, entails going from data to design. It entails getting beyond if stuff coming from China, what is the point of entry, and what share of it is a particular commodity? And, getting down into, where do I need to go? Where in town should I go to actually make my investment and improving the freight environment? What should I do when I get there? What kind of information will tell me what I can do that freight environment? It is often a much more granular and much less broad types of data source that we look at. There are three data views that we find looking at a city that is very helpful. One is the spatial a quantitative view. Another is a top-down view, which is kind of a GIS locational view. And another is cross-sectional. For example, the baseline quantitative view is where we go and we look at traffic patterns and destinations, but we are looking at saying, how many businesses and people, how many workers? How many resources are available? In this case, you compare a city in 2016 to what the congested future of the city may look like with respect to a particular site. And you can see some information about the households and the businesses, and how easy or hard it is to get there. That is the spatial a quantitative way that we might look at an urban freight system. Also, we can look from the top down. This example, coming from the Center for Neighborhood Technology, looks at Chicago. We can look and see where the industrial opportunities are in yellow. But where are they in relation to the different types of freight facilities? You can see the freight modal infrastructure is overlaid with the available development site to get an idea of how to use that information, that spatial data, to hone in on where we are going. But also, it is important to think about also the cross-sectional view of dat. To start saying, how much do we know about our streets? How are we using every foot in our right of way? How are we balancing things like sidewalks or housing or livability or beautification? How are rebalancing that with the cross docking at the loading and unloading requirements of our deliveries or other freight activity? So, the cross-sectional view comes relevant when you get into designing and delivering a project for the urban environment. So, then the question is, once again, how do we make the best and highest use of this information? When we look at the urban setting and we say that here are some areas where there are opportunities for freight, what do we do when we get there and what does the data tell us about what we should be doing? Mike has some interesting examples from a couple places where we see this happening.
Mike Brown
Okay, thank you Chandler. What I have here is a photo of the pre-World War II type of business environment. A time when business was thriving, multimodal, the type of place that a lot of urban planners are trying to re-create right now. But, in that thriving environment, I managed to find one person that does not really like thriving in business environments. So, I decided I ought to follow his career to see what he was up to. Let me see if I can tell with this guy is going to do. I followed him along and he decided to become a traffic engineer, because he figured as a traffic engineer, he could create huge messes that are terrible situations. This is the type of thing he started to do. He created these highways that become lined with all kinds of haphazard developments and end up being too fast during parts of the day. But then other parts of the day, they are far too slow, too congested, terrible. He could not have devised a better strategy to mess up places and make them so that they do not thrive so well. So how do we fix that kind of a thing? I want to walk through a scenario of how to take Chandler's idea of data and apply it to design.
Here's a situation very common to industrial areas. They usually access a freeway interchange, and that interchange becomes overwhelmed with too much traffic. The yellow and red shows the left turns and there is a just too much for the traffic signals to do. They become very complex and overly congested. Here is taking a look at that area in the aerial view, this is a location in Atlanta that we are looking at. One of the things that you could do to help revise this space and make it more valuable for business and general livability across the board, one thing to do is if there are already frontage roads, you convert those frontage roads into one-way frontage roads, or install them if they are not already there. If you do that, one of the first things that can happen is that this additional pathway then has access to the freeway, which relieves the previous arterial quite a bit. So, if you were going to do that and connect another roadway, then another thing to investigate is why not consider one-way flow through here at the same time. With one-way traffic you can get better efficiency and you can do it with less right-of-way footprint. So, if that was the one-way couplet scenario, you would need to have some cross-street connections to get from one to the other. But those cross street opportunities and the general better accessibility would help activate additional parcels in there, making this a more valuable place in general. You also have the previous arterial, which is no longer carrying a heavy load, so this is something that we can look at with that.
So back here, today view of the interchange, I wanted to show you a design modification and how we can use traffic data to analyze the before and after scenario. With the view on the right, we have the one-way frontage roads that are crossing with one-way arterial east and west, and the left turns are made as a square about kind of thing. That is a highly efficient system. This is another location in Atlanta that is very similar to that, where you have a diamond interchange before and the amount of congestion in this case goes up to more than four minutes per vehicle. But, when you reconfigure it in that crossing one-way pairs like that, and you have four different signal intersections, but each one has a maximum delay of 30 seconds. So, the great majority of trips will get through here in less than one-minute delay, relative to 4 before. That is something that improves the general accessibility for freight and all kinds of things and helps creates value in real estate and business value in the space.
So back then to this view of, all right, if we have cured some of the congestion and relocated some of the traffic, what do we do with that previous arterial? Here is a view of one of the options that you can do. Before it was just an ugly street, plenty of pavement, not much room for sidewalk or anything. But now with this reinvention, you can have on street angled parking, so some of your employee parking needs can be moved off to on street. Because it is a heavily industrial area, we may still want to make sure that curb radius and other sorts of features are accommodating of trucks for those times when trucks really need to be on that street for some reason. But another thing you can do is start to infill and create livable and walkable environment and additional services that are not usually common in industrial areas. This is a view of how that might occur.
There are a lot of wide setbacks on the arterial. Some of these industrial buildings, you could start to build on those wide setbacks and create retail opportunities and offices, front of the building offices, and maybe even apartments, so that some of the workers in these freight areas have a chance to live close to work and not need to own a car in order to get to work. That is the type of thing that you have, and with that sort of design approach, you cannot stop business from coming, and it melted even the Grinch's heart, so he is a happy guy. A reformed traffic engineer, finally. Here are a few strategies that we do have better analytics that improve traffic flow and can be good for both places making and freight flow at the same time. Here are a few resources that were used in this presentation, one of them was streetplan.net, that was a cross-section tool that our company produces, and it is free. You go there and build yourself a cross-section. And, the Metro Scape tool, go ahead and get a hold of this if you are interested in any of this.
Some of those links are not coming up in the right color, if you have a PDF version of the presentation available, you should be able to see a couple more links to get to some of these live tools.
Jennifer Symoun
Thank you, Mike and Chandler. We have a little bit of time remaining, I know that we have a number of questions that came in so we will start going to the questions. We will see how quickly we can get through them, so if the presenters can try to be brief with their answers. We will start with the questions for Florida. Joel, on some of your slides, there is a question on why the percents are not equal to 100.
Joel Worrell
When we did the study, we focused on looking at the full number of class IX that fell within the certain categories. We have four categories (full, empty, partially empty, and cubed out). So we took the percentages of that we were, that you saw the screen, and we looked at finding a number that fits that category and divided that by the entire amount of the state. So, the site-specific numbers you saw will not add up to 100. It's a percentage of the entire holding state.
Jennifer Symoun
Thank you. Another question for you, just to clarify, it looks like from your slides, there are a lot of empty trucks that depart Florida after making their inbound deliveries?
Joel Worrell
That is what we are seeing. There is a large imbalance between this phenomenon of trucks coming into Florida, dropping off their load, and leaving. It is not as bad as what we thought originally, what we saw in the anecdotal information that we were first putting out there. The assumptions we are looking at the data were 75%, we saw a lot less with the observed data.
Jennifer Symoun
How do the results compare to VIUS survey estimates of empty trucks for Florida?
Joel Worrell
We did not compare with the vehicle information use survey. But we did look at reasonable checks with transition staff and we thought some of the numbers looked pretty close. We compared the Transearch 2011 information, and that was not provided here, but what we can do is share the link of our study, which is on our website, and we can provide that to our users.
Jennifer Symoun
What industry data do you use?
Joel Worrell
I am not sure if this is pertaining to some additional data sets that we compared to. I mentioned that we use the FAF and Transearch. But we really use the FDOT data so any information that did not go into the presentation. It's basically the DOT's way of motion sensors that deployed across the state.
Bruce Lambert
Jennifer, I want to follow up on that, Joel did a great presentation for the ITTS member states earlier this year. That video was captured on the ITTS YouTube channel. If somebody wants to search it, they can hear him talk for 27 minutes on this study.
Jennifer Symoun
All right, thank you. Thomas, another question, can you share a link with the VOT study?
Thomas Hill
The VOT study was not compiled into a report that is in a releasable form. It was derived from state a preference of surveys, but not really compiled for a report.
Jennifer Symoun
All right. Another question, with an increase of 30% in port capacity, would that increase total freight demand and therefore increase total truck traffic on roadway network for local drayage to the port and total truck VMT compared to baseline scenario? Could you explain where the VHT reduction came from?
Thomas Hill
I knew somebody would call me out on that one. The port location probably should have been more down in the Peninsula in Tampa or Miami in that case. The question is right at a point there is well taken in that we use that port and for purposes of discussion. Let's assume that the freight was not being consumed by trucks, that way, it made it easier to talk through the example. The idea wasn't to talk about the goals and output of the tools. The question is on target and maybe I should have chosen a different location and showed that, because we import a lot of commodities in the state of Florida, it doesn't long distance trips. It would have probably borne out better in the discussion, but, if I were to do this again, I would use a different example.
Jennifer Symoun
How do you apply those flows to the network? What level of network have you been able to go down to? (State highways? county highways; local roadways?)
Thomas Hill
In the Florida Statewide model, the freight and components, we have the state highway plus all major local roads. Our resolutions are pretty tight. And we do not use a here network or anything like that. But, we do have a fine grade resolution. The entire model into the system was set up so that it could be pulled out of the statewide model and incorporated directly into the regional models, which have a much finer roadway network than the statewide model.
Jennifer Symoun
What commodity correspondence table (STCC vs. SCTG) was used to consolidate the different commodity types? Is it also available online?
Thomas Hill
The consolidation was conducted in-house. So using both tables and essentially broadening the third table.
Jennifer Symoun
How is the Transearch data compared to FAF 4 data? Do the total tonnage by freight mode match with each other when aggregated at the statewide level?
Thomas Hill
The difference between the two is that FAF is almost twice as much. It was not the comparison. We could not use that comparison because of the way that the data is reported for both Transearch and FAF. So what the project does is it looks at the likelihood of flows between one area and another area. That is the gist of the algorithm, if you will. Looking at the likelihood and not so much the tonnage. At the end of the day, you try to report the tonnage, but at the aggregate level it is a flow comparison. So it is more about the flow comparison that is it is about the tonnage.
Jennifer Symoun
What exports are going to Switzerland and Brazil?
Bruce Lambert
I actually know the answer to that because Ed Lee called me on it. And there was, at one time, a huge shipment of commercial gold that went from Florida to a bank in Switzerland.
Jennifer Symoun
Interesting. Did you also compare ton-miles with the fused flows? If yes, how do they compare?
Thomas Hill
I do not think we did that at the end of the day.
Jennifer Symoun
All right. And is there any concern in the fusing of TS and FAF data and then outreach efforts at local and regional levels that you might reidentify previously scrubbed proprietary data from Transearch?
Thomas Hill
We do not report any of the Transearch data. We use Transearch internally to derive the algorithm, but we do not report any of the Transearch data at all in our report. It is really about the methodology and the algorithm and applying that algorithm then to the FAF data set.
Jennifer Symoun
Thank you. We have two questions for Eric now. Eric, and I know that you posted some links, is there any additional information (online or off) regarding road to rail studies?
Erik Johnson
I am actually in the process of doing this right now. If you Google Virginia Truck Rail Diversion Study, it actually returns a pretty healthy document. If someone is interested in a full compilation of what we have done, have them contact me and I am more than happy to help.
Jennifer Symoun
Thank you. Another question, how did you identify freight generators?
Erik Johnson
First, I contacted our economic development partnership, who basically track with business property across the state. They had some information in a GIS format that I reviewed. And once I started looking at it, I got the idea that if I look at aerials, and from aerials you can definitely tell large buildings and you can see their docks and their basic footprint, and just started adding to that map. I had to move some stuff around because some of the locations that they had were improperly located. They were not in the right place, it did not match the actual business and that was supposed to be there, or the dot was on a house, which was probably an owner's mailing address. And, we just went from there, and that is how I did it. I am probably about 95% complete for the whole Commonwealth. I would like to add that the database, the attributes of the layer, include things like what is the code? What are the addresses? How many docks does it have, or bays? Does it have rail access as well? Those kinds of things. In the future, I would like to share this information with some of our partners that are generating models for us, to better inform the models that we get for the purpose of going out with VTrans and other studies that we do.
Jennifer Symoun
Thank you. We have two questions left in the chat pod that I will get through quickly and we will wrap up. The first question, Chip Millard of the Federal Highway Administration is on the line, maybe I will put that out to you? There are many freight networks that are not compatible from a node/segment standpoint. HPMS, NHFN, FAF, NHS, PFN, STAA for example. Is there any thought about reconciling these various networks with each other?
Chip Millard
You put me on the spot a little bit. I am not the best person, as I don't deal a lot with the freight data. There are people in my office that deal with some of these data sources and networks more than I do. I do not believe there is any sort of thought about reconciling the networks. I think that a lot of these networks measure different things. Some of them do overlap, some of them do not. Mind you, I realize that the challenge of having some of these networks be different from each other does create some challenge. I hear what the question is saying, I am not fond of that myself. But to my knowledge I do not think that there is any effort in terms of reconciling. I would need to follow up internally. To find out for sure if that is the case.
Jennifer Symoun
One last question, can any of the Florida speakers comment on the applicability of the Tropicana-Ocean Spray collaboration that got a lot of press about 5-6 years ago to achieving other efficiencies in terms of mode optimization and filling empty backhauls?
Florida DOT Presenters
We are not too familiar with that. Something to look into for the future though, thank you for the comment.
Jennifer Symoun
Okay, thank you. All right, well, we are out of time and it looks like we made it to all the questions. Thank you all for attending today's seminar and thank you to all of our presenters The recorded version of this event will be available within the next few weeks on the Talking Freight website. Registration is not yet available for the January webinar but once it is information will be sent through the Freight Planning LISTSERV. The Freight Planning LISTSERV is the primary means of sharing information about upcoming seminars. I encourage you to join the LISTSERV if you have not already done so. Thank you and happy holidays and happy new year everyone!
