Innovations in Freight Demand Modeling and Data Symposium by alicejenny


									SHRP2 C20: Freight Demand Modeling and Data Improvement Strategic Plan

Innovations in Freight Demand Modeling and
Data Symposium
Conference Summary

James M. Brock, Avant IMC, LLC
Chair, Symposium Planning Team

September 14-15, 2010
Crowne Plaza Hotel
Herndon, VA

Hosted by
Avant IMC, LLC
E2 Engineering
Gannett Fleming, Inc.

Sponsored by
Transportation Research Board of the National Academies
Federal Highway Administration

                                                                                            SHRP2 C20: Freight Demand Modeling
                                                                                             and Data Improvement Strategic Plan

                                                       Table of Contents
Summary ......................................................................................................................................... 3
    Purpose ........................................................................................................................................ 3
    General Observations .................................................................................................................. 3
    Process......................................................................................................................................... 5
    Day 1 ........................................................................................................................................... 6
       Session 1. Regional Freight Model Development ................................................................... 6
       Session 2: Alternative Techniques for Modeling Freight Transport ....................................... 6
       Session 3: The Application of Econometric and Statistical Methods for Freight Modeling ... 7
       Session 4: International Perspectives on Modeling Freight Movements ................................ 7
    Day 2 ........................................................................................................................................... 7
       Session 5: Data Collection and Visualization Techniques for Analyzing Freight Travel
       Patterns .................................................................................................................................... 7
       Session 6: Microsimulation Approaches to Freight Transportation Forecasting .................... 8
    Closing Discussion ...................................................................................................................... 8
Appendix A - Feedback on Symposium and its Future ................................................................ 11
    1. Innovative Ideas of Promise .................................................................................................. 11
    2. More Research Needed ......................................................................................................... 13
    3. Moving Forward with Freight Demand Forecasting Community ......................................... 15
    4. Tell Us Your Thoughts .......................................................................................................... 16
Appendix B – Attendee List ......................................................................................................... 18
Appendix C – Research Papers and Presentations ........................................................................ 22

                                                                                                                                             Page 2 
                                                            SHRP2 C20: Freight Demand Modeling
                                                             and Data Improvement Strategic Plan



The Transportation Research Board of the National Academies, Second Strategic Highway
Research Program, sponsored an interactive symposium centered on innovations in freight
demand modeling and data, hosted by Jim Brock of Avant IMC, LLC, and Richard Easley of E2
Engineering. The symposium is an integral part of the SHRP2-C20 initiative, aimed at fostering
fresh ideas and new approaches in freight demand modeling and associated data. This report
summarizes the symposium process and feedback.

The symposium was held on September 14-15, 2010, at the Crowne Plaza in Herndon, VA, with
about 50 in attendance. The symposium featured fifteen presentations (out of sixteen – one
presenter was unable to attend) selected to address the challenge of finding the next generation of
freight demand models. Each attendee was given a packet including a CD with a copy of all of
the presentations, and feedback forms for attendee comments and suggestions.

Freight traffic has been growing at a rate faster than passenger traffic. Freight demand behavior
is more dynamic and heterogeneous than passenger demand. There is considerable and complex
interaction between international and domestic flows, public and private interests, and logistics
behavior. Understanding and being able to forecast freight traffic is a critical input to planning
for future highway capacity. The purpose of the symposium was to incite a dialogue and learning
to bring the freight modeling practice closer to real world, practical, relevant freight model
generation and outputs, and—importantly—further the science of freight demand modeling.

General Observations

Freight-Modeling community is small and fragmented; there is a dearth of MPO and DOT
freight planners, while many academic and consulting researchers. Some attributed this to
economic times, but some indicated many MPOs and DOTs do not enjoy having planners
dedicated to freight modeling.

The symposium seemed to be very well regarded by the participants; who could take it over in
future years? (Perhaps there is a different answer in the short-term and the long-term.)

There was at times an almost shocking disconnect between the modelers' interests and the
decision-makers' needs; how do we bridge the gap? How do we get the academics pointed in a
more useful direction? There are a lot of freight models being developed that really have little or
no policy relevance.

There appeared to be three main research dimensions coming out of the symposium –
Knowledge, Models, and Data:

                                                                                             Page 3 
                                                         SHRP2 C20: Freight Demand Modeling
                                                          and Data Improvement Strategic Plan

1. Knowledge
      a. Gap in understanding in public sector about the underlying economic drivers of
         freight traffic, and therefore a gap in understanding differences in freight and
         logistics modeling. More attention to tour-based and supply chain models and data.
      b. What are the decision variables that should be considered in a model?
      c. Additional research is needed in this area of understanding how the systems operate.
         Once there is sufficient knowledge of how the system operates, then it can be
         determined what data is needed.

2. Models
     a. Differences in modeling approaches between long haul freight traffic models and
          urban delivery models.
     b. Many models attempt to explain the present system, but there is a need for more
          predictive models.
     c. There is also a major disconnect between logistics practitioners/modelers and freight
          practitioners/ do we bridge this gap?
     d. An important distinction is often lost: travel link approaches versus O/D-based
     e. Many of the models were not designed to be predictive; rather they were an attempt
          to explain the current state of events for their particular data set.
     f. Many of the models seemed to be the first step or limited in their applicability in
          their current state. Many of the presenters acknowledged the limitations of the model
          they were presenting.
     g. Very few models are multimodal.
     h. Knowing how companies tend to react and how they fit into the aggregate world is

3. Data
      a. Data is an important limitation of modeling; need to reconcile and integrate sources
         of data.
      b. Something like the Ontario CV survey (truck intercept survey) perhaps conducted in
         several locations in the US to help with freight model calibration.
      c. Something analogous to the ITE Trip Generation Manual but for freight: land
         use/employees in, tons and truck trips out.
      d. Travel time reliability data for major routes or O/D pairs...for long-haul trucking,
         some carriers seem to trade off distance and vehicle operating cost for greater
         reliability of delivery windows
      e. It seems as if the private sector (at least the Wal Marts and Con-Ways of the world)
         may be a lot more willing than we thought to part with their freight data holdings
         (which are huge) if confidentiality is maintained and they see a benefit for

                                                                                        Page 4 
                                                            SHRP2 C20: Freight Demand Modeling
                                                             and Data Improvement Strategic Plan

           themselves. Some sort of standards need to be set for datasets, data transfer,
           the private carriers are not driven insane in the process.
        f. California's on-line freight data portal (Cal-Fred) is a great idea and could be
           replicated at the regional level...mega-regions seem to be the ideal level at which to
           do this.
        g. There was a wide variety of data sources used to create the models.
        h. Some participants felt strongly that there is a limitation in the data; which limits the
           production of good models. Others disagreed stating the issue is not data
           availability; but rather fully understanding how the freight system works/operates is
           what is needed to generate good models. It is not possible to generate a valid freight
           model if one does not understand how the system operates. One problem is that
           every study is different. Need to find areas of commonality – this would be a good
           area of study.

The two-day symposium was organized into six sessions, based on six categories of areas of
research interest as determined by the papers topics submitted. Each session was comprised of
two or three presentations, twenty minutes in duration, followed by a fifteen minute question-
and-answer discussion involving the entire audience and the presenter. Immediately following
the Q & A, the next speaker was introduced, and so on.
These freight modeling presentations also competed for a $1,000 prize. The symposium included
a diverse audience representing academia, public sector practitioners, and private industry.
Participants examined, evaluated, and promoted innovative and promising advances in freight
demand modeling, data collection, and freight forecasting research methods. Local, state-level,
regional, and domestic and international models were presented.
The actual papers submitted with the respective presentations are contained in the appendices.
Each presentation topic is identified below with a sequential number for easy reference to the
number in the name of the associated electronic file. The electronic files for the presentations and
the papers can be retrieved from


                                                                                             Page 5 
                                                                   SHRP2 C20: Freight Demand Modeling
                                                                    and Data Improvement Strategic Plan

Day 1
Welcome Remarks: Jim Brock, Avant IMC, LLC
Jim opened the proceedings with a welcome and overview of the symposium and its objectives.

Private Sector Perspective on Freight Demand Modeling, Randy Mullet, Con-way Inc., Keynote

Randy challenged the group to be aware of the private sector logistics economic and business
drivers, and how they impact freight movement and private-sector decisions. Only 13.4% of
freight travels more than 500 miles over its entire supply chain; and over the next 30 years, the
modal shift is expected to remain constant. Only 1% of freight moves by rail, and rail only plans
at a regional level. Freight movement is “liquid” and will find the path of least-resistance and
optimal-cost in the supply chain, and on road networks in particular. He further described how
logistics modelers view freight movement differently than public-sector freight modelers, and
stressed the need to find a common language. His comment,” freight is freight is freight”,
challenging the notion of modelers needing to know what is in the trucks elicited a strong
response from the group. He also indicated that carriers such as Con-way have a lot of data that
might be used on the public-sector side, but indicated that there needed to be common standards
and a value proposition presented to the private sector for the use of the data. His final
suggestion was to have the private and public-sector groups begin a dialogue to pursue common
modeling interests.

Session 1. Regional Freight Model Development 
   1. Using the Oregon Statewide Integrated Model for the Oregon Freight Plan Analysis
   Presenter: Tara Weidner, Parsons Brinckerhoff
   Co-author: Becky Knudson, Oregon Department of Transportation

   2. Microsimulation of Commodity Flow in the Mississippi Valley Region
   Presenter: Alan Horowitz, University of Wisconsin – Milwaukee

Session 2: Alternative Techniques for Modeling Freight Transport
   3. Generation of a U.S. Commodity Flows Matrix Using Log-Linear Modeling and Iterative
       Proportional Fitting (Freight Analysis Framework 3)
   Presenter: Frank Southworth, Oak Ridge National Laboratory
   Co-author: Bruce E. Peterson, Oak Ridge National Laboratory
   Note: Although his paper was selected, Frank Southworth was unable to attend and did not present.
   His paper is included in Appendix C.

   4. Analysis and Multi-Level Modeling of Truck Freight Demand
   Presenter: Huili Wang, University of California at Berkeley
   Co-authors: Ching-Yao Chan, University of California at Berkeley;
   Kitae Jang, Northwestern Polytechnical University, Xi’an, Shaanxi, P.R. China

   5. Modeling Commercial Vehicle Daily Tour Chaining
   Presenter: Jane Lin, University of Illinois at Chicago
   Co-authors: Minyan Ruan and Kazuya Kawamura, University of Illinois at Chicago

                                                                                                Page 6 
                                                                    SHRP2 C20: Freight Demand Modeling
                                                                     and Data Improvement Strategic Plan

Session 3: The Application of Econometric and Statistical Methods for Freight Modeling
    6. A Bayesian Hierarchical Network for Truck Demand Modeling
    Presenter: Mihalis Golias, University of Memphis
    Co-authors: Maria Boile, Rutgers University; Stephanie Ivey, University of Memphis

    7. Freight Demand Modeling Using Econometric Models
    Presenter: Talha Muhammad, HDR, Inc.
    Co-author: Pierre Vilain, HDR, Inc.

Session 4: International Perspectives on Modeling Freight Movements
    8. Modeling Truck Route Choice Behavior by Traffic Electronic Application Data
    Presenter: Tetsuro Hyodo, Tokyo University of Marine Science and Technology
    Co-author: Yasukatsu Hagino, Transport and Socioeconomic Research Division, Institute of Behavioral

    9. Innovative Freight Transportation Framework for Flanders
    Presenter: Tabitha Maes, Universiteit Hasselt (Belgium)
    Co-authors: Katrien Ramaekers, An Caris, Tom Bellemans, and Gerrit K. Janssens, Universiteit Hasselt

    10. A Model System for Forecasting National and International Freight Transport in Norway
    Presenter: Inger Beate Hovi, Institute of Transport Economics (Oslo, Norway)
    Co-author: Wiljar Hansen, Institute of Transport Economics (Oslo, Norway)

Day 2
Opening Remarks: Thoughts on Visualization of Data in Freight Operations, Ron Hughes, TRB
Visualization Committee/North Carolina State University
Ron’s remarks highlighted and promoted the use of visualization tools to analyze and evaluate
data. His point of emphasis was that visualization tools is not a new way to present PowerPoint
slides or spreadsheet graphics, which are 2D; nor is it 3D or 4D – it is the development of new
visualization and computational methods, called visual analytics.

Session 5: Data Collection and Visualization Techniques for Analyzing Freight Travel Patterns
    11. Fusing Public and Private Truck Data to Support Regional Freight Planning and Modeling
    Presenter: Chen-Fu Liao, University of Minnesota

    12. Online California Freight Data Repository for Freight Modeling and Analysis
    Presenter: Andre Tok, University of California at Irvine
    Co-authors: Miyuan Zhao, Joseph Y.J. Chow, Stephen G. Ritchie, and Dimitri Arkhipov, University of
    California at Irvine

    13. Analysis Tool to Process Passively-Collected GPS Data for Commercial Vehicle Demand
        Modeling Applications
    Presenter: Bryce Sharman, University of Toronto
    Co-author: Matthew Roorda, University of Toronto

                                                                                                           Page 7 
                                                                      SHRP2 C20: Freight Demand Modeling
                                                                       and Data Improvement Strategic Plan

Session 6: Microsimulation Approaches to Freight Transportation Forecasting
    14. A Tour-Based Urban Freight Demand Model Using Entropy Maximization
    Presenter: Qian Wang, State University of New York at Buffalo
    Co-author: Jose Holguin-Veras, Rensselaer Polytechnic Institute

    15. A Firm-Based Freight Demand Modeling Framework: Capturing Intra-firm Interaction and
        Joint Logistic Decision-Making
    Presenter: Qi Gong, University of Wisconsin – Madison
    Co-author: Jessica Y. Guo, University of Wisconsin – Madison

    16. A Hybrid Microsimulation Model of Urban Freight Travel Demand
    Presenter: Rick Donnelly, Parsons Brinckerhoff
    Co-authors: Marcus Wigan, University of Melbourne; Russell Thompson, Monash University

Closing Discussion

A facilitated discussion was held at the close of the proceedings; the following are comments
made by the group:

       Papers were great ideas, excellent work.

       Need more submissions from private sector.

       Shared innovations in travel modeling last spring was a good conference.

       Suggest in these economic times subsidize MPOs and state DOTs, and other public-
        sector practitioners to come to the table. Unless you believe recession is going away
        soon, maybe get sponsorships for attendees – need to travel and need to learn.

       Panel where you have a policy person, Con-Way, discuss what are the issues that their
        agencies are facing such as policies, greenhouse gases. Not necessarily papers, but what
        issues they are having, what freight models can help or solve their issues.

       Small conference is good. Instead of having a prize, take the money and put in a
        scholarship fund – call for practitioners at state levels. What do you want to take away
        from this, if they can’t get here, we’ll help you pay for it. If doing a modeling session in
        other conferences, get up and spend 5 minutes explaining what you are doing, but not
        really a presentation. Give them an opportunity to get feedback from that.

       Within a 50-mile radius of this location, probably several MPOs and none are in

       Maybe get awareness and visibility through AASHTO.
                                                                                                   Page 8 
                                                         SHRP2 C20: Freight Demand Modeling
                                                          and Data Improvement Strategic Plan

   A lot of what was talked about here was generally what has been done recently, or what
    is work in progress. What is the best way to get ideas that are newly formed, create new
    ideas/creative ideas to go whole new directions, not just directions of working on paths
    that are already started – encourage whole new paths.

   Some of the public sector people would not be able to come to the symposium. Not same
    as in-person discussion, but may consider a 2-hour web seminar, brainstorming session
    for ideas, discussions with the public sector participating.

   Thought of having state DOTs or MPOs coming in stating ‘here are some issues we are
    dealing with.’ I like that, but not sure how this makes sense when you put out a call for
    papers. This information is helpful before presentation is prepared. If I’ve already
    prepared something, then you tell me you don’t need that, you need something
    completely different, I may feel like ‘what am I doing here.’

   Lots of good ideas, is there any collaboration, does anyone see this possible.

   Regarding MPO/state/DOT, could potentially have an update webinar to get their points.

   Do think need to get ideas of what policy makers are trying to deal with. This is critical.
    Also becomes the basis for them feeling like it is a worthy thing to participate in. If they
    don’t see the value because they are dealing with something different than we are
    working on, then they won’t be very interested.

   Is there a way – web participation – any way start conceiving and matching up, this is
    how you get RSS feed – a way to take this information and start to build something from

   The format that TRB uses for a lot of their conferences is call for papers, present papers,
    then have breakout discussion. Didn’t do that here, maybe use that kind of format next
    time? Having a 2 or 3 hours on Wednesday afternoon to conduct breakout discussions
    may be good for future.

   Need participation thru TRB for travel modeling as well as freight modeling – present
    what have done here, whet their appetite (possibly at the TRB meeting) and garner more

   Discussion was very good, good facilitating was done to encourage discussion.

   Webinar, FHWA talking freight always looking for subjects, would be good way for state
    DOTs/MPOs to get their freight models out there – 150 to 200 folks monthly on these
    sessions, bringing freight modelers and state DOTs/MPOs together.

   Commodity versus freight, kind of a conflict.

                                                                                          Page 9 
                                                             SHRP2 C20: Freight Demand Modeling
                                                              and Data Improvement Strategic Plan

       Freight descriptor versus freight predictor; if does not predict anything, what are we

       Freight modeling versus freight policy/transportation operations tool. Freight modeling,
        where is freight going versus environmental, infrastructure improvements, congestion
        policies, parking policies, models to be able to do that versus where freight is going.

       Putting all those different competing things together in one model is very difficult, and
        possibly worthy of more discussion.

Award Presentation

The award was presented to Tetsuro Hyodo, Tokyo University of Marine Science and
Technology; and the Symposium was adjourned after David Plazak, TRB, made closing remarks.



                                                                                            Page 10 
                                                           SHRP2 C20: Freight Demand Modeling
                                                            and Data Improvement Strategic Plan

Appendix A ‐ Feedback on Symposium and its Future 

The following is a complete listing of all of the comments received for both days.

1. Innovative Ideas of Promise
   Integration of the logistics modeling into an overall freight system of modeling
    More discussion on the value of a suite of models in freight
    All of the presentations were excellent
    Use of data sources such as weigh-in-motion stations and overweight truck permits
    The work with micro simulation of commodity freight flows
    Many freight modeling tools had been presented and discussed. Obviously each model can
     answer some certain questions and shortcomings. Is there a need to develop general data
     needs for developing models? Is there a list of national data resources that can help
    Integrating logistics models with demand models appears to be the next logical step. This
     is similar to the demand for MPO type models that are more responsive to operational
    Distinct differences between long distance and delivery models, but both are important.
    European approach/desire to looking at multimodal freight chains with public sector hopes
     to generate interest in public/private partnerships in rail, road, and waterways models to
     incorporate multimodal freight are needed.
    Potential for oversize/overweight permits as a data source for traffic assignment models
    Daily tour chain models—Approach to modeling daily chain linkages
    Microsimulation of Mississippi Valley Region detailed simulation of commodities
    Bayseian networks for improving/synthesizing data.
    Tour-based freight demand forecasting methods for metropolitan/urban areas.
    Aggregation models may work well too given the limited information available.
    Tour-based microsimulation approaches
    Some neat data collection methods
    Advanced statistical approaches—Bayes
    Modeling truck route choice behavior by traffic electronic application data
    Modeling commercial vehicle daily tour chaining
    Contrast between freight modeling and logistic objectives of focus at a shorter time period
    Great datasets—Ontario and TXDOT. More research needed!
    Dynamic traffic assignment (60 min intervals) for assigning long trips in Mississippi
     Valley model
                                                                                         Page 11 
                                                           SHRP2 C20: Freight Demand Modeling
                                                            and Data Improvement Strategic Plan

   Multiple tours/day—How important is this and for what industries?
   Freight demand model at the freeway levels (not macroscopic scale)
   Freight demand modeling vs. logistics modeling
   Freight vs. commodity approach
   Data availability vs. model sophistication
   Tour-based vs. tour chaining approach
   Importance of capturing the commonality of logistics management. The linkage between
    logistics management and freight demand modeling
   Using Bayesian modeling tools for modeling truck data seems like an ideal solution to
    getting around poor/limited data which may not be statistically significant
   Interest in using truck permitting (Japan) to identify truck flows and better routes.
   Use of scenarios (economic growth, fuel prices, etc.) for different futures
   Use of WIM and classification data
   Microsimulation, tour-based models
   Hearing from logistics planners from a range of companies-to inform freight modeling and
   Do the optimization methods logistics folks use really apply regionally (are they only used
    and are descriptive of large trucking companies but not small companies or owner-
   Agent-based models
   Multi-level models
   Hybrid approach to tour-based modeling
   Using the firm as the dependent variable in modeling freight instead of establishments
   Use of GPS as a freight planning tool
   Premise of modeling firms versus establishment
   The data tools from California
   Does CalFred provide a model databank for each state or should it become a national data
    bank that states can tap?
   Data scalability (Ron’s session)––need to find out more about that
   “Connecting the Dots”––used first for visualization but we need to connect between the
    different databases (GPS, Assessor, etc.) to better understand the system
   Personally, I was extremely interested in the aggregate modeling approach presented by
    Qian Wang and Rick Donnelly. Also, the database repository created by UC Irvine is
    extremely useful for researchers and practitioners as finding suitable data can often be

                                                                                            Page 12 
                                                             SHRP2 C20: Freight Demand Modeling
                                                              and Data Improvement Strategic Plan

    The trip generating systems that was presented in nearly all presentation today were new
     approaches to me. I need inspiration for our future work with the CIS for Norway, and
     have picked up many ideas.

2. More Research Needed
   “Shippers” decision characteristics or factors
    It seems that most of the limitations discussed here are related to data. Many presenters
     showed innovative analyses and models, which were based on available data more than
     mathematical and practical concepts.
    National data reporting standard to support freight planning and modeling
    Solicit list of needs from state DOT or MPO to determine what questions need to be
     answered and what data needs, where can we find the data and the possibility of merging
     private data as Randy kindly offered the possibility of sharing private data.
    The relationship between “descriptive (discrete choice) model” and “logistics optimizing
     model” would be discussed. It relates the main issue of this symposium.
    What is “Freight demand modeling”? If just describes results of optimized logistics
     activities? If so, more attention to logistics model should be considered.
    Route choice, departure time choice, international connections and transfers
    Nationwide multimodal data, vs. the district PIERS data of what enters and leaves the port
     and URCS data of what rides the rail. More is needed to capture ship to rail to truck and
     vice versa.
    Factors in logistics models that should be reflected in freight models to represent shipper-
     carrier responses to public policies.
    Tour based and supply chain modeling
    Relationships between monetary flows, commodity flows and vehicle model flows
    Impact of fuel prices(or alternative fuel policies) on freight industries
    Data needs that meet the modeling needs
    Lot more understanding of freight decision processes and factors
    Data sources and data fusion
    Need NHTS for freight
    Integrating demand-side frameworks with logistics/supply-side frameworks.
    Data gathering
    How to include more agent interactions
    Mode choice models
    More disaggregated data
    Multiple tours/day

                                                                                         Page 13 
                                                             SHRP2 C20: Freight Demand Modeling
                                                              and Data Improvement Strategic Plan

   Where can models cut corners?
   Need to distinguish/develop models for intra-metro areas vs. freeway vs. local corridors
   Need for multiple models to address freight demand and logistics planning
   Data repository: Creating an inventory/database of all freight data available to researchers
    and practitioners
   Good modeling exercise guidelines
   Data collection guidelines for regional freight surveys and 3PL surveys
   Have to acquire more freight data from private sectors. What’s the knowledge in logistics
    management important for freight modeling? How to incorporate them in freight demand
   Additional information on applicants for all presentations (shared before selection)
   How truck models can inform logistics decisions
   How can we marry the private data with public metrics and provide beneficial direction to
   Parallels and differences between freight and passenger modeling…to help professionals
    who are familiar with person demand modeling to understand how to deal with freight.
   Quantifying and explaining the benefits/costs of policy decisions to the public. Voters
    typically have a limited understanding of the supply chain. Examples: Nighttime truck
    restrictions; sitting of tanker facilities.
   Data with enough detail to understand logistics practices
   Implementable logistics-based model components
   Relevance and importance of logistics to statewide regional freight modeling and
    forecasting by public agencies
   Trip tour modeling methods and advantages (based on real data)
   Developing statewide/regional freight planning and forecasting models that are sensitive to
    economic conditions and policy options, without inordinate data and calibration
   Model validation findings and methods
   Understanding statewide and urban goods movement systems better
   Randy Mullet asserts that only 13.4% of freight moves greater than 500 miles. Either he
    has a view reflective of Conway’s customers or most of our FAF and foreign trade
    statistics are wrong. Can both sources be reconciled?
   Standardizing best practices for specific applications
   Additional refinement to firm based model––identify efficiency of the data
   Expansion of modeling framework for tow based hybrid microsims

                                                                                           Page 14 
                                                              SHRP2 C20: Freight Demand Modeling
                                                               and Data Improvement Strategic Plan

    Data fusion
    Standardizing best practices for specific applications
    How to integrate the multiple data sources and maintain the database
    Correlate actual speeds with modeled speed of trades to make the Highway Capacity
     Manual more sensitive to trades
    So where should the research agenda go next? What innovations can be combined to
     create better work?
    Compiling is making databases more available
    Comparing transferability of parameters/models to other regions
    Interaction between different vehicles is shared and divided traffic streams under different
     weather conditions and congestion conditions. We have HOV during specific hours, how
     about Heavy Vehicle Operating Lanes?
    I was surprised that none of the American presentations were focusing on intermodality
    Making GPS data representative of all tracking
    GPS on tractor versus on trailer/container versus on shipment
    Effective visualization

3. Moving Forward with Freight Demand Forecasting Community
   Go to 1 day format
    Suggest rotating location to foster more involvement
    Try to involve others than just those involved with freight demand modeling (as a judge or
     two, for example)
    I hope this will be an annual event. It’s a very good opportunity to exchange ideas and
     discuss with academies, practitioners, and government people
    Bayesian methods
    Multi-level modeling of truck freight demand
    First of all, I think this symposium should go continue on an annual basis. Where it will be
     held every year may depend on who the next host is going to be. Personally, I’m OK with
     having the symposium in DC. On the other hand, having it in different places every year
     may also make it attractive.
    Somewhere in east coast or west coast…better to be around campus.
    Should be held at University Transportation Centers/Institutes in different regions
    Dulles Airport location is good as well
    Western US next time. At least west of Mississippi

                                                                                           Page 15 
                                                             SHRP2 C20: Freight Demand Modeling
                                                              and Data Improvement Strategic Plan

    ITS at UC Irvine (my institution) could be interested
    Midwest seems to have a disproportionate amount of research; Maybe a location in
     Minnesota, Wisconsin or Illinois.
    It’s curious that only two speakers showed up (the majority of the audience). You’d think
     this would be more broadly interesting to the freight modeling crowd. Maybe this would
     be better positioned as a workshop at TRB annual meeting?

4. Tell Us Your Thoughts
   Like the time limit on presentations
    Like the “prize” for best/innovative idea
    Think some of the presentations suffered from lack of content—anything to help with that
     would be good such as ensuring that every presentation has: 1) Problem Statement; 2)
     Approach to address problem, and; 3) Evaluation of effectiveness
    Consider having different categories for the presentations and award multiple winners (e.g.
     national and multi-state vs. state and vs. metro area)
    I came to hear what’s going on about the work with modeling freight demand, and received
     inspiration especially about work with commodity flow modeling
    I really enjoyed this symposium and I have learned a lot. Best aspects include the intense
     discussions after most presentations, which I have not seen at other conferences
    I liked the international speakers who brought the perspective of international freight
     planning and modeling
    Round table system is very good
    Candid questions. Good give and take! Good humor and encouragement
    Small group. A lot of interaction great atmosphere for sharing
    Like discussion between speakers sessions
    Like small conversational group that provides context (good and bad) about speakers
    Choice of speakers/topics might be more rigorous—e.g. ensure model has been developed
     with sufficient data before acceptance
    I like the small setting so you get a chance to talk to everyone in the symposium
    Diverse topics of presentation that cover wide range of modeling techniques
    I like the discussions about modeling and data, however the discussions were a little short
    Well prepared presentations
    Could be a working group at the end of presentation group to discuss presenters/critique
    Maybe some more diverse topics—A lot of truck presentations. Maybe something on
     ports/rail interests if available

                                                                                          Page 16 
                                                             SHRP2 C20: Freight Demand Modeling
                                                              and Data Improvement Strategic Plan

       Good presentations and discussion
       All positives. Small size (although up to 50-70 would be OK too); informality; selectivity
        and focus regarding presentations (very important!); Location near major airport; Hotel not
        too expensive
       Small number of really top notch people
       Great diversity in presentations
       Maybe some additional perspective on private data (work) sources (trucking company,
        3PC, large firm)
       Great discussion; Dynamic
       Like idea of scholarship for state DOT folks, as well as webinar
       Ditch the prize––perhaps a better incentive is publication at TRB (TRR) or similar outlet
       I am extremely pleased and impressed with this symposium. I had more feedback and
        discussion here than most (if not all) other conferences that I’ve attended.
       A common dinner would have gotten the group more familiar
       It was a pleasure to come and hear what is on in the US!
       The presentations were put together in a good way, so that the themes fit together. Positive
        that the call for papers was limited to 3-5 pages!



                                                                                            Page 17 
                                                           SHRP2 C20: Freight Demand Modeling
                                                            and Data Improvement Strategic Plan

Appendix B – Attendee List                    
Saad Alhussein                               Alice Cheng   
Associate Professor                          President 
King Saudi University                        Cheng Solutions, LLC 
Addiriyah                                    25 Prospect Park West 
Riyadh, Saudi Arabia 23211                   Brooklyn, NY 11215‐1706 
Ph: +966505121617                            Ph: 347‐277‐4424                  
Patrick Anater (Pat)                         Ted Dahlburg   
Senior Transportation Planner/Associate      Manager of Freight Planning 
Gannett Fleming                              Delaware Valley Regional Planning Commission 
207 Senate Ave                               190 North Independence Mall West, 8th Floor 
Camp Hill, PA 17011                          Philadelphia, PA 19106 
Ph: 717‐763‐7212, x2659/Fax: 717‐763‐8150    Ph: 215‐592‐1800                   
James Brock (Jim)                            Rick Donnelly   
President & CEO                              Principal Associate 
Avant IMC, LLC                               Parsons Brinckerhoff, Inc. 
2157 Market Street                           6100 Uptown Blvd, NE, Ste 700 
Camp Hill, PA 17011                          Albuquerque, NM 87110 
Ph: 717‐234‐5225                             Ph: 505‐881‐5357                 
Ching‐Yao Chan                               Richard Easley  
Program Leader/Transportation Safety         President 
Research                                     E‐Squared Engineering 
California PATH, Univ. of California at      43969 Tavern Dr., #200 
Berkeley                                     Ashburn, VA 20147 
Bldg 452, Richmond Field Station, UC         Ph: 703‐858‐5588 
Berkeley                                     Fax: 703‐724‐0983 
Richmond, CA 94804                           reasley@e‐ 
Ph: 510‐665‐3659                                          Sharon Easley  
                                             Project Manager 
Keith Chase                                  E‐Squared Engineering 
Vice President                               43969 Tavern Dr., #200 
Gannett Fleming                              Ashburn, VA 20147 
207 Senate Ave                               Ph: 703‐858‐9545 
Camp Hill, PA 17011                          Fax: 703‐724‐0983 
Ph: 717‐763‐7212, x2108                      seasley@e‐ 
Fax: 717‐763‐8150                                                          

                                                                                             Page 18 
                                                             SHRP2 C20: Freight Demand Modeling
                                                              and Data Improvement Strategic Plan

Mihalis Golias (Mike)                          Tetsuro Hyodo (Tetsu)   
Assistant Professor                            Professor 
University of Memphis                          Tokyo University of Marine Science  
112A Engineering Science Bldg,                      and Technology 
3815 Central Ave                               2‐1‐6, Etchujima, Koto‐ku 
Memphis, TN 38152                              Tokyo, Japan 135‐8533 
Ph: 901‐678‐3048                               Ph: +81‐3‐5245‐7386               
Qi Gong (Chi)                                  Barbara Ivanov (Barb)   
Graduate Student/Research Assistant            Freight Systems Division Co‐Director 
University of Wisconsin‐Madison                Washington State Dept. of Transportation 
1249A Engineering Hall, 1415 Engineering Dr    310 Maple Park Avenue SE 
Madison, WI 53706                              Olympia, WA  98504‐7407, WA 98504‐7407 
Ph: 608‐334‐0541                               Ph: 360‐705‐7931                        
Jose Holguin‐Veras                             Susie Lahsene   
Professor                                      Regional Transportation and Land Use Manager 
Rensselaer Polytechnic Institute               Port of Portland 
110 8th St, J&C 4030                           7200 N.E. Airport Way 
Troy, NY 12180                                 Portland, OR 97218 
Ph: 518‐276‐6221                               Ph: 503‐415‐6517                           
Alan Horowitz                                  Chen‐Fu Liao   
Professor of Civil Engineering                 Sr. Systems Engineer 
University of Wisconsin‐Milwaukee              University of Minnesota 
PO Box 784                                     500 Pullsbury Dr. SE 
Milwaukee, WI 53201                            Minneapolis, MN 55311 
Ph: 414‐229‐6685                               Ph: 612‐626‐1697                      
Inger Beate Hovi                               Jane Lin  
Chief Research Economist                       Associate Professor 
Institute of Transport Economics               University of Illinois at Chicago 
Gaustadalleen 21                               842 W. Taylor St (M/C 246) 
Oslo, Norway NO‐0349                           Chicago, IL 60607 
Ph: +4722573834                                Ph: 312‐996‐3068                
Ronald Hughes (Ron), Ph.D.                     Tabitha Maes  
TRB Visualization Committee/                   Ph.D. Student   
North Carolina State University                IMOB ‐ University Hasselt 
Centennial Campus, Box 8601                    Wetenschapspark 5 bus 6 
Raleigh, NC 27695‐8601                         Diepenbeek, Belgium 3590 
Ph: (919) 515‐8523/Fax: (919) 515‐8898         Ph: +32 011 269133                     

                                                                                           Page 19 
                                                             SHRP2 C20: Freight Demand Modeling
                                                              and Data Improvement Strategic Plan

Cameron Millard                                David Plazak    
Freight Analyst                                Senior Program Officer 
Cambridge Systematics                          Transportation Research Board/SHRP 2 
4800 Hampden Lane, Suite 800                   500 5th Street, NW 
Bethesda, MD 20814                             Washington, DC 20001 
Ph: 301‐347‐0100, x119/Fax: 301‐347‐9101       Ph: 202‐334‐1834/Fax: 202‐334‐3471                   
Talha Muhammad                                 Stephen Ritchie (Steve)  
Economist                                      Professor 
HDR Inc                                        University of California, Irvine 
500 Seventh Ave                                Inst. of Transportation Studies, 4000 AIR Bldg 
New York, NY 10018                             Irvine, CA 92697 
Ph: 212‐542‐6063; 646‐872‐7954                 Ph: 949‐824‐4214/Fax: 949‐824‐8385             
C. Randall Mullett (Randy)                     Matthew Roorda (Matt)              
Vice President                                 Assistant Professor 
Con‐way Inc.                                   University of Toronto 
575 7th St NW                                  Dept of Civil Engineering, 35 St. George St 
Washington, DC 20004                           Toronto, ON M5S 1A4 
Ph: 202‐637‐0994                               Ph: 416‐978‐5976  
Diana Olney                                    Rolf Schmitt    
Director, Business Development                 Team Leader of Freight Mgmt. and Operations 
IHS Global Insight                             FHWA 
1150 Connecticut Ave, NW, Suite 401            1200 New Jersey Ave, SE 
Washington, DC 20036                           Washington, DC 20590 
Ph: 202‐481‐9265/Fax: 202‐481‐9301             Ph: 202‐366‐0408                   
Maren Outwater                                 Bryce Sharman   
Director                                       Ph.D. Candidate 
Resource Systems Group                         University of Toronto 
917 E. Spooner Rd                              Dept of Civil Engineering, 35 St. George St 
Fox Point, WI 53217                            Toronto, ON M5S 1A4 
Ph: 425‐269‐9684/Fax: 802‐295‐1006             Ph: 647‐618‐3900                  
Ram Pendyala                                   Robert Shull (Bob)         
Professor                                      Vice President 
Dept of Civil and Environmental Engineering    PTV America, Inc. 
Arizona State University                       1145 Broadway Plaza, Suite 605 
Mail Code: 5306                                Tacoma, WA 98402‐3583 
Tempe, AZ 85287                                Ph: 253‐272‐4440 Ext. 2605/Fax: 253‐736‐0640 
Ph: 480‐727‐9164                                                 

                                                                                                 Page 20 
                                                              SHRP2 C20: Freight Demand Modeling
                                                               and Data Improvement Strategic Plan

Scott Smith                                     Tara Weidner    
Operations Research Analyst                     Senior Planner 
US DOT/RITA/Volpe Center                        Parsons Brinckerhoff, Inc. 
55 Broadway                                     400 SW Sixth Ave, Suite 802 
Cambridge, MA 02142                             Portland, OR 97204 
Ph: 617‐494‐2588/Fax: 617‐494‐3260              Ph: 503‐478‐2342                    
Victor Suski (Vic)                              Wade White    
General Manager                                 President 
Shenandoah Express, Inc.                        Whitehouse Group 
344 Reliance Rd.                                40 East Joyce Lane 
Middletown, VA  22645                           Arnold, MD 21012 
Ph: 540‐869‐4036                                Ph: 202‐241‐2008/Fax: 866.846.5875                  
Yeow Chern Tok (Andre)                           
University of California, Irvine                 
Institute of Transportation Studies,  
Univ of California, Irvine 
Irvine, CA 92697‐3600 
Ph: 949‐690‐0737 
Huili Wang        
Visiting Graduate Student 
California PATH, Univ. of California at 
Bldg 452, Richmond Field Station, UC 
Richmond, CA 94804 
Ph: 510‐665‐3659 
Qian Wang         
Assistant Professor 
University at Buffalo,  
the State University of New York 
231 Ketter Hall, Univ at Buffalo North Campus
Buffalo, NY 14260‐4300 
Ph: 716‐645‐4365 

                                                                                          Page 21 
                                                           SHRP2 C20: Freight Demand Modeling
                                                            and Data Improvement Strategic Plan


Appendix C – Research Papers and Presentations 

Research papers and presentations can be accessed at 


                                                                                           Page 22 

To top