Review of U.S. Department of Transportation Truck Size and Weight Study: First Report: Review of Desk Scans (2014)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

22 MODAL SHIFT The MAP-21 [Section 32801(a)(6)] specification for the CTSW study modal shift analysis asks the study to estimate the following: (A) the extent to which freight would likely be diverted from other surface transportation modes to principal arterial routes and National Highway System intermodal connectors if alternative truck configuration is allowed to operate and the effect that any such diversion would have on other modes of transportation; (B) the effect that any such diversion would have on public safety, infrastructure, cost responsibilities, fuel efficiency, freight transportation costs, and the environment; (C) the effect on the transportation network of the United States that allowing alternative truck configuration to operate would have; and (D) whether allowing alternative truck configuration to operate would result in an increase or decrease in the total number of trucks operating on principal arterial routes and National Highway System intermodal connectors. The modal shift analysis may be described as the most critical element of the CTSW study, because the magnitudes of all other effects of changes in size and weight limits depend on the changes in truck traffic. The committee’s review of the modal shift desk scan in light of the MAP-21 requirements led to the following observations. Is the Desk Scan Thorough? The desk scan is thorough in identifying past public-sector studies that included prospective estimates of the effects of changing truck size and weight limits on mode shares of freight traffic. The desk scan also

23 reviews past estimates or models of impacts of changes in truck size and weight limits on fuel efficiency, the environment, traffic flow, and highway cost recovery. Three gaps in the assessment of past studies may lead the study team to overlook resources useful in the CTSW study:  The diversion projections of the mode shift models are not compared in terms of their utility or credibility for their intended applications. This omission is especially important in view of limitations of the Intermodal Transportation and Inventory Cost (ITIC) model chosen for use in the CTSW study.  Alternative freight mode choice models deserve greater examination. The review covers only mode choice models used in studies of truck size and weight regulations. A broader review of freight demand modeling might have revealed other methods of analyzing the effect of changes in size and weight limits on freight flows.  Methods of estimating the effects of mode shifts, as required in the MAP-21 study charge, are not adequately covered. Some of these effects (e.g., safety effects and highway infrastructure effects) presumably will be estimated in other parts of the CTSW study; however, the desk scans do not appear to cover methods of estimating effects on freight transportation costs, cost responsibilities, fuel efficiency, or the environment. The committee’s concerns with regard to each of these gaps are explained below. Diversion Projections The studies described are those produced for past prospective evaluations of proposed changes in truck size and weight regulations by USDOT, TRB, and the Government Accountability Office, together with four recent state studies and three from other sources. The mode shift methodology used in each study is identified, and the three principal methodologies (disaggregate total logistics cost models like ITIC,

24 aggregate econometric models, and expert opinion) are compared qualitatively, primarily with respect to practicality of use. The review does not describe quantitative results of past model estimates, reliability of the estimates, or sources of uncertainty, and it does not compare projections with outcomes. Citation of any published critiques of ITIC or any external review of the model that the Federal Railroad Administration may have commissioned would be especially helpful. For example, the desk scan could cite the discussion of the limitations of ITIC, including the problem of all-or-nothing freight allocations, in the 2000 USDOT truck size and weight study (USDOT 2000, IV-10–IV-11). The desk scan discusses data availability related to commodity flows. It focuses on four data sources: the Freight Analysis Framework (a synthesis of data from various sources rather than an independent source), IHS Global Transearch, the Commodity Flow Survey, and the Surface Transportation Board carload waybill sample. This focus is appropriate, since these are the major data sources available. The desk scan also should have investigated the availability of other proprietary freight flow databases. The desk scan concludes that the ITIC model is most suitable for the CTSW study, primarily on the grounds that no other directly applicable model is available and that development of an aggregate econometric model would be impractical within the study schedule. The comparison of alternative models does not consider the suitability of the ITIC model to the aggregate analysis to be conducted in the CTSW study, as discussed in the desk scan and project plan. This problem is considered below in the section headed “Does the Desk Scan Interpret the Literature Reviewed Correctly?” Alternative Freight Mode Choice Models The desk scan would benefit from an overview of the fundamental concepts of modal diversion (that is, beyond studies examining solely the question of truck size and weight limits) and a review of other methodologies, in particular, econometric models of mode choice. Such a review would provide assurance to the public that the team is using the best possible methodologies given the constraints and

25 would contribute to advancing USDOT’s ability to analyze freight market issues in the future. The desk scan needs a more comprehensive literature review of econometric models. The scan includes hardly any sources from the academic literature, which is unfortunate since recent studies on logistics analysis may help inform the ITIC model. These publications are essential for understanding the underlying behaviors concerning mode choice. In particular, the team should examine the following: 1. Publications that have used the Commodity Flow Survey data (Samuelson 1977; McFadden et al. 1986; Abdelwahab and Sargious 1990; Abdelwahab and Sargious 1991; Abdelwahab and Sargious 1992; Abdelwahab 1998; Abdelwahab and Sayed 1999). Although they are dated, these publications provide insight into the effect of commodity type on mode choice. 2. Publications that have studied the topic of freight vehicle choice (Holguín-Veras 2002; Cavalcante and Roorda 2010; Holguín-Veras et al. 2011), which is central to the CTSW study. 3. Publications that have used or developed supply chain models for the study of mode choice (Hall 1985; Leachman 2008). A literature review that covers some of the academic literature related to diversion has been prepared by Winebrake et al. (2012). On the question of whether changes in limits will induce a change in the total volume of freight traffic, the scan cites a single reference, produced for the 2000 USDOT truck size and weight study. The desk scan acknowledges that change in total traffic volume is a key question because it affects safety and infrastructure and has other consequences; therefore, citations of alternative models or newer estimates of induced freight traffic would be valuable. The ongoing Research Project 44 of the National Cooperative Freight Research Program, Impacts of Policy-Induced Freight Modal Shifts (TRB 2013), is expected to estimate econometric models and, possibly, develop a major revision of the ITIC model. These models may enhance USDOT’s capabilities of analyzing freight mode and vehicle choice. Unfortunately, the models are not likely to be ready before

26 the fourth quarter of 2014. Effects of Mode Shifts The desk scan should identify methods of estimating the effect that diversion would have on other modes of transportation, public safety, infrastructure, cost responsibilities, fuel efficiency, freight transportation costs, and the environment, if such effects are not addressed in other parts of the CTSW study.  Effect of diversion on infrastructure: Pavement and bridge impacts are to be covered in other parts of the CTSW study. However, infrastructure effects of diversion may include issues related to the rail system as well as consequences for rail yards and transfer facilities. Such items are not included in the desk scan.  Effect of diversion on cost responsibilities: A number of cost responsibilities might arise from diversion and affect shippers, carriers, and consumers. Also, because diversion affects the quality of infrastructure, the cost elements of this infrastructure (and how those costs are shared between the public and private sectors) may be important to consider.  Effect of diversion on fuel efficiency: This section of the desk scan focuses on the Environmental Protection Agency’s GEM (Greenhouse Gas Emissions Model) primarily, with some discussion of other approaches to determine trade-offs between truck and rail. Hardly any nongovernmental literature is cited in this area, but much exists with respect to the energy and environmental trade-offs of truck versus rail. The challenge here is that many studies are “top–down” and do not account for the specific operations of a truck or locomotive. The “bottom–up” calculations are more accurate but are not as applicable to a wide network. 1 In the end, various energy consumption factors may need to 1 The literature has referred to analyses that use fleet averages to estimate mode-specific energy use or emissions factors as “top–down” analyses. For example, a top–down fuel efficiency factor for trucks (BTU/ton-mile) is calculated by taking the total energy use reported for the trucking sector and dividing by the total ton-miles of freight activities for this sector. The results provide fleet averages that may “shroud the true variability that exists

27 be calculated in a bottom–up fashion for a set number of truck configurations, routes, and commodities and then applied in a top–down fashion for a national network of transportation operations. Some of the problems with top–down calculations are explained by Comer et al. (2010) and Winebrake and Corbett (2010).  Effect of diversion on the environment: The comments above related to energy consumption apply to the effect of diversion on the environment. In addition, the desk scan is relatively silent on the non– greenhouse gas emissions shifts that could occur with respect to diversion. Moving freight from rail to truck (or vice versa) could have significant impacts on particulate matter, oxides of nitrogen, and oxides of sulfur, for example. There is also a spatial dimension to these pollutants, and it should be recognized that both the amount of pollution and the location of that pollution (which ultimately leads to exposure to populations) are important. Some of these issues are discussed in the academic literature, particularly in analyses that examine waterborne freight as an alternative to land-based freight, since waterborne freight could reduce exposure of populations to pollution even if the overall emissions are higher. Is the Desk Scan Missing Literature, Case Studies, Models, or Data That Would Help Achieve the Study Goals? The preceding section identifies literature relating to alternative mode choice models and models and studies relevant to estimating the effects of changes in truck size and weight limits that may be of value to the CTSW study. with respect to shipping goods” (Comer et al. 2010). Another approach for estimating energy use or emissions factors in the freight sector is the “bottom–up” approach. In this approach, specific information about the route; travel speed; and vehicle, locomotive, or vessel characteristics is applied to calculate factors used in mode selection. The value of a bottom–up approach is that it creates data that are more realistically calibrated to the specific freight diversion question at hand.

28 Does the Desk Scan Interpret the Literature Reviewed Correctly? The desk scan does not adequately describe the differences between the ITIC mode choice model and the alternative models considered or potential shortcomings of ITIC for the CTSW study application. The committee understands that ITIC, a disaggregate single-shipper model, is being used in the CTSW study to determine the optimal mode and vehicle to transport nationwide freight flows, which are represented in the method as aggregate county-to-county flows by commodity. This method implicitly assumes perfect cooperation and a homogeneous level of service across shippers and perfect consolidation of the cargo. Although these assumptions may be needed on account of the lack of readily available alternative modeling approaches, they lead to some complications that should be taken into account. The optimal mode or vehicle obtained from that process is likely to have a larger capacity than the ones that would be obtained without the assumption of perfect cooperation and consolidation; consequently, projections may overstate the rail share of traffic or the average weight or volume of truck loads. This effect should be kept in mind and accounted for when the results are interpreted. Moreover, using ITIC on the data described on pages 30–39 of the desk scan will create all-or-nothing assignments for each origin– destination–commodity pairing. This should be mentioned and addressed. Does the Desk Scan Synthesize the Literature and Draw Appropriate Conclusions? Beyond the qualitative comparisons in Tables 1 and 2, the desk scan does not attempt to synthesize the literature. Two kinds of synthesis are needed for the study:  A synthesis of experience in the use of alternative methods of estimating mode shares, in terms of reliability and applicability, that leads to an assessment of the current state of the art. (That is, are reliable predictions of the effect of policy changes on mode shares feasible with available models?)

29  A synthesis of quantitative results of past estimates of mode shift effects of changing truck size and weight limits. Readers assessing the credibility of the CTSW study final report will need to compare the mode shift estimates with those of earlier studies and to understand the sources of differences, which may arise from real-world changes over time (e.g., changes over time in freight markets), differences in modeling assumptions, or differences in the policies simulated in the various studies. The committee’s overall impression of the desk scan is that its intent is to justify a prior decision about the method to be used in the CTSW study. There is not a logical flow from literature review to synthesis to conclusion. The desk scan represents more a listing of reports and literature, followed by a conclusion that is likely based on availability of models, time to complete the study, familiarity with the methods, and budget. References Abbreviations TRB Transportation Research Board USDOT U.S. Department of Transportation Abdelwahab, W. M. 1998. Elasticities of Mode Choice Probabilities and Market Elasticities of Demand: Evidence from a Simultaneous Mode Choice/Shipment-Size Freight Transport Model. Transportation Research Part E: Logistics and Transportation Review, Vol. 34, No. 4, pp. 257–266. Abdelwahab, W. M., and M. Sargious. 1990. Freight Rate Structure and Optimal Shipment Size in Freight Transportation. Logistics and Transportation Review, Vol. 26, No. 3, pp. 271–292. Abdelwahab, W. M., and M. A. Sargious. 1991. A Simultaneous Decision-Making Approach to Model the Demand for Freight Transportation. Canadian Journal of Civil Engineering, Vol. 18, No. 3, pp. 515–520. Abdelwahab, W., and M. Sargious. 1992. Modelling the Demand for Freight Transport: A New Approach. Journal of Transport Economics and Policy, Vol. 26, No. 1, pp. 49–70.

30 Abdelwahab, W., and T. Sayed. 1999. Freight Mode Choice Models Using Artificial Neural Networks. Civil Engineering and Environmental Systems, Vol. 16, No. 4, pp. 267–286. Cavalcante, R., and M. J. Roorda. 2010. A Disaggregate Urban Shipment Size/Vehicle-Type Choice Model. Presented at 89th Annual Meeting of the Transportation Research Board, Washington, D.C. Comer, B., J. J. Corbett, J. S. Hawker, K. Korfmacher, E. E. Lee, C. Prokop, and J. J. Winebrake. 2010. Marine Vessels as Substitutes for Heavy-Duty Trucks in Great Lakes Freight Transportation. Journal of the Air and Waste Management Association, Vol. 60, July, pp. 884–890. Hall, R. 1985. Dependence Between Shipment Size and Mode in Freight Transportation. Transportation Science, Vol. 19, No. 4, pp. 436–444. Holguín-Veras, J. 2002. Revealed Preference Analysis of Commercial Vehicle Choice Process. Journal of Transportation Engineering, Vol. 128, No. 4, pp. 336–346. Holguín-Veras, J., N. Xu, G. de Jong, and H. Maurer. 2011. An Experimental Economics Investigation of Shipper–Carrier Interactions on the Choice of Mode and Shipment Size in Freight Transport. Networks and Spatial Economics, Vol. 11, No. 3, pp. 509–532. Leachman, R. C. 2008. Port and Modal Allocation of Waterborne Containerized Imports from Asia to the United States. Transportation Research Part E: Logistics and Transportation Review, Vol. 44, No. 2, pp. 313–331. McFadden, D., C. Winston, and A. Boersch-Supan. 1986. Joint Estimation of Freight Transportation Decisions Under Non-Random Sampling. Discussion paper, Harvard University. Samuelson, R. D. 1977. Modeling the Freight Rate Structure. Center for Transportation Studies, Massachusetts Institute of Technology. TRB. 2013. NCFRP 44 [Active]: Impacts of Policy-Induced Freight Modal Shifts. http://apps.trb.org/cmsfeed/TRBNetProjectDisplay.asp?ProjectID=3534. USDOT. 2000. The U.S. Department of Transportation’s Comprehensive Truck Size and Weight Study: Volume III: Scenario Analysis. Aug.

31 Winebrake, J. J., and J. J. Corbett. 2010. Improving the Energy Efficiency and Environmental Performance of Goods Movement. In Climate and Transportation Solutions: Findings from the 2009 Asilomar Conference on Transportation and Energy Policy (D. Sperling and J. S. Cannon, eds.), University of California, Davis. Winebrake, J. J., E. H. Green, B. Comer, J. J. Corbett, and S. Froman. 2012. Estimating the Direct Rebound Effect for On-Road Freight Transportation. Energy Policy, Vol. 48, Sept., pp. 252–259.