Comprehensive Truck Size and Weight Limits Study: Linkage between the Revised Desk Scans and Project Plans Report
Chapter 2: Modal Shift Comparative Analysis
2.1 Purpose
This section highlights linkages between the Revised Desk Scan and the project plan developed for the modal shift analysis, the energy and environmental impact analysis, and the traffic operations impact analysis in the Modal Shift Comparative Analysis. The focus is on how the desk scans shaped the general technical approach outlined in the project plans and the specific data and analytical techniques available to produce information needed to meet overall 2014 CTSW Study objectives.
2.2 Modal Shift Analysis Linkages
The most direct linkage between the desk scan and the project plan for the modal shift analysis was the identification of required outputs from the modal shift analysis. These outputs included not only aggregate changes in mode choice, but also changes in the distribution of truck traffic by operating weight, vehicle configuration, and highway functional class. These latter outputs are required to estimate impacts of truck size and weight scenarios on infrastructure, safety, compliance, traffic operations, energy consumption and environmental emissions.
The primary focus of the desk scan was to identify past studies that had developed outputs corresponding to the outputs required for the 2014 CTSW Study. The scope of the desk scan was broadened, however, to include studies that focused primarily on the issue of modal choice, but that might be able to be extended to produce the full range of outputs required for the current study.
The modal shift desk scan identified several different approaches that have been used in past studies to estimate impacts of potential truck size and weight policy changes on modal shifts. Those methods were generally categorized as disaggregate approaches, aggregate econometric approaches, and expert opinion. Disaggregate approaches use characteristics of actual or representative shipments as the basis for estimating modal shifts. They are data intensive, especially if actual data are used, and require analytical tools that capture the major transportation and logistics costs that shippers/carriers consider when making mode choice decisions. Most of the disaggregate studies identified in the desk scan used the Intermodal Transportation Inventory Cost (ITIC) model. This model is described in detail in the Revised Desk Scan. Others used different formulations of economic order quantity logistics cost models that consider the same types of transportation and non-transportation logistics costs as are included in ITIC to analyze mode choice decisions. The disaggregate approaches were found to be particularly robust for the type of analysis required for the CTSW Study since they analyzed mode choice decisions on a shipment by shipment basis, thereby allowing detailed assessments of the impacts of network restrictions, commodity attributes, and vehicle attributes on the choice of mode, vehicle configuration, vehicle operating weights, and VMT by highway functional class. Econometric approaches generally are based on estimates of the cross-elasticities of demand for one mode as a function of changes in prices for another mode. For instance if the price of truck transportation falls as a result of truck size and weight policy changes, the demand for rail transportation would be expected to fall based on the cross-elasticity of demand with respect to truck prices. Econometric approaches generally are applicable only to choices between truck and other modes, and generally have not been used to estimate shifts in traffic between different truck configurations. Several freight demand modeling studies have examined choices between type of carrier (truckload, less-than-truckload, parcel delivery) based on characteristics of individual shipments, but like other econometric studies, these freight demand study approaches could not produce the detailed outputs that must be considered in the CTSW Study. The third major approach to estimating modal shifts is reliance on opinions of transportation experts who are familiar with freight transportation markets in the area being studied. Expert opinion has most often been used in State studies where there was insufficient budget to apply disaggregate approaches.
Disaggregate approaches that explicitly consider total transportation and non-transportation logistics costs associated with the use of different modes and different vehicle configurations thus were judged to be superior to other mode choice modeling approaches for purposes of the 2014 CTSW Study. In reviewing the disaggregate approaches that had been used in previous studies, the one most frequently used was the ITIC Model. That model had been used in the USDOT Comprehensive Truck Size and Weight Study, 2000 (2000 CTSW Study), the Western Uniformity Scenario Analysis, and in an analysis of potential mode shifts along the I-81 corridor in Virginia. Using that model for the 2014 CTSW Study has several advantages. First, it facilitates a comparison of results of the 2014 CTSW Study with those other studies, especially the 2000 CTSW Study. Second, ITIC was developed by USDOT and has been used both by FHWA and FRA. This reduces any claims that the model is biased toward one mode or the other. Third, the ITIC model has undergone recent updates that reduced the need for extensive updating. Fourth, ITIC is publicly available, which was an important study criterion. No other publicly available models that had the capabilities of ITIC to meet study requirements were uncovered in the desk scan. If a superior analytical tool had been found, that certainly would have been considered for use in the 2014 CTSW Study.
2.2.1 Commodity Flow Data
Another important linkage between the desk scan and the project plan was the data required to produce outputs required from the modal shift analysis. The desirability of using a disaggregate approach to modeling mode choice required that a disaggregate commodity flow database be identified.
As noted, ITIC is very data intensive and it was important to identify nationwide commodity flow databases that contained shipments by mode, by commodity, and by origin and destination. Several databases were identified including the Commodity Flow Survey; Transearch, a proprietary database maintained by IHS Global Insight; the Freight Analysis Framework (FAF) developed by the Federal Highway Administration (FHWA); and the Surface Transportation Board's (STB) Carload Waybill Sample. No other nationwide commodity flow databases that would meet study requirements were identified during the desk scan. Advantages and disadvantages of each database are discussed in detail in the desk scan report. Ultimately the FAF was selected as the database for truck movements because it is more complete than the Commodity Flow Survey which is a data source for the FAF and it is more accessible than the proprietary Transearch database.
An important consideration in selecting the FAF was whether the data could be disaggregated to a finer level of geography than the 123 zones into which origins and destinations are reported. With just 123 zones, many States are represented by just a single zone. This was insufficient geographic detail for purposes of network routing and analyzing the impacts of restricting access for triples to a limited network of highways. Oak Ridge National Laboratory disaggregated the FAF to produce county-to-county flows. FHWA does not disaggregate the FAF to this level of detail because the accuracy of individual flows is not as high as with the 123 zone-to-zone flows, and use of county-to-county flows by State and local planning agencies could produce unreliable results when used for infrastructure investment decisions. However for a national-level policy study, the lower accuracy of individual flows was more than offset by the ability to provide greater network resolution and to analyze how restricting network access might affect shifts to triples.
The FAF includes rail shipments, but data in the STB's Carload Waybill Sample are more detailed in terms of origin and destination, the type of equipment used, rates charged, and whether short-line railroads were involved in the rail moves. Because of this greater detail, the Waybill Sample was used as the database for rail moves. The Waybill Sample had also been used in the 2000 CTSW Study.
Another important set of data required for all impact assessments including modal shift was the base case distribution of traffic by vehicle class, operating weight, and highway functional class. These data served as control totals for estimating overall shifts in VMT by vehicle class, operating weight, and highway functional class which in turn are important in estimating safety, infrastructure, energy and environmental impacts of truck size and weight policy changes. All of these impacts are sensitive to changes in traffic by vehicle class, weight, and highway functional class. The development of the base case traffic estimates is summarized in the modal shift desk scan. This same type of VMT breakdown by vehicle class, weight group, and highway functional class was used in the 2000 CTSW Study, but for the 2014 CTSW Study the data were broken down into more vehicle classes and more weight groups to add precision to estimates of infrastructure impacts.
In addition to information on the distribution of total traffic by vehicle class, ITIC also required information on the body types used to haul various commodities since different body types have different operating costs, payloads, and other operating characteristics. The only nationwide source of information on characteristics of the vehicles used to haul various commodities is the Vehicle Inventory and Use Survey (VIUS) conducted by the Census Bureau. The last VIUS was conducted in 2002 and it was recognized that data might not precisely reflect operations in 2011, the base year for the 2014 CTSW Study. While information in the VIUS is dated, the importance of reflecting body type and other operational characteristics of vehicles used to haul various commodities was essential to the analysis and the 2002 VIUS data were used. Updating the VIUS data is a key research need.
In summary, the linkage between the desk scan and the project plan for the modal shift analysis was largely driven by the study requirements. Estimates of modal shifts were important in their own right, but were perhaps more important as the basis for estimating safety, infrastructure, energy, environmental and traffic operations impacts associated with truck size and weight policy changes. The importance of producing the best estimates possible of these various impacts dictated that analytical techniques and data that would provide detailed changes in VMT by vehicle class, operating weight, and highway functional class be used if available. The 2000 CTSW Study and the Western Uniformity Scenario Analysis both provided guidance on the analytical tools, data sources, and study approaches that could provide the best estimates of impacts of the truck size and weight policy options. If those analytical techniques and data sources were not available, the modal shift project plan would have been quite different and would not have been able to provide the detailed estimates of traffic shifts needed to estimate other impacts.
2.3 Energy and Environmental Analysis Linkages
As with the modal shift analysis, study requirements were important considerations in reviewing the literature on heavy truck fuel consumption and emissions and in formulating a plan to conduct the analysis. The modal shift analysis produced changes in VMT by vehicle class, operating weight, and highway functional class, and methods were required that could take all those factors into consideration when estimating impacts on fuel consumption and emissions.
The literature review indicated an evolution of approaches to estimating heavy truck fuel consumption and emissions. The evolution was driven in part by Federal regulations that set maximum emission levels and the need to develop methods to objectively measure emissions from different vehicles. Because regulations were aimed at truck tractors, these methods focused on truck tractors rather than the tractor-trailer combination as a whole. Increasingly the methods included tire rolling resistance and aerodynamic drag in addition to engine efficiency in estimating fuel consumption and emissions.
The Environmental Protection Agency, the Northeast States Center for a Clean Air Future (NESCCAF) and other environmental groups have been investigating a broad range of potential strategies to reduce greenhouse gas and other environmental emissions, including the use of trucks with higher gross vehicle weights that could move more freight with each trip. These studies required analysis not only of emissions associated with the truck tractor, but also consideration of the impact of different trailer lengths, numbers of trailers, and numbers of tires to understand emissions associated with the whole vehicle combination.
Other key developments found in the literature were improved vehicle simulation models that made it possible to extrapolate findings from one vehicle class to another and the development of drive cycles that represented the mix of driving conditions that vehicles would encounter in actual use. These capabilities suited requirements to estimate fuel consumption and emissions by different vehicle classes operating at different weights on different highway classes. The project plan for estimating fuel consumption and environmental emissions associated with the truck size and weight scenarios being analyzed in the 2014 CTSW Study was developed based on the capabilities of the analytical tools and data found in the desk scan.
The specific modeling tools chosen for the analysis were those used in the 2009 NESCCAF study of options for reducing CO2 emissions associated with heavy trucks. Those same analytical tools also are being used by members of the Study team in an on-going project for the National Highway Traffic Safety Administration (NHTSA) and had been verified as part of that NHTSA project. Tire rolling resistance and vehicle aerodynamic drag coefficients for different vehicle classes were based on work for the NESCCAF study that had examined fuel consumption and emissions for heavy 6-axle tractor-semitrailers, 33-foot doubles, and triple trailer combinations among other vehicle classes. The NESCCAF study focused on long-distance trucking operations and the drive cycles used in that study were consistent with that focus. Drive cycles for the 2014 study were modified to be more representative of the full range of trucking operations.
Thus, as for the modal shift analysis, study requirements strongly influenced the preliminary project plan for energy and environmental emissions analysis and they also focused the desk scan on past studies that had conducted the same types of analyses as were required for the 2014 CTSW Study. Analytical techniques and data were discovered that met study requirements, and the final project plan was structured around the use of those techniques. Unlike the modal shift analysis, the methods and data were quite different from those used in the 2000 CTSW Study since the state-of-the-art in emissions modeling has advanced so much since the 2000 CTSW Study.
2.4 Traffic Operations Analysis Linkages
As noted in the traffic operations desk scan, traffic operations involves a number of specific elements including maintaining speed on grades; weaving, merging, and changing lanes; highway capacity and level of service; and maneuvering through signalized intersections. A common thread in all of these elements is the impact on vehicle delay and traffic congestion. Since delay and congestion costs are critical factors affecting all traffic, the focus of the traffic operations analysis was on estimating those two items, although past truck size and weight studies were scanned for analyses of all aspects of traffic operations.
As with modal shift and energy and environmental analysis, the starting point in developing the preliminary project plan for traffic operations was the study requirements. The product of the modal shift analysis was changes in VMT by vehicle class, operating weight and highway class. The objective of the traffic operations analysis was to translate those changes in VMT into changes in traffic operations.
The Highway Capacity Manual (HCM) is the recognized source of relationships between traffic volumes and highway level of service, delay, and congestion. Over the years relationships in the HCM have been updated and refined. The HCM was last updated in 2010; relationships in that edition of the HCM were the basis for estimating how changes in traffic resulting from truck size and weight scenarios would affect levels of delay and congestion costs.
An important step in estimating traffic delay is to translate truck volumes into passenger car equivalents (PCEs). PCE values are specified in the HCM to account for the effects of different volumes and characteristics of truck traffic. The value of PCEs depends on the operating speed and grade of the highway section, the vehicle's length, and its weight- to-horsepower ratio which measures how a vehicle can accelerate. PCE values in the HCM do not include the vehicle lengths, however. The desk scan identified several studies that had estimated the PCEs for different vehicle configurations not included in the HCM. PCE values estimated for the 2000 CTSW Study were selected for use in the 2014 CTSW Study.
The HCM is not a network or system analysis tool. Techniques are required to apply relationships in the HCM to system-wide highway conditions to estimate total delay and congestion costs. An analytical tool had been developed for the 2000 CTSW Study to estimate changes in system-wide delay. This tool was selected for use in the 2014 CTSW Study, but speed volume relationships had to be updated to reflect changes in the 2010 HCM.
The analytical tool requires data on highway characteristics on different highway functional classes. The critical highway characteristics are the percentage of different types of highways with different grades and the percentage of different types of highways that are congested with volume/service flow ratios greater than or equal to 0.8. The source for those characteristics was FHWA's Highway Statistics publication.
The only nationwide study that had attempted to estimate changes in highway delay and congestion costs associated with truck size and weight policy options was the 2000 CTSW Study and only two State studies were found that included estimates of impacts on delay and congestion costs. Analytical tools had been developed for the 2000 CTSW Study that met requirements of the current study, but relationships in those tools had to be updated to reflect changes in highway characteristics since the 2000 CTSW Study and changes in speed - flow relationships in the 2010 HCM.
previous | next
