Separation of Vehicles—CMV-Only Lanes (2010)

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.

3.1 Introduction This chapter presents the results of the work conducted as part of Task 8 (Performance Eval- uation) of the project. The primary objective of the performance evaluation task was to conduct a comprehensive comparative assessment of the performance of different truck-only lane con- cepts, as well as concepts without truck-only lanes. The analysis is based on a detailed review of analytical/modeling studies of truck-only lane projects and field data from real-world truck-auto separation concepts in the United States and internationally, to the extent that data were available. The performance assessment effort did not involve any new modeling work. Rather, we evaluated each truck-only lane concept against a standard set of performance measures, which are discussed in the following sections: • Performance Evaluation Approach describes the main types of truck-only lane concepts (scenarios) included in the analysis, the types of system improvements (alternatives) consid- ered for the relative performance assessment, the selected set of performance measures, the data sources used for the evaluation, and the methodology used to compile, summarize, and assess the results from the various data sources. • Performance Evaluation Results provides a discussion of the performance analysis results from each of the selected data sources, as well as a comparative summary and analysis of results for each performance measure across data sources to arrive at a consistent assessment of the relative performance of truck-only lane concepts compared to other types of system improve- ments (alternatives). 3.2 Performance Evaluation Approach The study team’s performance evaluation was based on a procedure that pulled together data reported in other studies in a way that allowed the team to compare the results of, and develop consistent performance metrics for, a variety of different truck-only lane concepts. This approach, shown in Figure 3.1 relied on existing data and information available from evaluations of truck- only lane concepts—the study team did not define new alternatives to be evaluated, develop new data, or conduct new modeling or other detailed analyses. A description of each of the steps shown in Figure 3.1 is presented in this section. Although this approach presents certain limitations in terms of the types of alternatives that can be compared, the types of performance measures that can be evaluated reliably, and the abil- ity to validate the results of the analysis, it does provide a consistent basis by which different truck- only lane concepts can be compared. This kind of apples-to-apples comparison allowed the researchers to identify the most promising concepts and scenarios for further evaluation in a more 24 C H A P T E R 3 Performance Evaluation

controlled setting. In addition, this approach clearly defines some of the most critical research and data gaps and helps frame ideas for future research as discussed in the final chapter of this report. Step 1. Identification of Corridor Scenarios The first step in the performance evaluation approach involved the identification of key corri- dor scenarios for the performance evaluation of truck-only lanes. Two main generic scenarios were identified for the analysis, which included: (1.) long-haul intercity corridors and (2.) urban corri- dors. These two types of corridor scenarios are broadly representative of the major types of corri- dors for which truck-only lanes have been proposed in the past. • Long-haul corridors include intercity corridors that serve long-haul truck and auto traffic demand. Truck traffic along these corridors is predominantly composed of large (five or more axle) combination trucks moving long-haul freight. Some examples of key long-haul corridors in the U.S. carrying high truck volumes include I-15 between Barstow and Las Vegas, I-94 between Chicago and Detroit, I-5 between Bakersfield and Sacramento, and I-90 between Cleveland and Buffalo. Some of the characteristics of long-haul corridors that would make them particularly good candidates to consider for potential implementation of CMV-only lanes include travel demand (auto and truck traffic volumes), congestion, and LCV network connec- tivity. Additional discussion of these issues is presented in a later section. • Urban corridors include short-haul corridors in urban areas serving as primary access routes to major freight facilities (such as seaports) or major auto and truck travel corridors in major metropolitan areas. Examples of such corridors include the I-710, SR 60, and I-15 freeways in Performance Evaluation 25 Step 1. Identification of Corridor Scenarios Step 2. Selection of Alternatives Step 3. Selection of Performance Measures and Metrics (within Each Measure) for Performance Evaluation Step 5. Post-Processing of Performance Results for Consistency in Comparisons Step 4. Identification and Review of Data Sources (Modeling/Analytical Studies) for Performance Evaluation within Each Corridor Scenario Figure 3.1. Steps involved in truck-only lane performance evaluation.

Southern California; the I-395, I-95, SR 112, and SR 836 corridors serving the Port of Miami, Florida; and the I-75, I-85, and I-285 freeways in metropolitan Atlanta. Step 2. Selection of Alternatives The second step was to select the range of operational and infrastructure investment alternatives against which to evaluate the performance of different truck-only lane scenarios. It is important to note that because our truck-only lane performance evaluation relied on existing feasibility studies and analyses, the range of potential investment alternatives to consider was limited to those eval- uated as part of these existing efforts. Table 3.1 describes the types of alternatives that the study team assessed for both of the truck- only lane scenarios. Since the applicability and viability of various system improvement options is a function of corridor characteristics, the set of alternatives selected for the performance evalua- tion was specific to each of the two scenarios. To understand the relative performance benefits of truck-only lanes, a no-build alternative was included in all of the analyses in order to compare the performance of truck-only lanes with an alternative without truck-only lanes. Additional alter- natives were included in the evaluation process, depending on the availability of performance data. Step 3. Selection of Performance Measures and Metrics Because different truck-only lane scenarios may have different performance objectives, appro- priate performance measures and metrics must be evaluated for each truck-only lane scenario. It is more critical to assess potential productivity gains along long-haul corridors, for instance, than 26 Separation of Vehicles—CMV-Only Lanes Scenario Alternatives Description No-build (without LCV operations) Includes all committed improvement projects to be implemented along the study corridor in the future, without LCV operations No-build (with LCV operations) Includes all committed improvement projects to be implemented along the study corridor in the future, along with LCV operations on general purpose lanes CMV-only lanes without LCV operations Includes implementation of truck-only lanes along the study corridor (in addition to the projects included in no-build), but without LCV operations Long-Haul Corridors CMV-only lanes with LCV operations Includes implementation of truck-only lanes along the study corridor (in addition to the projects included in no-build), along with LCV operations on truck-only lanes No build Includes all committed improvement projects to be implemented along the study corridor in the future Additional general purpose lanes Includes implementation of additional general purpose lanes along the study corridor (in addition to the projects included in no-build as well as Transportation Supply Management/Transportation Demand Management (TSM/TDM*) strategies) Urban Corridors CMV-only lanes Includes implementation of truck-only lanes along the study corridor (in addition to the projects included in no-build as well as TSM/TDM strategies), without the application of tolls Note: *TSM strategies include projects that optimize transportation system supply in a region to handle demand. For example, managed lanes optimize capacity of a roadway network by varying bidirectional lane capacities of a roadway for the morning and evening peak periods to account for varying commute travel demand in these time periods. TDM strategies include projects that optimize the demand on the transportation system, to encourage optimal system capacity utilization. An example of a TDM strategy is congestion pricing, which manages demand during congested time periods through the use of user fees. Table 3.1. Alternative types by corridor scenario.

along urban corridors. Conversely, congestion relief or travel time reliability benefits might be more relevant for urban corridors. It is important to note that this section is solely dedicated to analyzing the relative performance of truck-only lanes (based on consideration of a key set of per- formance measures) against other alternatives, without consideration of costs. Chapter 4 looks specifically at the net performance benefits of truck-only lanes compared to other alternatives based on an estimation of benefit-cost (B-C) ratios. The performance measures/metrics considered to guide the study team’s evaluation process are as follow: • Travel Time. This measure is used to quantify the mobility benefits of truck-only lanes com- pared to other alternatives. The benefits are quantified in terms of percent savings in travel time for the build alternatives (including the truck-only lane alternative) compared to the no-build alternative. For the truck-only lane alternative, travel time savings are estimated primarily for autos (and trucks) on general-purpose lanes (and for trucks on truck-only lanes to the extent to which these savings were quantified in the reviewed studies). • Travel Time Reliability. Travel time reliability is measured by the percent change in incident- related (nonrecurrent) delay for the build alternatives (including the truck-only lane alternative) compared to the no-build alternative. The following key aspects of truck-only lanes expected to contribute to reliability improvements along a corridor include: – Reduction in accidents due to mobility improvements on the general purpose lanes (due to diversion of trucks to truck-only lanes), – Accident reduction due to truck-auto separation and improvements in general flow condi- tions due to less weaving of trucks and autos, and – Increased capacity on the general purpose lanes resulting in improved processing of bottle- necks during incidents (increased capacity will result in increased efficiency in alleviating the traffic impacts of incidents). • Productivity. This measure is used to quantify the productivity enhancements realized by the private sector (shippers and carriers) due to operations on truck-only lanes. The parameter used to quantify productivity benefits is the increase in net earnings (revenue less cost) per ton mile29 for carriers. This parameter is computed based on (1.) the contribution of speed improvements on truck-only lanes to increased productivity of trucking operations (i.e., increased trucking industry earnings per ton mile from improvements in truck operating speeds); and (2.) contri- bution of LCV operations to productivity enhancements (increased net earnings per ton mile due to increase in truck payloads). • Safety. Safety benefits are quantified in terms of reduction in the number of accidents on gen- eral purpose lanes for the build alternatives (including the truck-only lane alternative) relative to the no-build alternative. Truck-only lanes also contribute to safety improvements along a cor- ridor by improving mobility on the general purpose lanes, and by reducing interactions between autos and trucks by diverting trucks to truck-only lanes. Step 4. Identification and Review of Data Sources As described earlier, the study team’s evaluation approach relied on data and information from existing planning and analytical/modeling studies of truck-only lane projects conducted in the United States. Although there have been a number of truck-only lane projects, studies, and Performance Evaluation 27 29Earnings per ton mile is an appropriate parameter to quantify productivity benefits of truck-only lanes because it can capture increased earnings for truckers due to improved speeds and an increase in payloads. Improved speeds (for truckers on truck-only lanes) would result in a reduction in operating costs per mile, which would translate into increased net earnings per mile. An increase in payloads (from LCV operations on truck-only lanes) would result in reduction in operating costs per ton, which would result in increased net earnings per ton.

initiatives in the United States, the study team only used those studies that analyzed long-haul or urban corridors and provided performance analysis results that could be used, either directly or after post-processing, for the performance evaluation process. The following data sources were used for performance evaluation: • Long-haul corridors – Reason Foundation study on long-haul truck corridors,30 – Western Uniformity Scenario Analysis,31 – I-35 Trade Corridor Study,32 and – Georgia Statewide Truck Lane Needs Identification Study.33 • Urban corridors – I-710 Major Corridor Study,34 – I-15 Comprehensive Corridor Study,35 – Georgia Statewide Truck Lane Needs Identification Study,36 and – PSRC FAST corridor study.37 Tables 3.2 through 3.5 describe the alternatives considered by each of these studies, and provide information on which of the key performance metrics described in this section (travel time, travel time reliability, productivity, and safety) are considered in these studies. It is clear from this analysis that different truck-only lane scenarios have different performance objectives. Long-haul corridor scenarios, for instance, are concerned with improving overall travel time, productivity, and safety (in cases where corridor configurations and truck-auto interactions are having safety impacts). As part of the performance analysis, these studies typically compared truck-only lane concepts against a no-build alternative (and, in some cases, against the benefits 28 Separation of Vehicles—CMV-Only Lanes 30R. W. Poole, Jr. and P. Samuel, Toll Truckways: Increasing Productivity and Safety in Goods Movement, Reason Foundation, http://www.fhwa.dot.gov/download/hep/freightplanning/talkingfreight3_16_05bp.ppt. 31U.S.DOT, Western Uniformity Scenario Analysis: A Regional Truck Size and Weight Scenario Requested by the Western Governors’ Association, April 2004. 32HNTB and Wilbur Smith Associates, I-35 Trade Corridor Study: Recommended Corridor Investment Strategies, Texas Department of Transportation, September 1999. 33Georgia Department of Transportation, Statewide Truck Lane Needs Identification Study, “Technical Memo- randum 3: Truck-Only Lane Needs Analysis and Engineering Assessment,” April 2008. 34Los Angeles County Metro, I-710 Major Corridor Study: Final Report, March 2005. 35Southern California Association of Governments, I-15 Comprehensive Corridor Study, December 2005. 36Georgia Department of Transportation, Statewide Truck Lane Needs Identification Study, “Technical Memo- randum 3: Truck-Only Lane Needs Analysis and Engineering Assessment,” April 2008. 37Kuppam, et al., Evaluating Freight Mobility on a Regionwide Basis Using EMME/2—Freight Action Strategy (FAST) Truck Model for the Puget Sound Region, 16th International EMME/2 User’s Group Conference, Albuquerque, New Mexico, March 2002. Study No-Build General Purpose Lanes with LCVs Truck Lanes with LCVs Truck Lanes without LCVs Reason Foundation Study Western Uniformity Scenario Analysis I-35 Trade Corridor Study ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Georgia Statewide Truck Lane Needs Identification Study Table 3.2. Alternatives considered within long-haul corridor truck-only lane studies.

and impacts of LCV operations on truck-only lanes). Conversely, urban corridor scenarios are more concerned with improving overall travel time, travel time reliability, and safety. Truck-only lane concepts in these areas were compared with other congestion-reduction alternatives, such as additional general purpose (GP) lanes and transportation system management strategies. As a result, these different scenarios use different performance metrics—long-haul corridors use travel time, productivity, and safety measures, while urban corridors calculate travel time, reliability, and safety measures. Recognizing these key differences, the study team structured their evaluation approach to nor- malize only those performance metrics that are applicable to each corridor scenario, as shown in Table 3.6. Step 5. Post-Processing of Performance Results In order to ensure consistent comparisons of performance results, the study team found it nec- essary to normalize the results from studies in order to develop a common metric for comparisons. The team used post-processing factors in cases where the desired performance metric was not esti- Performance Evaluation 29 Study No-Build Mixed-Flow Lanes Truck-Only Lanes TSM/TDM Strategies I-710 Major Corridor Study I-15 Comprehensive Corridor Study Georgia Statewide Truck Lane Needs Identification Study PSRC FAST Corridor Study ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Study Travel Time Productivity Safety Reason Foundation Study Western Uniformity Scenario Analysis I-35 Trade Corridor Study Georgia Statewide Truck Lane Needs Identification Study ✓ ✓ ✓✓ ✓ ✓ ✓ Study Travel Time Reliability Safety I-710 Major Corridor Study I-15 Comprehensive Corridor Study Georgia Statewide Truck Lane Needs Identification Study PSRC FAST Corridor Study ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ Table 3.3. Alternatives considered within urban corridor truck-only lane studies. Table 3.4. Performance metrics described within long-haul corridor truck-only lane studies. Table 3.5. Performance metrics described within urban corridor truck-only lane studies.

mated in a study, or where limitations were identified in the approach used by studies in the esti- mation of specific performance metrics. Post-processing tools and sources included the following: • Intelligent Transportation System (ITS) Deployment Analysis System (IDAS). IDAS is an analytical tool that provides post-processing factors for the performance evaluation of trans- portation capacity improvement strategies. Although the tool is intended to analyze the per- formance benefits of ITS deployments, the factors can be used to assess other transportation investments as well (such as infrastructure projects), as long as information is available on the level of capacity improvement achieved on the transportation system. The post-processing factors used from IDAS for the performance evaluation included the following: – Accident rates (number of accidents per million VMT) by type of accident for autos and trucks as a function of V/C38 ratio, for the estimation of safety benefits (reduction in accidents); and – Incident delay rates (hours of incident-delay per vehicle mile) as a function of V/C and num- ber of lanes, for the estimation of reliability benefits (savings in incident-related delay). • Handbook for Planning Truck Facilities on Urban Highways39 (Douglas handbook) and New Jersey Turnpike Accident Data. The Douglas handbook serves as a compendium of key issues related to the planning, policy, and performance/feasibility evaluation of truck facilities. The handbook provides recommendations on approaches to analyze the performance benefits of truck facilities. The handbook, based on an analysis of historic accident statistics for the New Jer- sey Turnpike, recommends a 15% accident reduction factor due to truck-auto separation (not taking into consideration the safety benefits of capacity improvements). This 15% accident reduction factor is used in the study team’s analysis. (A summary of historic accident statistics on the dual-dual and non dual-dual sections of the New Jersey Turnpike and the basis for arriv- ing at a 15% accident reduction factor to account for the safety benefits of truck-auto separation is provided in Appendix B, Section B.3, which is available on the TRB website at www.TRB.org by searching for NCHRP Report 649/NCFRP Report 3). 3.3 Performance Evaluation Results The following sections provide a brief overview of each of the major truck-only lane studies iden- tified in Step 4, discuss important findings within the four key performance metrics identified in Step 3 (Selection of Performance Measures and Metrics), and identify key assumptions and data 30 Separation of Vehicles—CMV-Only Lanes Scenario Performance Measure Long-Haul Corridors Urban Corridors Travel Time Travel Time Reliability N/A Productivity N/A Safety* Note: * There appear to be some key limitations in the evaluation methodologies and tools used to assess the safety benefits of truck-auto separation. However, there is a body of research (described in subsequent sections) that will allow us to consider safety impacts using post-processing techniques. ✓ ✓ ✓ ✓ ✓ ✓ Table 3.6. Applicability of performance measures to scenarios. 38V/C is a parameter used to quantify the level of congestion on a roadway. It is calculated as the ratio of the level of demand (traffic volume) to the supply (capacity of roadway). Typically, V/Cs are measured for a specific time period (e.g., hourly or daily). 39J. G. Douglas, Handbook for Planning Truck Facilities on Urban Highways, August 2004. This reference was used primarily to estimate safety benefits consistently.

gaps that hindered the study team’s ability to draw specific conclusions from each of these scenar- ios. A more detailed discussion and summary of each of these studies is provided in Appendix B (Sections B.1 and B.2, which are available on the TRB website at www.TRB.org by searching for NCHRP Report 649/NCFRP Report 3). 3.3.1 Long-Haul Corridors The following sections present the results of the performance evaluation of truck-only lanes along long-haul corridors, based on a detailed review of studies listed in Step 4. Reason Foundation Study The Reason Foundation conducted an analysis40 of toll truckways that compared the incremen- tal productivity benefits of three types of truck-only lanes to mixed-flow facilities without LCV operations (the no-build scenario). In addition to a “standard” semi-trailer option (i.e., 80,000-lb, 53-ft trailers), two LCV options were evaluated—one allowing triple shorts and one allowing turn- pike doubles. The Reason Foundation study is useful in presenting an illustrative methodological framework for the analysis of productivity benefits and for providing estimates to determine the relationship between speed improvements and LCV operations, as well as productivity benefits measured in terms of increased earnings per truck per ton mile. Table 3.7 presents the produc- tivity benefits per truck estimated by the Reason Foundation study to provide insights into the following areas: • Relative productivity benefits of truck-only lanes (with and without LCV operations) compared to a no-build alternative to understand the relative contributions of travel time savings and LCV operations on truck-only lanes to productivity benefits and Performance Evaluation 31 40R. W. Poole, Jr. and P. Samuel, Toll Truckways: Increasing Productivity and Safety in Goods Movement, Rea- son Foundation, http://www.fhwa.dot.gov/download/hep/freightplanning/talkingfreight3_16_05bp.ppt. Mixed Freeway Semitrailers Truckway Semitrailer Truckway Triple Short Truckway Turnpike Double Payload (Pounds) 45,000 45,000 67,500 90,000 Metric Tons 20 20 30 40 100-mi Delivery (2004 Freight Rates) $500 $500 $750 $1,000 Average Speed on the Road 38 mph 60 mph 60 mph 60 mph Miles Driven in 8-h Shift 228 360 360 360 Revenue from 6-h Payload at 2004 Rates $1,140 $1,800 $2,700 $3,600 Variable Costs $684 $684 $1,007 $1,165 Earnings (Revenue—Costs) $456 $1,116 $1,693 $2,435 Earnings (per Ton-Mile) $0.100 $0.155 $0.157 $0.169 Productivity Benefit (% Increase in Earnings per Ton-Mile) (Truck Lanes Compared to Mixed-Flow Lanes) – 55% 63% Productivity Benefit (% Increase in Earnings per Ton-Mile) (Truck Lanes with LCVs Compared to Truck Lanes without LCVs) – – 5% Source: Adapted from Poole, Jr., R. W., and P. Samuel, Toll Truckways: Increasing Productivity and Safety in Goods Movement, Reason Foundation, 2005, http://www.fhwa.dot.gov/download/hep/freightplanning/talkingfreight3_16_05bp.ppt. Table 3.7. Reason Foundation study’s truck-only lane productivity benefits.

• Incremental productivity benefits to trucks from LCV operations on truck-only lanes compared to standard truck operations on truck-only lanes (this provides insights into the productivity benefits associated with increased payloads alone, without taking into consideration the produc- tivity benefits from travel time savings). Performance Results Productivity Benefits. As shown in Table 3.7, the productivity benefits of truckways can be attributed to travel time savings, as well as increased payloads from LCV operations. Assuming an equal split of triple-short and turnpike double trucks under the truck-only lanes with the LCV operations alternative, some of the key findings from the Reason Foundation analysis results on productivity benefits are summarized below. • A large share of the productivity benefits of truck-only lanes are observed to be associated with travel time savings (55% increase in earnings per ton mile from travel time savings compared to 63% from LCV operations, which include the benefits from travel time savings). Thus, the incremental benefits to trucking productivity from LCV operations are observed to be small compared to the benefits from travel time savings. This is because of the assumptions in the Reason Foundation analysis, wherein it is assumed that the no-build alternative experiences significant congestion. • To quantify the relative productivity benefits of LCVs compared to standard truck operations without considering the contribution of travel time savings to productivity improvements, a comparison was made of the earnings per ton mile between the truck-only lane alternative with- out and with LCV operations. The study team observed that LCV operations provide only a 5% increase in earnings per ton mile compared to standard truck operations on truck-only lanes. Travel Time Improvements. The study did not analyze travel time savings. Safety Improvements. The study did not analyze safety benefits. Assumptions and Data Gaps. Some of the key assumptions and data gaps in the Reason Foundation approach that potentially impact the ability to draw conclusions about the actual performance benefits of truck-only lanes are as follow: • The productivity benefits from the study are based on arbitrary assumptions regarding the con- gestion conditions in the no-build and truckway alternatives. The study assumes a very high average level of congestion in the no-build alternative. As a result, a large share of the produc- tivity benefits accrue from travel time savings compared to the benefits from increase in payload. Due to this assumption, the results from the study could potentially be inconclusive in provid- ing insights into the incremental productivity benefits of LCV operations on long-haul corri- dors, particularly among corridors with low congestion levels. • The study does not provide any insights into the diversion potential of truck-only lanes (due to their performance benefits), and only provides estimates for productivity benefits per truck. Thus, the study, although providing a useful analytical framework for the evaluation of produc- tivity benefits of truck-only lanes, can not be used to assess the total productivity benefits of implementing truck-only lanes along a corridor (since these benefits are inherently tied to the total trucks that divert to the truck-only lanes). Western Uniformity Scenario Analysis The Western Uniformity Scenario Analysis, conducted by U.S.DOT in 2004, analyzed the impacts of lifting the LCV freeze and allowing uniformity in LCV operations (weights and dimensions) among western states with current LCV operations based on a key set of performance criteria, including safety, pavement, bridge and infrastructure costs, shipper costs, energy consumption, environmental quality, and traffic operations. To understand the productivity benefits of LCVs 32 Separation of Vehicles—CMV-Only Lanes

compared to standard truck operations on general purpose lanes, this section presents the results from the study on shipper cost savings from uniform LCV operations. The study estimated total annual shipper cost savings resulting from uniform LCV operations in western states of around $2 billion per year, as shown in Table 3.8. The vast majority of these savings would accrue to shippers diverting shipments from standard trucks to LCVs, with minor benefits resulting from rail to LCV diversions and more competitive rail rates provided to shippers. Performance Results Productivity Improvements. According to the study, the implementation of LCV operations on existing facilities (to achieve uniformity in LCV operations in all the western states considered in the analysis) will result in total productivity benefits (in terms of shipper cost savings) of close to 4% (around $2 billion annually) compared to the no-build scenario. These benefits are associ- ated with reduced operating costs to the trucking industry from shifting to LCV operations. Travel Time Improvements. The study did not analyze travel time savings benefits due to LCV operations on mixed-flow lanes. Safety Improvements. The study, although providing a comprehensive comparative discus- sion of accident rates associated with LCVs and non-LCVs, does not quantify the safety benefits of the LCV uniformity alternative compared to the base case (no-build). Assumptions and Data Gaps. Some of the key assumptions and data gaps in this study that potentially impact the ability to gain insights and draw conclusions on the performance benefits of truck-only lanes are discussed below. • The results from this study have been included in the performance evaluation to assess the rel- ative improvements in productivity from LCV operations compared to standard truck opera- tions. Since the two alternatives in the study do not involve any system capacity improvements, the results provide insights into the productivity benefits solely associated with increased pay- loads from LCV operations. • The study quantifies productivity benefits of LCV operations in terms of shipper cost savings, while the total productivity benefits accruing to the trucking industry (in terms of increased earnings from LCV operations, for example) are not reported. Since shipper cost savings only account for a share of the total productivity benefits, the results from the study only provide the lower threshold of the total productivity benefits of LCV operations. • Since the study does not consider CMV-only lanes as part of the LCV uniformity scenario, results from the study cannot be used to assess the performance benefits of truck-only lanes with LCV operations. Performance Evaluation 33 Source of Savings Amount (Millions of 2000 Dollars) Percentage of Change Truck to Truck Diversion 2,036 3.9% Rail to Truck Diversion 3 .01% Rail Discounts 26 .11% Total 2,065 n/a Source: Reprinted from U.S.DOT, Western Uniformity Scenario Analysis: A Regional Truck Size and Weight Scenario Requested by the Western Governors’ Association, April 2004, http://www.fhwa.dot.gov/policy/otps/truck/wusr/wusr.pdf. Table 3.8. Annual shipper cost savings from Western Uniformity Scenario Analysis.

• The study assumes close to 50% diversion of tractor-semi-trailer shipments to LCVs for the LCV scenario, but there is inadequate explanation of the rationale and data to support this assumption. Also, since this assumption is for an LCV uniformity scenario that does not include truck-only lanes, this assumption is expected to not be applicable to a truck-only lane alternative with LCV operations. I-35 Trade Corridor Study Completed in 1999, the I-35 Trade Corridor Study evaluated a set of alternatives with the objective of arriving at recommended corridor investment strategies to improve local, intrastate, interstate, and international service along the I-35 corridor between Laredo, Texas, and Duluth, Minnesota, in the future (2025). The results from the study provide some useful insights into the performance benefits of truck-only lanes along long-haul corridors, compared to alternative investment strategies. The primary performance criteria evaluated in the study included savings in vehicle operating costs, travel time savings, and safety benefits. Performance Results Productivity Improvements. The study quantified the productivity benefits of the truck-only lane alternative in terms of reduction in truck operating costs. The study estimated that the truck- only lane alternative will provide around 36% savings in truck operating costs compared to the no- build alternative. These savings result from increased travel speeds as well as increased payloads from LCV operations on the truck-only lanes. Travel Time Improvements. Travel time savings estimates for the various alternatives were derived using the regional travel demand model, which employs volume-delay functions (VDF) to generate vehicle speeds (and travel times) as a function of congestion parameters such as V/C. The study estimated that the truck-only lane alternative provides around 21% savings in travel time compared to the no-build alternative. Safety Improvements. Since the I-35 study did not estimate percent reductions in acci- dents associated with truck-only lanes, an attempt was made to quantify the approximate percent improvement in safety, based on the estimation of percent reduction in passenger car equivalents (PCEs), assuming a PCE factor of 2.0 for trucks, on the general purpose lanes. The percent change in congestion (V/C) along the corridor for the truck-only lane alternative (relative to the alter- native without truck-only lanes) was assumed to be the same as the percent change in PCEs, since the general purpose lane capacity along the corridor does not change under the two alternatives, and safety benefits are directly proportional to change in corridor congestion. Results from the study indicate an average safety improvement of 38% along the corridor due to truck-only lanes. Since these benefits are solely associated with congestion reduction and do not include the safety benefits of truck-auto separation, an additional accident reduction factor of 15% (as recommended by the Douglas handbook) was applied to the results, providing a total percent reduction in acci- dents for the CMV-only lane alternative of around 47% compared to the no build alternative. Assumptions and Data Gaps. Some of the key assumptions and data gaps in this study that potentially impact the ability to draw conclusions about the actual performance benefits of truck- only lanes are discussed below. • The estimates in the study on the diversion of trucks to the truck-only lanes are observed to be very optimistic, compared to the diversion rates under LCV operations derived in other studies41 (note that the study referenced is ongoing, and the results from the study on LCV diver- sion rates are potentially subject to change as more detailed analyses of the diversion potential 34 Separation of Vehicles—CMV-Only Lanes 41FHWA, Technological Challenges and Policy Implications for LCVs on Exclusive Truck Facilities, I-90 Gap Clos- ing Scenario, Draft Evaluation Results.

of LCV operations are conducted). The study estimated that the trucks on the truck-only lanes (adjusted for LCVs) accounted for 79% of total trucks. This implies that the assumption in the study on the actual diversion of trucks to the truck-only lanes would be greater than 79%, since the adjustment for LCVs would result in a net reduction of total truck volumes on the truck- only lanes. Although part of this diversion would be driven by congestion relief considerations, the diversion would also be impacted by productivity improvements offered by LCV operations, the propensity of the trucking market (type of commodities and origination-destination [O-D] patterns) toward shifting to LCVs, and other LCV operational considerations (e.g., equipment and operational costs). It is unclear if these issues associated with LCVs were considered in the study in arriving at the diversion rates (and corresponding utilization of the truck-only lanes). Since the performance benefits of the truck-only lane alternative are directly linked with the assumptions on the level of diversion of trucks from the general purpose lanes, it is difficult to draw conclusions from the results of this study on the performance benefits of truck-only lanes without adequate information on the inherent factors driving these diversion estimates. • The study uses a regional travel demand model to quantify the travel time savings benefits of the truck-only lane alternative associated with speed improvements on the general purpose lanes from diversion to truck-only lanes. This approach could potentially underestimate the travel time savings benefits of the truck-only lane alternative, since travel demand models typically underestimate the congestion relief impacts of truck diversion based on their assump- tions on truck PCE factors. • The alternatives defined in the study prevented the ability to compare the performance benefits of the truck-only lane alternative against an alternative with additional mixed-flow capacity (Appendix B provides a detailed description of the alternatives considered in this study.). • The study does not consider the impacts of differences in time-of-day distributions between trucks and autos on the utilization and associated performance benefits of the truck-only lane alternative, which could potentially impact the results reported in the study on the relative per- formance benefits of the truck-only lanes. Georgia Statewide Truck Lane Needs Identification Study The Georgia Statewide Truck Lane Needs Identification Study (i.e., the Georgia study) was con- ducted to evaluate the feasibility of implementing truck-only lanes on Georgia’s statewide highway network. The main objectives of the study included quantifying the performance benefits of truck- only lanes (relative to an alternative without truck-only lanes), identifying potential corridors for implementation (based on certain feasibility criteria such as truck volumes, congestion, and mar- ket accessibility), and assessing the benefits and costs of implementing truck-only lanes. The hori- zon year for the study was 2035. The initial phase of the study considered all the major Interstate facilities and access controlled state routes in Georgia for the truck-only lane needs identification analysis. An initial screening process (based on a qualitative performance evaluation process) was undertaken to evaluate the feasibility of truck-only lanes along these corridors, which resulted in the identification of a set of “candidate corridors” showing the greatest potential for the implementation of truck-only lanes to meet the freight and transportation needs in the state. The candidate long-haul corridors identi- fied in the study to have the greatest potential for truck-only lanes included the following: • I-75 (southern segment) between I-285 (south end) and I-475 (near Macon). This segment was divided into the following subsegments: – Segment 3A: Henry/Butts County line to I-285 and – Segment 3B: I-475 to Henry/Butts County line. • I-75 (northern segment) between I-285 (north end) and Georgia/Tennessee boundary. This segment was divided into the following subsegments: – Segment 4A: I-285 to Bartow/Gordon County line and – Segment 4B: Bartow/Gordon to Tennessee. Performance Evaluation 35

• I-85 (southern segment) between I-285 and Georgia/Alabama boundary. This segment was divided into the following subsegments: – Segment 6A: Alabama to Coweta/Troup County line and – Segment 6B: Coweta/Troup County line to I-285. • I-85 (northern segment) between I-285 and Georgia/South Carolina boundary. This segment was divided into the following subsegments: – Segment 7A: I-285 to Gwinnett/Jackson County line and – Segment 7B: Gwinnett/Jackson County line to South Carolina. • Segment 8: I-20 (western segment) between I-285 and Georgia/Alabama boundary. • Segment 9: I-20 (eastern segment) between I-285 and Georgia/South Carolina boundary. Appendix B provides a detailed description of the candidate corridors, auto and truck traffic demand along these corridors and how these were generated, and the performance benefits esti- mates for the alternatives considered in the study. Performance Results Productivity Benefits. The study did not look at productivity benefits as an exclusive perfor- mance metric in evaluating the performance benefits of truck-only lanes. Therefore, productivity benefits were derived using a post-processing analysis based on the Reason Foundation approach to estimating productivity benefits of truck-only lanes presented in Table 3.7. This post-processing analysis used the speeds from the Georgia study as inputs along with other assumptions on freight rates and truck variable costs (consistent with the assumptions used by the Reason Foundation and presented in Table 3.7), to derive the relative increase in trucking industry earnings due to truck- only lanes. Table 3.9 presents the productivity benefits to truckers for a select set of corridor seg- ments (Segments 3B, 4A, 4B, 6B, and 7A) due to usage of truck-only lanes (without and with LCV operations) compared to the no-build alternative. Although the study did not consider LCV operations, the post-processing analysis also consid- ered a truck-only lane alternative with LCV operations, to assess the incremental productivity ben- efits of truck-only lanes with LCV operations (due to increased payloads). The productivity benefits results are summarized below. • Truck-only lanes without LCVs more than double the productivity of trucking operations (in terms of increased annual trucking industry earnings) compared to the no-build alternative; • Truck-only lanes with LCV operations provide close to 7% incremental productivity benefits compared to truck-only lanes without LCV operations, due to the productivity benefits of increased payloads; and • The incremental productivity benefits due to increased payloads are observed to be significantly lower compared to the productivity benefits from travel time savings on the truck-only lanes, due to high congestion conditions on many of the intercity corridor segments, particularly those falling within the outer-limits of the Atlanta metropolitan area. Travel Time Improvements. Travel time savings of truck lanes are estimated as the savings in vehicle hours traveled (VHT) between the no-build and the truck-only lane alternatives. Total VHT for autos and trucks are derived for the study alternatives using travel demand models for each of the candidate corridor segments, and these results are presented in Table 3.10. VHT savings are considered to be representative of travel time savings under the assumption that traffic volumes do not change significantly between the no-build and the truck-only lane alternatives. Truck lanes along the intercity corridors considered in the study are estimated to provide 20% savings in travel time compared to the no-build alternative. Safety Improvements. Safety performance evaluation of alternatives was conducted in terms of change in the number of fatal accidents, using crash rate data from the Georgia DOT, as a func- 36 Separation of Vehicles—CMV-Only Lanes

Segment 3B: I-75 South Segment 4A: I-75 North Segment 4B: I-75 North Segment 6B: I-85 South Segment 7A: I-85 North Without TOL2 No LCV3 LCV4 Without TOL No LCV LCV Without TOL No LCV LCV Without TOL No LCV LCV Without TOL No LCV LCV Speeds (mph) 61 64 64 27 52 52 56 62 62 38 57 57 29 51 51 Freight Rate (per 100 mi) 500 500 875 500 500 875 500 500 875 500 500 875 500 500 875 Miles per 8-h Day Shift (6 h Driving) 366 384 384 162 312 312 336 372 372 228 342 342 174 306 306 Revenue per Day Shift 1,830 1,920 3,360 810 1,560 2,730 1,680 1,860 3,255 1,140 1,710 2,993 870 1,530 2,678 Variable Costs 684 684 1,086 684 684 1,086 684 684 1,086 684 684 1,086 684 684 1,086 Net Earnings 1,146 1,236 2,274 126 876 1,644 996 1,176 2,169 456 1,026 1,907 186 846 1,592 Earnings per Ton Mile 0.16 0.16 0.17 0.04 0.14 0.15 0.15 0.16 0.17 0.1 0.15 0.16 0.05 0.14 0.15 % Increase in Earnings per Ton Mile 3 8 261 287 7 12 50 59 159 178 Source: Cambridge Systematics, Inc. (based on the Reason Foundation approach and data inputs from the Georgia study). Notes: 1. Productivity benefits of truck-only lanes are estimated separately in the table for each of the major corridor segments in the Georgia study. The corridor segments in the table include 3B, 4A, 4B, 6B, and 7A. The key data inputs used for estimating productivity benefits include speeds, freight rates, and variable costs. Speed data is derived from the Georgia study, while assumptions on freight rates and variable costs are taken from the Reason Foundation methodology for the estimation of productivity benefits. 2. Refers to the no-build alternative (without truck-only lanes [TOL]). 3. Refers to the truck-only lane alternative with standard truck (no LCV) operations. 4. Refers to the truck-only lane alternative with LCV operations. Table 3.9. Productivity benefits estimates of truck-only lanes using the Reason Foundation methodology applied to data from the Georgia study,1 2035.

Without Truck-Only Lanes With Truck-Only Lanes % Travel Time Savings1 Corridor Segments Facility2 4-Mile Buffer3 12-Mile Buffer4 Region5 Facility 4-Mile Buffer 12-Mile Buffer Region Facility 4-Mile Buffer 12-Mile Buffer Region 3A: I-75 South 0.3 1.0 2.6 10.3 0.2 0.8 2.5 10.1 42 15 7 2 3B: I-75 South 0.1 N/A N/A 38.1 0.1 N/A N/A 38.1 0 N/A N/A 0 4A: I-75 North 0.3 1.1 3.3 10.3 0.2 1.0 3.1 10.0 23 11 5 2 4B: I-75 North 0.1 N/A N/A 38.1 0.1 N/A N/A 38.1 8 N/A N/A 0 6B: I-85 South 0.1 0.5 1.9 10.3 0.1 0.5 1.8 10.3 9 6 1 0 7A: I-85 North 0.2 1.1 3.7 10.3 0.2 1.0 3.6 10.1 11 5 3 1 8: I-20 West 0.2 0.7 2.5 10.3 0.1 0.6 2.8 10.2 20 8 -15 1 9: I-20 East 0.1 0.6 2.6 10.3 0.1 0.6 2.6 10.2 13 2 1 0 Total VHT and Average Savings 1.2 4.9 16.5 137.7 1.0 4.5 16.4 137.0 20 9 1 0 Source: Adapted from Georgia Department of Transportation, Statewide Truck Lane Needs Identification Study, Technical Memorandum 3: Truck-Only Lane Needs Analysis and Engineering Assessment, April 2008. 1. Percent savings in travel time are calculated as the percent reduction in VHT between the “with truck-only lanes” and “without truck-only lanes” alternatives. 2. The cells in this column represent daily VHT (in millions) for users on each of the Interstate corridor segments. 3. The cells in this column represent daily VHT (in millions) for users traveling on roadways within a 4-mi buffer area on either side of each of the Interstate corridor segments. 4. The cells in this column represent daily VHT (in millions) for users traveling on roadways within a 12-mi buffer area on either side of each of the Interstate corridor segments. 5. The cells in this column represent daily VHT for users on the entire roadway network (within the study area). Table 3.10. Travel time savings due to truck-only lanes, estimates from the Georgia study, 2035 (percentage change in VHT [millions]).

tion of roadway facility type and congestion. Clearly, the method used for accident estimation in the study did not consider the incremental safety benefits associated with truck-auto separation. As a result, additional post-processing was conducted to account for the safety benefits of truck- auto separation, by applying an additional 15% accident reduction factor (as recommended by the Douglas handbook). Table 3.11 presents the safety benefits estimates from the study. Truck lanes along the intercity corridor segments considered in the study are estimated to reduce accidents by 44% compared to the no-build alternative, as a result of congestion reduction as well as separation of trucks and autos. Assumptions and Data Gaps. Some of the key assumptions and data gaps in this study that potentially impact the ability to draw conclusions about the performance benefits of truck-only lanes are as follow: • The study only considers the performance benefits of truck lanes compared to a no-build alter- native and does not provide insights into the relative performance of truck lanes compared to adding mixed-flow capacity. This appears to be particularly relevant, since some of the corridor segments considered in the study experience significant congestion. • The study uses travel demand models to estimate the travel time savings benefits of truck lanes compared to the no-build alternative. Since trucks account for a large share of total traffic vol- umes along most of the corridor segments (close to 40% on average), the estimation of the con- gestion relief benefits of eliminating trucks from the general purpose lanes has to accurately consider the congestion impacts of trucks (when trucks represent a large share of the total traf- fic volumes) through the use of representative PCE factors. Typically, this is a limitation with travel demand models, which do not consider variable PCEs. Consequently, the results from the study could be underestimating the travel time savings of truck lanes, particularly along the con- gested corridor segments. • The safety benefits results from the study are based on the congestion reduction benefits of truck lanes, and the safety benefits of truck-auto separation are accounted for by applying an addi- tional accident reduction factor as part of a post-processing analysis. Clearly, the safety benefits estimates are not based on robust analytical tools (such as simulation) that can capture the true safety benefits of truck-auto separation. Performance Evaluation 39 Estimated Annual Fatal Accidents Corridor Without TOL With TOL 15% Additional Adjustment for TOL Percentage of Savings in Accidents 3A: I-75 South 11 7 6 46% 3B: I-75 South 10 7 6 41% 4A: I-75 North 14 9 8 45% 4B: I-75 North 10 7 6 41% 6B: I-85 South 9 6 5 43% 7A: I-85 North 10 7 6 41% 8: I-20 West 10 6 5 49% 9: I-20 East 8 5 4 47% Total Fatal Accidents and Average % Savings 82 54 46 44% Source: Adapted from Georgia Department of Transportation, Statewide Truck Lane Needs Identification Study, Technical Memorandum 3: Truck-Only Lane Needs Analysis and Engineering Assessment, April 2008. Table 3.11. Safety benefits due to truck-only lanes, 2035 (fatal accident reduction).

• The study does not consider differences in time-of-day distributions between trucks and autos, which could impact the performance of truck-only lanes. 3.3.2 Urban Corridors I-710 Major Corridor Study The I-710 Major Corridor Study was initiated in 2001 to analyze future year traffic volumes, con- gestion, safety, and environmental issues along the I-710 corridor in Southern California, with the objective of developing transportation solutions to address these issues. Traffic forecasts along the corridor for each of the alternatives were generated using a subarea travel demand model for the I-710 study area. These forecasts served as inputs in the estimation of key performance mea- sures, including V/Cs, speeds, travel times, and number of accidents. The following sections describe the results from the study related to travel time savings, reliability, and safety. Additional post-processing was conducted to estimate reliability (since the study did not esti- mate these benefits), and safety (this was done for the truck-only lane alternative since the study did not specifically account for the safety benefits of truck-auto separation). Appendix B provides a detailed description of the alternatives, performance measures considered, and performance ben- efits results from the study for each of the alternatives. The following sections summarize the key performance results from the study to gain insights into the relative performance of truck-only lanes compared to no-build and additional mixed-flow lane alternatives. Performance Results Travel Time Savings. The study evaluated mobility performance among alternatives in terms of speeds on general purpose lanes (and speeds on truck lanes in the case of the truck-only lane alternative) for each of the alternatives that were derived using a subarea travel demand model. These speeds were translated into equivalent travel time savings as part of a post-processing analy- sis. The relative savings in travel time for the build alternatives compared to the no-build alterna- tive are presented in Table 3.12. Following are some insights into the travel time savings benefits of truck-only lanes estimated in the study: • For the truck-only lane alternative, the travel time savings for trucks using the truck-only lanes are significantly higher (more than double) than the savings for autos and trucks on the general 40 Separation of Vehicles—CMV-Only Lanes Alternatives General Purpose Lanes Truck/Carpool Lanes TSM/TDM1 8% N/A Mixed-Flow Lanes (One Lane in Each Direction) 14% N/A Mixed-Flow Lanes (Two Lanes in Each Direction) with Additional HOV Lanes2 21% 37% Truck-Only Lanes 16% 33% Source: Adapted from Los Angeles County MTA, I-710 Major Corridor Study—Final Report, March 2005. Notes: 1. Includes improvement strategies such as added bus service for local area communities, ramp metering system on I-710, advanced technologies for traffic management, and motorist information systems for route choice decision making based on traffic congestion. 2. Four additional bidirectional lanes between SR 91 and SR 60 and six additional bidirectional lanes between Ocean and SR 91. Table 3.12. Percent travel time savings compared to no build, 2025 (northbound lanes, P.M. peak period).

purpose lanes. This is because of the lower level of congestion on the truck-only lanes compared to the congestion relief provided on the general purpose lanes due to truck diversion. • The relative travel time savings benefits on the general purpose lanes between the truck-only lane and mixed-flow lane (one lane in each direction) alternatives are observed to be not that signif- icant, even though the truck-lane alternative has higher capacity. This could be attributed to the lower level of contribution of trucks to congestion in the P.M. peak period. Reliability Benefits. The I-710 study did not estimate the reliability benefits associated with each of the build alternatives. In order to estimate percentage change in reliability for each of the build alternatives compared to the no-build, the study team used a post-processing approach. This approach involves estimating total nonrecurrent/incident-related delays for each of the alter- natives, and comparing these estimates for the build alternatives against the no-build alternative to arrive at percent change in incident-related delays. Total nonrecurrent delay per vehicle mile is estimated using post-processing factors as a func- tion of V/C and number of lanes. The approach involves estimating nonrecurrent delay per vehi- cle mile for various sections of the I-710 corridor based on information on V/C and number of lanes from the I-710 Major Corridor Study. The nonrecurrent delay estimates per vehicle mile are then averaged out for the corridor, and multiplied by the total VMT along the corridor (from the I-710 Major Corridor Study) for each of the alternatives. IDAS post-processing factors do not consider the improvements in travel time reliability for the truck lane alternative associated with truck-auto separation. Since nonrecurrent delays are directly proportional to, and can be assumed to have a linear relationship with, the number of accidents, a 15% accident reduction factor (as recommended by the Douglas handbook) is applied to the non- recurrent delay estimates for the truck-only lane alternative from the I-710 study. The final esti- mates for percentage change in reliability for the build alternatives in the I-710 study compared to the no-build alternative are presented in Table 3.13. The percentage change in reliability for the truck lane alternative in Table 3.13 includes relia- bility improvements both on the GP and truck lanes. Since reliability improvements on the truck lanes are expected to be significantly higher than those for the GP lanes (as was observed from the Georgia study), the reliability improvements on the GP lanes for the truck-only lane alternative are expected to be lower than the 59% figure for the truck-only lanes. Safety Benefits. The I-710 study analyzed safety benefits among alternatives in terms of reduc- tion in accidents under each of the build alternatives compared to the no-build alternative. These results are presented in Table 3.14. In order to determine percent improvements in safety relative to the no-build alternative, addi- tional post-processing of the results from the I-710 study was conducted, using IDAS safety factors Performance Evaluation 41 Alternatives Total Annual Savings in Nonrecurrent Delay (Hours) Percentage of Reliability Improvement A. No Build – – B. TSM/TDM 2,375 15% C. Mixed-Flow Lanes (One Lane in Each Direction) 7,403 47% D. Mixed-Flow Lanes (Two Lanes in Each Direction) with Additional HOV Lanes 9,935 63% E. Truck-Only Lanes 9,308 59% Source: Adapted from Los Angeles County MTA, I-710 Major Corridor Study—Final Report, March 2005. Table 3.13. Percent change in travel time reliability.

as a function of V/C, to determine annual accidents under the no-build alternative. The approach and results from this post-processing are shown in Table 3.15. Combining the results from Table 3.14 and Table 3.15, Table 3.16 presents the percent improve- ment in safety for each of the build alternatives. For the truck-only lane alternative, an additional 15% reduction factor was applied to the total accidents to account for the safety benefits of truck- auto separation (as recommended by the Douglas handbook). 42 Separation of Vehicles—CMV-Only Lanes Alternatives Reduction in Accidents* B 316 C 554 D 480 E 539 * When compared to Alternative A. Source: Adapted from Los Angeles County MTA, I-710 Major Corridor Study—Final Report, March 2005. Table 3.14. Annual accident reduction (2025). V/C 1.42 Daily VMT 4,400,000 Fatality Rate (Accidents per Million VMT) 0.0066 Injury Rate (Accidents per Million VMT) 0.71 Property Damage Rate (Accidents per Million VMT) 0.9192 Daily Fatality Accidents 0 Daily Injury Accidents 3 Daily Property Damage Accidents 4 Total Daily Accidents 7 Total Annual Accidents (Assuming a Factor of 300) 2,159 Source: Adapted from Los Angeles County MTA, I-710 Major Corridor Study—Final Report, March 2005. Table 3.15. Annual accidents under the no-build alternative, 2025. Alternatives Annual Accidents Additional Reduction Due to Truck-Auto Separation Net Annual Accidents Percentage of Accident Reduction No Build 2,159 N/A 2,159 – TSM/TDM 1,843 N/A 1,843 15% Mixed-Flow Lanes (One Lane in Each Direction) 1,605 N/A 1,605 26% Mixed-Flow Lanes (Two Lanes in Each Direction) with Additional HOV Lanes 1,679 N/A 1,679 22% Truck-Only Lanes 1,620 243 1,377 36% Source: Adapted from Los Angeles County MTA, I-710 Major Corridor Study—Final Report, March 2005. Table 3.16. Percent safety improvement, 2025.

Some key insights from the performance benefits results from the I-710 study are as follow (potential limitations associated with these results are discussed in the next section): • The study provides insights into the relative performance of truck-only lanes compared to no-build and additional mixed-flow lane alternatives for travel time savings, reliability, and safety performance measures; • Relative improvements in reliability and safety benefits of truck-only lanes (compared to additional mixed-flow lanes) are observed to be higher compared to travel time savings because, in addition to congestion relief (which contributes to reliability and safety improve- ments), the separation of trucks and autos in the truck-only lane alternative also provides incremental reliability and safety benefits; and • The truck-only lane alternative provides the highest safety benefits compared to all other alternatives. Assumptions and Data Gaps. Some of the key assumptions and data gaps in this study that potentially impact the ability to draw conclusions about the performance benefits of truck-only lanes are as follow: • The mixed-flow and truck-only lane alternatives considered in the study have different capac- ities. Therefore, without a B-C analysis, it is difficult to assess, from the results, the relative over- all performance of truck-only lanes compared to adding mixed-flow capacity. • The travel time savings benefits are calculated using the speed outputs from a subarea travel demand model. As discussed earlier, the applications of travel demand models to analyze the mobility performance of truck-only lanes have potential limitations, especially in congested corridor conditions (such as in the case of the I-710 corridor), particularly regarding assump- tions related to truck PCE factors, which could result in an underestimation of the actual travel time savings benefits of truck-only lanes. • The study provides mobility performance results for the build alternatives in terms of average speeds for the P.M. peak period, which are translated into travel time savings benefits as part of a post-processing analysis. However, travel time savings for only the P.M. peak period are inad- equate in assessing the performance of truck-only lanes since differences in time-of-day patterns of trucks in the A.M. peak and mid-day time periods could potentially affect the performance of the truck-only lane alternative relative to adding mixed-flow capacity. Georgia Statewide Truck Lane Needs Identification Study The Georgia Statewide Truck Lane Needs Identification Study used a two-pronged approach in the performance evaluation and needs analysis process for truck-only lanes. The first step in the process involved the analysis of individual corridor segments and the quantification of perfor- mance benefits of truck lanes on these corridors. The results from this analysis for the long-haul corridors in the state were discussed previously in the long-haul corridor performance evaluation section. The second step involved the development of truck-only lane systems (combination of individual corridors) and a detailed evaluation of the performance benefits of truck-only lane sys- tems compared to a system without truck-only lanes. The performance measures evaluated in the study for the analysis of truck-only lanes along the above-mentioned corridor systems included travel time savings, reliability, and safety benefits. The study used a combination of the Georgia statewide model, the Atlanta Regional Commission (ARC) travel demand model, and the Savannah Metropolitan Planning Organization (MPO) model to generate traffic forecasts on the corridor systems, and estimate travel time and reliabil- ity performance measures for each alternative. Safety performance was analyzed using crash rates available from GDOT, in terms of number of accidents using V/C and VMT outputs from the model. Appendix B provides a detailed description of the metropolitan corridor systems, alterna- tives, and performance measures considered in the study. The following sections present the key performance benefits results from the study to gain insights into the relative performance of truck- only lanes compared to the no-build alternative. Performance Evaluation 43

Performance Results Travel Time Savings. Travel time savings due to truck-only lanes were evaluated in terms of changes in VHT in 2035 for each of the corridor systems. Analysis of change in VHT to evaluate travel time savings provides a conservative estimate for travel time savings, because part of the VHT change is contributed by increase in traffic volumes on the corridor due to increased capac- ity. Table 3.17 summarizes these results. The average 2035 change in VHT due to the implemen- tation of truck-only lanes, by location, is as follows: • Facility: −17%, • Corridor Buffer—4 mi: −11%, • Corridor Buffer—12 mi: −8%, and • Region: −6%. 44 Separation of Vehicles—CMV-Only Lanes Source: Adapted from Georgia Department of Transportation, Statewide Truck Lane Needs Identification Study, Technical Memorandum 3: Truck-Only Lane Needs Analysis and Engineering Assessment, Tables 43, 52, and 51, April 2008. Fatal Injury Total Region Facility GP Lanes Truck- Only Lanes Corridor Buffer = 12 Miles Corridor Buffer = 4 Miles 2030 Estimated Annual Crashes Buffer Index System 1 I-75/I-20W/ I-285W System 2 I-75/I-675/ I-20W/I-285W System 3 I-75/I-85N/ I-20W/I-285 All System 4 I-75/I-85N/ I-675/I-285 E Truck- Only Lanes Truck- Only Lanes Truck- Only Lanes Truck- Only Lanes No Project No Project No Project No Project Percent Change Percent Change Percent Change Percent Change Vehicle Hours Traveled per Day (Millions) Reduction in Fatal Crashes -40% -1% 0% -40% -1% 0% -39% -1% 0% -43% -1% 0% 18 12 1,686 7,860 30 1,727 7,955 23 15 2,262 10,531 38 2,288 10,540 30 20 2,910 13,553 50 2,944 13,564 17 13 1,705 7,948 30 1,708 7,867 -19% -14% -15%-19%0.48 0.59 0.680.800.850.980.480.59 -11% -11% -10%-11%2.37 2.81 3.343.704.114.602.502.68 -8% -9% -8%-8%5.48 6.20 7.217.867.618.395.715.97 -5% -8% -7%-5%10.11 10.66 9.9310.669.8210.6610.1210.66 -42% -35% -34%-43%72% 124% 82%124%77%118%70%124% -81% -78% -80%-79%43% N/A 45%N/A40%N/A44%N/A Table 3.17. Change in daily VHT (2035), buffer index (2035), and estimated annual crashes (2030).

Reliability Benefits. Reliability benefits in the study were evaluated in terms of change in the Buffer Index which is the ratio of the extra time travelers must build into the trip when planning their travel (to ensure reaching their destination on time 95% of the time), to the average travel time. The Buffer Index was estimated as a function of the Travel Time Index (TTI, which is defined as the ratio of the congested travel time to the free-flow travel time), using model outputs for con- gested and free flow travel times for the no-build and truck-only lane alternatives. The reliability benefits (in terms of percent change in the Buffer Index) were evaluated both for trucks using the truck lanes, as well as for autos and trucks using the general purpose lanes. See Table 3.17. Under the no-project scenarios, travelers allowed for significantly high buffer times (118% to 124%) to reach their destination on time 95% of the time (implying that their total travel times were more than double their average travel times). The implementation of truck lanes was esti- mated to result in significant reduction in buffer times for trucks on truck lanes (percent reduc- tion of 80%). The reduction in buffer times for autos and trucks on the general purpose lanes were not as significant as on the truck lanes, but still was observed to be notable (percent reduc- tion of close to 40%). The 2035 average percentage reduction in buffer times based on the results is as follows. • Trucks on truck-only lanes = 80%. • Autos/trucks on general purpose lanes = −39%. Safety Benefits. Safety benefits were evaluated in terms of percent reduction in injury and fatal- ity accidents due to the implementation of truck lanes compared to the no-build alternative for each of the corridor systems. See Table 3.17. Since the Georgia study did not specifically account for the safety benefits of truck-only lanes accruing from truck-auto separation, a 15% reduction factor was applied to the total, fatality, and injury accidents estimates for the truck-only lane alter- native. These results are presented in Table 3.18. Average fatality accident reduction of 50% due to the implementation of truck-only lanes is observed. The percent reduction for injury accidents is estimated to be lower (average reduction of 16%), which is as expected, since a large share of the fatality accidents are influenced by the involvement of trucks (while this share is significantly lower for injury accidents). Assumptions and Data Gaps. Since the performance results from the study for the urban cor- ridors scenario are based on the same procedures and tools as described for the study under the long-haul corridor scenario, the assumptions/data gaps and their impacts on the ability to assess the relative performance of truck-only lanes described under the long-haul corridor scenario are applicable here as well. The procedure used to derive the reliability benefits estimates from the study, using the Buffer Index as the performance metric, has the following potential impacts on the accu- racy of the results: • The Buffer Index is estimated as a function of the Travel Time Index, which implies that the analysis only considers the reliability benefits associated with congestion reduction without con- sidering the benefits of truck-auto separation. • The Travel Time Index estimates are derived from model outputs for congested and free flow travel times, and the limitations of travel demand models in assessing the mobility performance of truck-only lanes have been described earlier. Thus, the Travel Time Index estimates from the model for the truck-only lane alternative may be miscalculated, which would impact the accu- racy of the Buffer Index estimates. Puget Sound Region Freight Action Strategy (FAST) Corridor Analysis The FAST corridor analysis project in the Puget Sound region involved the development and application of a regional truck model (the FAST truck model) to evaluate the benefits associated Performance Evaluation 45

with a variety of transportation investments impacting goods movement in the four-county Puget Sound region. The analysis involved conducting model runs for a set of alternatives and comparing model outputs for a set of performance measures against the future no-build alter- native. The primary performance measures evaluated in the study included travel time savings in terms of change in VHT, and change in delays (congested travel time—free-flow travel time). The following alternatives were considered in the study, in addition to the no-build alternative: • Alternative 1. “Operational” improvements of facilities, such as upgrading arterials to free- ways, interchange improvements, and capacity improvements for trucks; • Alternative 2. “Infrastructure–related” improvements consisting mainly of adding general purpose lanes and truck lanes along the larger corridor network; • Alternative 3. Addition of truck-only lanes along the I-405 corridor; • Alternative 4. Addition of truck-only lanes along the I-5 corridor; and 46 Separation of Vehicles—CMV-Only Lanes No-Builda Truck-Only Lanesb Truck-Only Lanes (After 15% Reduction)c Safety Benefits (Percentage of Reduction due to Truck-Only Lanes)d Fatality Accidents System 1e 30 18 15 49% System 2 30 17 14 52% System 3 50 30 26 49% System 4 38 23 20 49% Average 50% Injury Accidents System 1 1,708 1,686 1,433 16% System 2 1,727 1,705 1,449 16% System 3 2,944 2,910 2,474 16% System 4 2,288 2,262 1,923 16% Average 16% Total Accidents System 1 7,867 7,860 6,681 15% System 2 7,955 7,948 6,756 15% System 3 13,564 13,553 11,520 15% System 4 10,540 10,531 8,951 15% Average 15% Source: Adapted from Georgia Department of Transportation, Statewide Truck Lane Needs Identification Study. Notes: a. The cells in this column represent number of accidents for the no-build alternative. b. The cells in this column represent number of accidents for the truck-only lane alternative. c. The cells in this column represent the number of accidents for the truck-only lane alternative after accounting for accident reduction due to truck-auto separation (a 15% reduction factor is applied to the accidents in the “Truck-Only Lanes” column based on recommendations from the Douglas handbook). d. The cells in this column represent the % reduction in accidents (safety benefits) due to truck-only lanes. This is estimated from the accidents under the “Truck-Only Lanes (after 15% reduction)” and “No Build” columns. e. Systems 1 through 4 represent combinations of individual corridors analyzed in the Georgia study. Table 3.18. Safety benefits estimates of truck-only lanes using accident data from the Georgia study, 2030.

• Alternative 5. Changes in land-use patterns for the 2020 no-build scenario, and associated impacts. Appendix B provides a detailed description of the performance measures and benefits results from this study for each of the alternatives. The following sections summarize the key perfor- mance benefits results from the study. Performance Results Travel Time Savings. Travel time savings is the only performance measure quantified in the study as part of the alternatives analysis process. As mentioned previously, these benefits are gen- erated in terms of percent change in VHT for each of the build alternatives compared to the no- build alternative using VHT outputs from the FAST truck model. Figure 3.2 presents the VHT results from the alternatives analysis. Table 3.19 summarizes the percentage change in VHT for the truck-only lane alternatives (Alter- natives 3 and 4) compared to the no-build alternative, by vehicle class and total vehicles. Assuming the total VMT to be relatively the same along the I-5 and I-405 corridors, average travel time savings from truck-only lanes are estimated to be around 9% compared to the no-build alternative. Performance Evaluation 47 -25% -20% -15% -10% -5% 0% 5% 10% Auto Light Trucks Medium Trucks Heavy Trucks Total Vehicles Alternative Pe rc en t C ha ng e i n D el ay H ou rs fr om 20 20 B as el in e Alternative #1 Alternative #2 Alternative #3 Alternative #4 Alternative #5 Source: Adapted from Arun R. Kuppam and Maren L. Outwater, Evaluating Freight Mobility on a Regionwide Basis Using EMME/2 —Freight Action Strategy (FAST) Truck Model for Puget Sound Region, submitted for the 16th International EMME/2 User’s Group Conference, Albuquerque, NM, March 18–20, 2002. Figure 3.2. Percentage change in vehicle hours traveled (VHT) from 2020 future baseline. Autos (%) Light- Heavy (%) Medium- Heavy (%) Heavy- Heavy (%) Total Vehicles (%) Truck-Only Lanes (Alternative 3) -10 -4 -10 -8 -8 Truck-Only Lanes (Alternative 4) -14 6 -19 -23 -10 Source: Adapted from PSRC FAST corridor study. Table 3.19. Change in VHT, truck-only lanes, compared to no-build, 2020.

Assumptions and Data Gaps. Some of the key assumptions and data gaps in this study that potentially impact the ability to draw conclusions about the performance benefits of truck-only lanes are as follow: • The study does not consider an exclusive mixed-flow lane alternative to allow for the compar- isons of mobility performance benefits of truck-only lanes compared to adding mixed-flow capacity. • A travel demand model (FAST truck model) is used to evaluate the travel time savings ben- efits of truck-only lanes. The limitations of travel demand models in assessing the mobility performance of truck-only lanes has been discussed earlier. • The study does not consider other key performance measures such as safety and travel time reliability in the alternatives analysis, which are important metrics to consider, particularly as part of the performance evaluation of truck-only lanes. • The study does not consider differences in time-of-day distributions between trucks and autos in analyzing the travel time savings benefits of truck-only lanes by time of day. An assessment of the actual mobility benefits of truck-only lanes should be based on a time-of-day analysis, to take into consideration the inherent differences in time-of-day patterns of trucks and autos in an urban corridor environment. I-15 Comprehensive Corridor Study The I-15 study was sponsored by the Southern California Association of Governments (SCAG), the California Department of Transportation (Caltrans), and the San Bernardino Associated Governments (SANBAG) with the primary objectives of analyzing right-of-way needs along the corridor, assessing the feasibility and costs of implementing truck lanes, and performing a com- prehensive evaluation of transportation needs along the corridor to feed the development of a long-range improvement plan and implementation strategy for the corridor. The study con- ducted an initial screening evaluation of a comprehensive list of alternatives, which were narrowed down to a final set of five alternatives for detailed screening and alternative selection process: no- build, TSM/TDM, HOV lanes, full corridor dedicated truck lanes, and reversible managed lanes. Traffic (truck and auto) volume forecasts and a mobility performance measure (V/C) for each of the alternatives were generated using the 2004 SCAG Regional Transportation Plan (RTP) model. Appendix B provides a detailed description of the alternatives, and performance benefits results estimated in the study. Table 3.20 shows the 2030 forecast for truck and auto volumes along the corridor by segment (the study analyzed travel demand along the corridor for seven segments) under each of the alter- natives. Following are some key insights from the data: • Truck volumes are projected to represent a notable share (more than 25%) of the total traffic volumes along most segments of the corridor. • The data show an increase in auto traffic volumes along the corridor for each of the build alter- natives (HOV lanes, truck lanes, and managed lanes) relative to the no-build alternative due to the availability of additional capacity and the shifting of demand onto the corridor from adja- cent facilities. In the case of the HOV lane alternative, this shifting of demand compensates for the reduction in auto traffic along the corridor due to shifting of demand from single- occupancy to multiple-occupancy vehicles. • Auto volumes are projected to be slightly lower for the TSM/TDM alternative compared to the no-build alternative because of the implementation of increased transit service under this alternative. • There is only a marginal increase in truck volumes under the build alternatives relative to the no-build alternative. The following sections summarize the key performance benefits results from the study to gain insights into the performance benefits of truck-only lanes. 48 Separation of Vehicles—CMV-Only Lanes

Performance Results Travel Time Savings. The I-15 study analyzed the mobility performance of each of the alter- natives based on the estimation of V/C, using the SCAG RTP model. For the current analysis, these estimates were post-processed to arrive at travel time savings benefits of the build alternatives rel- ative to the no-build alternative. The steps involved in this process include conversion of V/C esti- mates to average speeds along corridor segments using a volume-delay function (VDF)42. The speed improvements for the build alternatives are then converted to equivalent travel time savings. The travel time savings benefits estimated from this process are as follows for the 2030 forecast of the northbound P.M. peak period average travel time savings relative to no-build: • TSM/TDM (0% savings), • HOV lanes (16% savings), • Truck-only lanes (23% savings), and • Managed lanes (29% savings). Reliability Benefits. Reliability benefits are estimated in terms of reduction in incident-related delay for the build alternatives relative to the no-build alternative using IDAS post-processing fac- tors for delay per VMT as a function of congestion (V/C) and number of lanes. The incident-related delay per VMT estimates from IDAS for each of the segments of the I-15 corridor (based on con- gestion and number of lanes) are presented in Appendix B. Using the percent change in daily VMT estimates (derived from Table 3.20, assuming average trip lengths do not change significantly between alternatives), and the delay per VMT estimates from IDAS, the percent change in incident-related delay for the build alternatives relative to the no-build are calculated and presented in Table 3.21. Safety Benefits. Safety benefit comparisons between alternatives are conducted by estimating total accidents (fatality, injury, and property damage) per million VMT as a function of V/C along the corridor (by segment) using the IDAS software, and estimating percentage change in total acci- dents using the percent change in VMT estimates between alternatives. For the truck-only lane Performance Evaluation 49 No Build TSM/TDM HOV Lanes Truck Lanes Managed Lanes Segment* Autos Trucks Truck % Autos Trucks Truck % Autos Trucks Truck % Autos Trucks Truck % Autos Trucks Truck % 1 103,133 34,749 25% 103,473 34,777 25% 108,985 34,835 24% 110,093 35,237 24% 113,332 34,755 23% 2 98,240 42,047 30% 98,704 42,004 30% 103,340 42,167 29% 105,404 43,079 29% 108,787 42,013 28% 3 101,608 48,330 32% 100,414 48,268 32% 102,100 48,464 32% 105,329 48,664 32% 98,301 48,120 33% 4 142,320 49,575 26% 141,585 49,547 26% 142,922 49,920 26% 147,598 50,161 25% 142,574 49,663 26% 5 96,515 39,559 29% 96,333 39,532 29% 100,713 39,885 28% 104,212 40,316 28% 99,034 39,681 29% 6 163,796 27,704 14% 163,500 27,521 14% 175,326 30,172 15% 173,552 32,319 16% 173,871 27,717 14% 7 223,860 42,250 16% 224,995 42,422 16% 245,335 46,071 16% 244,013 56,828 19% 234,762 42,477 15% Source: Adapted from Southern California Association of Governments (SCAG), I-15 Comprehensive Corridor Study, December 20, 2005, as found at http://www.scag.ca.gov/goodsmove/pdf/I-15_Comprehensive_Corridor_Study.pdf.[cf29] Note: * Segments are defined as Segment 1—Mojave River Crossing to Bear Valley Road; Segment 2—Bear Valley Road to US 395; Segment 3—U.S. 395 to SR 138; Segment 4—SR 138 to I-215; Segment 5—I-215 to I-210; Segment 6—I-210 to I-10; and Segment 7—I-10 to SR 60. Table 3.20. Daily auto and truck volumes along I-15 by segment and alternative, 2030. 42For the current analysis, the following VDF was used: Speed = 65/(1 + [1.16 * (V/C)4.33]).

alternative, an additional 15% reduction factor is applied to the total accidents estimate to account for the safety benefits of truck-auto separation. Table 3.22 presents these results. Assumptions and Data Gaps. Some of the key assumptions and data gaps in this study that potentially impact the ability to draw conclusions about the performance benefits of truck-only lanes are as follow: • The I-15 study does not consider a conventional mixed-flow lane alternative as part of the alter- natives analysis process. The alternative that comes closest to a conventional mixed-flow lane alternative is the managed lanes alternative. However, differences in operational characteristics of conventional mixed-flow and managed lanes imply that even with similar capacities, the per- formance characteristics of these alternatives are expected to be quite different. Thus, the results from the study are inconclusive in providing insights into the relative performance benefits of truck-only lanes compared to additional mixed-flow lanes. • The study, as with other studies described earlier, uses a travel demand model to evaluate the mobility performance of truck-only lanes, and based on the assumptions related to PCE factors, this process could potentially underestimate the travel time savings benefits of truck-only lanes. • Since IDAS factors only account for the reliability benefits of capacity improvement but do not consider the benefits of truck-auto separation, the actual reliability benefits of truck-only lanes are expected to be higher than estimated in Table 3.21. 50 Separation of Vehicles—CMV-Only Lanes Segment TSM/TDM (%) HOV Lanes (%) Truck-Only Lanes (%) Managed Lanes (%) 1 0 -86 -88 -86 2 0 5 -88 -85 3 -1 -79 -82 -97 4 -1 -66 -70 -91 5 0 -86 -88 -87 6 0 -78 -81 -78 7 1 -70 -74 -71 Average 0 -66 -82 -85 Table 3.21. Percent change in incident-related delay relative to no-build (P.M. peak period, northbound), 2030. Truck Lanes Segment TSM/TDM (%) HOV Lanes (%) From IDAS (%) After 15% Reduction Factor Adjustment (%) Managed Lanes (%) 1 0 -8 -21 -33 -4 2 0 -8 -20 -32 -3 3 -1 0 -12 -25 -15 4 -1 0 -12 -25 0 5 0 -9 -20 -32 -10 6 0 7 -10 -24 6 7 1 10 -7 -21 -19 Average 0 -1 -15 -27 -6 Table 3.22. Percent change in total accidents compared to no-build (P.M. peak period, northbound), 2030.

• The study only looked at the performance of alternatives for the A.M. and P.M. peak periods. Although this would suffice for the other alternatives, analyzing the impact of time-of-day dis- tributions of autos and trucks in the mid-day period on utilization of truck-only lanes and the overall performance of the truck-only lane alternative would be important to assess the applica- bility and effectiveness of truck-only lanes in meeting performance objectives based on corridor auto and truck demand characteristics. 3.4 Comparative Summary of Results and Conclusions This section draws together the results of a number of different studies that all concluded in their own context that truck-only lanes of different configurations provide positive benefits and may be a preferred choice for improvements in both long-haul corridors and urban corridors. Although the particular studies that are drawn on for performance evaluation are by no means exhaustive of all the studies that have been conducted that evaluate truck-only lanes, they are representative of the general configurations that have been evaluated and the methodologies that have been used to conduct these evaluations. Several general conclusions can be obtained from reviewing these per- formance evaluations. These conclusions are presented and discussed first, then followed by a dis- cussion of conclusions that relate to each of the specific corridor scenarios (long-haul and urban corridors) that formed the basis of this chapter. 3.4.1 General Conclusions There are several general conclusions that can be drawn from the performance evaluations. These include the following: • The primary analytical tool used to evaluate truck-only lanes is a traditional travel demand model that incorporates truck trip tables. This limits the analysis to the types of travel behav- iors that are best captured in these models and misses many of the key potential benefits of truck-only lanes. Operational benefits of truck-only lanes are largely absent from the analyses or are estimated on the basis of relationships between operations and V/C ratios. • The benefits of truck-only lanes are highly sensitive to the methods and assumptions regard- ing diversion of trucks to these lanes (or truck lane utilization). In most of the analyses, these methods and assumptions have been driven by the level of congestion in the mixed-flow facil- ity and the data on origin-destination (O-D) patterns of the truck-only lanes. A more market- focused analysis of the drivers of truck lane utilization than what was found in most of the studies is needed to draw conclusions about the feasibility of truck-only lanes. • Assessing the performance benefits of truck-only lanes effectively requires the definition of appropriate alternatives as a standard of comparison. If a primary benefit driving the perfor- mance assessment is response to congestion, then a baseline alternative for comparison pur- poses should include other methods of alleviating congestion (e.g., equivalent capacity in general purpose lanes). • Despite these shortcomings, the performance evaluations do demonstrate that there are corri- dors with high volumes of truck traffic where truck-only lanes could provide benefits to freight users (and thus would achieve high levels of utilization) as well as to nonfreight users who would continue to use mixed-flow facilities. The reliance on travel demand models as the primary analytical tool used to evaluate per- formance of truck-only lanes presents a serious drawback to the analyses. It focuses much of the analysis on benefits that are derived from evaluation of general congestion conditions. The standard approach is to assign multiclass vehicle trip tables to a network with and without truck-only lanes. Routing decisions of the trucks are a function of travel times on the network given the O-D patterns reflected in the truck trip tables. Other benefits (such as productivity, travel time savings for users of the mixed-flow lanes, safety, and reliability) are generally derived Performance Evaluation 51

from relationships between V/C ratios (i.e., general recurrent congestion conditions) and these other benefit categories. The result is that evaluations in which there are high levels of conges- tion in the corridor generally will show benefits for adding capacity as truck-only lanes as long as the access and egress locations to the truck lanes are chosen to serve general O-D patterns of the trucks. However, this approach does not adequately address some of the following pur- ported benefits of truck-only lanes: • Safety benefits. Relationships between lane configurations and V/C ratios were used to estimate safety benefits in the performance evaluations and for the purposes of this project, these were adjusted by a simple factor to account for additional benefits that accrue from truck-auto sepa- ration. This simple factor was derived from the few studies that attempted to draw some data from the limited applications of truck-auto separations such as the New Jersey Turnpike and the limited applications of truck interchange bypasses. A more rigorous analysis of truck-auto inter- actions using simulation models and more detailed statistical analysis of how truck-involved accidents vary with the relative amount of truck traffic and congestion might identify additional safety improvements that are associated with truck-auto separations. Research conducted at the University of Tennessee43 on the safety benefits of truck-only lanes using simulation models has demonstrated that this would be a potentially useful technique to conduct this type of analy- sis. Appendix B (Section B.4) provides a discussion of this approach and the safety benefits results from this research. • Reliability benefits. The shortcomings with respect to the analysis of safety benefits extend to the analysis of reliability benefits as well. None of the literature sources report any attempts to examine how variability in speeds (and travel times) is related to truck-auto interactions whereas at least some efforts have been made to show how crashes might be reduced through truck-auto separations. A more thorough operations analysis capable of analyzing merge and weave behavior, near crashes, and the different rates of recovery from crashes that are truck- involved—versus those that are not—could provide a more definitive assessment of the reli- ability benefits of truck-only lanes. Current Highway Capacity Manual (HCM) approaches to conducting operational analysis are likely to be insufficient for this type of analysis because of limitations on the percentage of trucks in the traffic stream that can be handled by this approach. Again, simulation models appear to be a more appropriate tool. There may be some additional shortcomings associated with using travel demand models to ana- lyze the congestion or travel time savings benefits of truck-only lanes. A standard approach used in modeling the impacts of truck traffic on congestion is to convert truck volumes to passenger car equivalents (PCEs). However, given potential operational improvements associated with truck- auto separations and the variability of truck PCE factors as a function of demand (truck share of total traffic volume) and certain roadway geometric characteristics, current travel demand models may not be properly accounting for changes in PCE factors as trucks and autos are separated. Thus, it is possible that the performance evaluations conducted in the studies reviewed for this project may be underestimating the travel time savings associated with truck-only lanes. The general approach to estimating truck lane utilization in the studies reviewed for this per- formance evaluation was the use of travel demand models where routing decisions are based on minimum time or cost path assignments. This may be an appropriate method for estimating diver- sion of trucks to truck lanes when the primary motivation for using the truck lanes is to avoid con- gestion, but a more market-focused analysis is necessary when considering routing decisions for LCVs in long-haul corridors. The motivation for LCV operation should be based primarily on 52 Separation of Vehicles—CMV-Only Lanes 43A. A. Adelakun, Simulating Truck Lane Management Approaches to Improve Efficiency and Safety of Highways in Knoxville, Tennessee, Master of Science thesis, University of Tennessee, Knoxville, December 2008.

increased productivity and potential net earnings increase to carriers. However, there are a variety of limiting factors that may reduce this benefit because of off-setting costs of operation and lim- ited markets. This issue is discussed in more detail later in this chapter in the discussion of conclu- sions under the long-haul corridor scenario. In light of how important congestion relief benefits appear to be in supporting positive conclu- sions about the value of truck-only lanes, it is disconcerting to find how few of the studies look at the performance tradeoffs between adding truck-only lanes versus adding more general purpose lanes in congested corridors. Inclusion of alternatives that add equivalent capacity for mixed-flow operations is important in determining the congestion relief benefits of truck-only lanes in com- parison with the benefits provided by adding mixed-flow capacity. Nonetheless, the results of the studies evaluated in this project suggest that there are likely to be many congested corridors with sufficient truck volume to ensure high utilization of truck-only lanes. Given that benefits associ- ated with truck-auto separation are not well captured with the current data and tools, these high levels of utilization suggest that truck-only lanes can be a desirable alternative to other forms of capacity expansion. This conclusion needs to be further explored with a more complete analysis of the time-of-day characteristics of congestion and truck traffic patterns. 3.4.2 Conclusions for Long-Haul Corridors The evaluation of truck-only lanes in long-haul corridors focused on three primary perfor- mance measures and two primary scenarios (with and without LCV operations; the “with LCVs” scenario only looks at longer-combination truck configurations operating on truck-only lanes (such as turnpike doubles and triple shorts), and does not consider special permit vehicles, such as oversize/overweight (OS/OW) trucks). As stated earlier in Section 3.2 (Step 3), the conclusions presented here are based on an analysis of performance benefits of truck-only lanes, without con- sideration of costs. These results are intended to provide an understanding of how truck-only lanes perform compared to other alternatives. The planning and policy/decision making processes for truck-only lanes call for the evaluation of the benefits of truck-only lanes (compared to other alternatives) in relation to costs. To address this issue, the next Chapter delves into the B-C per- formance of truck-only lanes compared to other alternatives. The performance measures analyzed in this section were as follow: • Productivity benefits. Measured in a variety of different ways, including reductions in operat- ing costs or increased net earnings to truckers per day, these benefits generally are compared to a no-build scenario. Productivity improvements are the result of reduced travel times for trucks on the truck-only lanes and increased payloads per unit cost for LCVs. • Travel time savings for mixed-flow traffic. By removing trucks from the mixed-flow lanes, travel time savings can be obtained by other motorists. • Safety benefits. Safety benefits for all motorists accrue from reduced congestion and truck-auto separations. The results of the performance evaluation for long-haul corridors are summarized in Table 3.23. The key conclusions from the review of studies on the performance evaluation of truck-only lanes along long-haul corridors are contained in the following discussion. • In selecting the long-haul corridor scenario, it was observed that there was a high level of inter- est in using truck-only lanes as a way of moving to LCV operations, thereby promoting greater freight efficiency in key freight corridors. Although the studies that were examined do show incremental benefits from LCV operations regardless of the method used to measure produc- tivity benefits, these incremental benefits associated with LCV operations are generally small as compared with the potential benefits associated with travel time savings achieved from the Performance Evaluation 53

Table 3.23. Summary comparison of performance evaluation results, long-haul corridors. Productivity Benefits Travel Time Savings Safety Benefits Truck-Only Lanes Truck-Only Lanes Truck-Only Lanes No-Build with LCV Operations (%) Without LCV Operations (%) With LCV Operations (%) Without LCV Operations (%) With LCV Operations (%) Without LCV Operations (%) With LCV Operations (%) Source – – 36a – 21 – 47 I-35 Trade Corridor Study – 106 120b 20 – 44 – Georgia Statewide Truck Lane Needs Identification Study – 55 63c – – – – Reason Foundation, Toll Truckways: Increasing Productivity and Safety in Goods Movement 4d – – - – – – Western Uniformity Scenario Analysis Notes: a. The study provides productivity benefits in terms of savings in truck operating costs, but without information on the share of the total revenues accounted for by operating costs, the % increase in truck earnings could not be determined from the study. b. These results are the relative productivity benefits of truck-only lanes (without and with LCV operations) compared to a no-build alternative. LCV operations provide around 6.8% incremental productivity benefits compared to standard truck operations on truck-only lanes (this is the productivity benefit associated solely with increase in payloads without considering the impacts of travel time savings on productivity improvements). c. LCV operations provide around 5% incremental productivity benefits compared to standard truck operations on truck-only lanes. d. The study reports productivity benefits of LCV operations on mixed-flow lanes (compared to standard truck operations on mixed-flow lanes) in terms of % savings in shipper costs. The information available from the study is inadequate to be able to translate these cost savings into equivalent productivity benefits in terms of increased truck earnings.

additional capacity in the truck lanes. This is because all of the long-haul corridors that were analyzed included large segments moving through congested metropolitan areas. In the long run, this may be a prerequisite for successful application of truck-only lanes even in long-haul corridors—that is, unless there is a substantial need for new freight capacity in the corridor because of congested operations, there may not be sufficient benefit from truck-only lanes to justify the investment. This is difficult to determine from the analyses conducted in the various studies of long-haul corridors and without a more complete B-C evaluation. • Incremental benefits of LCV operations relative to truck-only lanes without LCV operations are in the range of 4% to 7% and are much less than the magnitude of productivity benefits associ- ated with operations over long distances in reduced congestion (i.e., productivity benefits asso- ciated with travel time savings over long-haul distances). A reassessment of the travel time savings portion of productivity benefits in corridors with less congested operations would be beneficial since multistate corridors that are being considered for LCV operations often have long stretches of uncongested operations and truckers may be able to plan their routes to avoid traveling in metropolitan areas during periods of peak congestion. Since the productivity bene- fits are assessed on a per truck basis, it is possible that with high levels of truck usage even small relative productivity benefits may prove cost effective when compared to costs. This will be explored further in the next chapter on benefit-cost evaluations and points out the importance of the level of utilization in assessing the cost-effectiveness of truck-only lanes with or without LCV operations. • All of the studies that considered truck lane utilization (which were most often done with travel demand models) estimated high levels of truck diversion to the truck-only lanes, driven largely by the need to avoid congestion. Truck diversion rates of over 65% were reported in some studies and even higher rates of diversion were estimated in the I-35 study for LCV operations. As noted previously, market limitations suggest that much lower diversion rates for LCVs would be appropriate in less congested long-haul corridors because the markets may not be present to support this level of diversion. It appears that none of the studies considered mar- ket factors in assessing the potential rate of LCV usage on truck-only lanes. • Two of the studies did examine the potential travel time savings for the mixed-flow traffic in long-haul corridors with and without LCV operations and both studies showed travel time sav- ings on the mixed-flow lanes of around 20%. In both cases, there were significant fractions of the total corridor length that were very congested (around Atlanta, Georgia, and in the metro- politan areas along I-35 in Texas) and the additional capacity compared to a no-build condi- tion both encouraged truck use of the truck lanes and freed up capacity on the mixed-flow lanes. However, neither study looked at a comparison with an alternative to build additional mixed-flow capacity. • It should be noted that the large increases in productivity benefits to trucks using the truck-only lanes (due to both travel time savings and increased payloads in the case of LCV operations) sug- gest that it may be possible for the public sector to capture some of the value of the productiv- ity benefits of the truck lanes through tolling while still providing sufficient benefits to trucks so that drivers would continue to prefer the truck lanes. As part of this analysis, it would be impor- tant to assess the impacts of incremental productivity benefits of LCV operations on truck-only lanes on the feasibility of tolls. The Reason Foundation has conducted sensitivity analysis to assess the return on investment (ROI) for toll truckways under various toll, truck diversion rate, and facility cost scenarios. However, a more detailed evaluation of tolling concepts for specific congested long-haul corridors is needed to assess the viability of tolls on truck-only lanes under realistic long-haul corridor demand and operational conditions. • Results from the review of studies indicate safety benefits to be significant for truck-only lanes in long-haul corridors. This can be attributed to the mobility benefits provided by truck-only lanes along those intercity corridor segments experiencing notable congestion (such as along the outer boundaries of the Atlanta metropolitan area). Although the approach to estimating safety Performance Evaluation 55

benefits is not very sophisticated and is driven largely by improvements in overall V/C ratios, the post-processing of results conducted as part of the performance evaluation to some extent takes into account the safety effects of separating trucks and autos. Techniques that are able to focus more specifically on the benefits of truck-auto separation are likely to show even greater safety benefits. This is clearly an area where lack of field data is a critical deficiency. 3.4.3 Conclusions for Urban Corridors The evaluation of truck-only lanes in urban corridors focused on comparisons between addi- tional capacity in truck-only lanes and similar capacity additions in mixed-flow lanes. In order to properly assess the performance of truck-only lanes in comparison to additional mixed-flow lanes along urban corridors, it is necessary to have models that differentiate performance at different times of day (since trucks and autos may have different time-of-day demand characteristics), and the data on truck activity must properly take into account truck time-of-day demand. The per- formance measures that were considered in the performance evaluation of truck-only lanes for urban corridors included the following: • Travel time savings. Travel time savings were noted for autos (and trucks) on the general pur- pose lanes (and for trucks on the truck-only lanes to the extent to which these savings are reported in the reviewed studies). • Reliability. In order to evaluate projects on a consistent basis, reliability benefits are analyzed in terms of savings in nonrecurrent (incident-related) delay, which are estimated using inci- dent delay look-up factors as a function of V/C and number of lanes from IDAS, as described previously. • Safety. The approach used to evaluate safety benefits was the same as that used for long-haul corridors. The results of the performance evaluation for urban corridors are summarized in Table 3.24. The key conclusions from the review of studies on the performance evaluation of truck-only lanes along urban corridors are as follow: • As mentioned in the discussion of the assumptions and data gaps associated with the reviewed literature sources, the differences in capacity assumptions between the truck-only lane and mixed-flow lane alternatives and the omission of the mixed-flow lane alternative in the per- formance evaluation in some of the reviewed studies result in the findings being inconclusive or inadequate in assessing the relative performance benefits of truck-only lanes against mixed- flow lanes with similar capacity. This limitation is addressed in the next chapter as part of the B-C analysis. • Given that the principal tool used to conduct the evaluations was a traditional travel demand model, truck-only lanes in urban settings would only compete favorably with additional mixed- 56 Separation of Vehicles—CMV-Only Lanes Table 3.24. Summary comparison of performance evaluation results, urban corridors. Travel Time Savings Reliability Safety Mixed-Flow Lanes (%) Truck-Only Lanes (%) Mixed-Flow Lanes (%) Truck-Only Lanes (%) Mixed-Flow Lanes (%) Truck-Only Lanes (%) Source 14 16 47 59 26 36 I-710 Major Corridor Study 29 23 85 82 6 27 I-15 Comprehensive Corridor Study – 17 – 39 – 15 Georgia Statewide Truck Lane Needs Identification Study 9 – – – – PSRC FAST corridor study

flow lanes in settings with enough truck traffic throughout the day to achieve high levels of uti- lization during all periods. Further, the corridor would need to have reasonably congested con- ditions during the mid-day periods when truck traffic tends to peak, and the contribution of truck traffic to peak-period congestion also would need to be significant (so that diversion of trucks in the peak-period can contribute to notable relief in congestion). These conclusions point out the necessity of obtaining better data on PCE factors for trucks in truck-only lanes as compared to mixed-flow conditions with varying levels of congestion and percent trucks in the mixed-flow. It is very likely that trucks have higher PCE factors when in mixed-flow conditions even with the same level of traffic; thus the separation of trucks and autos could result in greater congestion relief than is indicated in these studies. • The impacts of truck-only lanes in urban corridors, as estimated in most of the existing studies, do not adequately address potential benefits of truck-auto separation in terms of reliability and safety. These impacts would be even more pronounced when measuring reliability and would be related to the safety benefits of truck-auto separations as well as the general traffic flow pat- terns. As noted previously, when trucks and autos mix, crashes are more likely and these crashes generally are severe. The time to recover from such crashes is greater than that needed for recov- ery from auto-only crashes and results in more significant impacts on reliability. The inability to capture the true reliability benefit of truck-auto separation in the performance evaluations is observed to be a major deficiency of the studies conducted to date. This chapter pointed out some deficiencies in prior studies with regard to the performance eval- uation of truck-only lanes for both long-haul corridors and urban corridors. Chapter 4, which follows, addresses some of these deficiencies by developing representative generic corridors for both the long-haul and urban corridor scenarios, developing an appropriate set of alternatives, monetizing the benefits so that they can be compared on a consistent basis, and evaluating these benefits in comparison to costs to get a better sense of the potential B-C tradeoffs of truck-only lanes against adding mixed-flow lane capacity. Although this addresses some of the deficiencies of prior analyses, there are other issues that can only be addressed through a combination of addi- tional field data collection, focus on specific corridors where actual traffic conditions can be taken into account, and the application of different analytical techniques (such as traffic simulation mod- eling). Some proposals for future research to address remaining deficiencies in the data and analy- sis methodologies to support the performance evaluation and B-C analysis of truck-only lanes are presented in Chapter 5 of this report. Also, the performance evaluations presented in this chapter did not consider the application of tolls on truck-only lanes. To understand the feasibility of tolls and their impacts on the performance of truck-only lanes, a discussion of tolls on truck-only lanes and their implications is presented in Appendix B (Section B.5). Performance Evaluation 57