Identification and Evaluation of the Cost-Effectiveness of Highway Design Features to Reduce Nonrecurrent Congestion (2014)

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

12 Chapter 1 presented and discussed the six primary sources of nonrecurrent congestion. Research in SHRP 2 Project L07 addressed these sources of unreliable travel times by iden- tifying various design treatments that may be considered by highway agencies to reduce nonrecurrent congestion. Initially, the scope of Project L07 focused on five of the six sources of nonrecurrent congestion: • Traffic incidents • Work zones • Traffic control devices • Special events • Demand fluctuations However, during the first year of the project, SHRP 2 expanded the scope of Project L07 to address weather as a cause of nonrecurrent congestion and to include design treatments that may be used to reduce nonrecurrent con- gestion related to snow and ice and other weather-related events. research Objective and Scope The objectives of this research were to (1) identify the full range of possible roadway design features used by transpor- tation agencies on freeways and major arterials to improve travel time reliability and reduce delays from key causes of nonrecurrent congestion, (2) assess their costs and opera- tional and safety effectiveness, and (3) provide recommenda- tions for their use and eventual incorporation into appropriate design guides. The research focused on geometric design treatments to reduce nonrecurrent congestion. However, some of these treat- ments are broader in scope than just geometric design. For example, some include traffic control, incident management, or motorist services. That is, some treatments of interest are directly related to geometric design, but other treat- ments have an important, but indirect, relationship to geo- metric design (e.g., they are supported by geometric design features). Three separate analyses of the design treatments were conducted to achieve the research objectives. The primary analysis was a traffic operational assessment that estimated a distribution of travel times on a freeway segment with a specific set of geometric and operational characteristics and then estimated the expected change in the distribution of travel times after the implementation of a treatment. This shift in the distribution of travel times provides information about delay savings and improved reliability of the roadway as a result of implementing a design treatment. A second- ary analysis of the safety implications of using the design treatments was also conducted. Although this analysis considered direct safety benefits of treatment installation, it focused on the indirect benefits associated with reduced nonrecurrent congestion. The research team explored the relationship between crash frequency by severity and level of service to develop a model for predicting the reduction in crashes due to a reduction in nonrecurrent congestion. These two analyses were then used as inputs, along with a user-defined treatment cost, into the benefit–cost analysis of treatments. The traffic operational analysis methodology developed in this research was intended to build from work completed in SHRP 2 Project L03. However, the products of Project L03 did not precisely meet the needs of the Project L07 analy- sis. The next section describes the evolution of the research approach for this analysis. Chapter 4 of this report provides a detailed description of the traffic operational analysis, Chap- ter 5 describes the safety analysis, and Chapter 6 describes the benefit–cost analysis. C h a p t e r 2 Research Approach

13 evolution of research approach for traffic Operational analysis The research team’s original concept for Project L07 was that delay measures (i.e., vehicle hours of delay) for specific design treatments for nonrecurrent congestion would be obtained through a combination of the following: • Direct calculation of performance measures from field data • Deterministic analysis techniques (primarily those of the Highway Capacity Manual [2]) • Microscopic traffic simulation • Qualitative methods, when necessary During the development of the work plan for Project L07, the research team for another SHRP 2 Reliability project (Proj- ect L03, Analytic Procedures for Determining the Impacts of Reliability Mitigation Strategies) anticipated that their reli- ability models would estimate vehicle hours of delay and then translate those delay estimates into reliability measures. Spe- cifically, it was anticipated that the models being developed by Project L03 could be very useful in translating the Project L07 delay measures into reliability measures, as follows: Treatment → ∆Event or physical or traffic characteristics → ∆Delay → ∆Reliability Therefore, this approach was recommended in the work plan for Project L07. However, the models that were actually developed by Proj- ect L03 and presented in the final report of that project estimate reliability measures directly without first quantifying vehicle hours of delay. Thus, Project L03 took a somewhat different approach to modeling than its team originally anticipated. Furthermore, as the Project L07 research team studied the Proj- ect L03 relationships and began to apply them to specific design treatments, some constraints and boundary conditions of the Project L03 models became apparent. In particular, the Project L03 models are most applicable to urban freeways in major metropolitan areas, but the scope of Project L07 included rural and small- and medium-sized urban areas, as well. In addition, the Project L03 models are most applicable to peak periods, but Project L07 focused on nonrecurrent congestion, which occurs at any time of the day or night. Therefore, the research team revised the approach for Project L07 as follows: • Reliability measures for design treatments were deter- mined using the Project L03 models directly for the condi- tions to which these models apply; this generally included time-slices (i.e., portions of the day) in which the demand- to-capacity (d/c) ratio was greater than or equal to 0.8. • Delay measures for the effect of design treatments were developed for a broader range of traffic conditions than those to which the Project L03 models apply (i.e., includ- ing traffic conditions representative of off-peak condi- tions in major urban areas, peak and off-peak conditions in small- to medium-sized urban areas, and peak and off- peak conditions for rural areas). These conditions gener- ally include time-slices in which the d/c ratio is less than 0.8. The delay measures were developed with simulation modeling for each design treatment to which simulation modeling was applicable. Thus, the operational effects of the design treatments were initially quantified with a combination of reliability measures from the Project L03 models and delay measures from simu- lation modeling. Ideally, however, the L07 research team and SHRP 2 hoped that reliability models could be developed for the full range of d/c ratios; that is, for congested and uncongested periods. Furthermore, the Project L03 models included a variable to account for rainfall (R0.05″), but the models did not account for snow conditions. To address these and other issues, SHRP 2 approved an extension of Project L07 to further develop and refine the analytical framework and the spreadsheet-based Analysis Tool that were developed earlier in the project. Spe- cifically, the extension of the project focused on the following: • Further development of the models to address the effects of snow and ice on the traffic operational effectiveness of design treatments • Further development of the models to address the effects of multihour incidents on the traffic operational effective- ness of design treatments • Analysis of existing data to improve the applicability of reliability models for time periods with d/c < 0.8 • Verification of the reasonableness of evaluation results for design treatments obtained with the spreadsheet-based Analysis Tool products of the research In addition to the documentation of the research found in this final report, the research plan included the development of two key products: a design guide for nonrecurrent con- gestion treatments, and an information dissemination plan. Through the course of conducting the traffic operational analysis and applying reliability models to assess the traffic operational effectiveness of design treatments, the research team also developed a spreadsheet-based treatment analy- sis tool. This analysis tool, which is accompanied by a user guide, represents a third key product in the research.

14 Design Guide The Design Guide catalogs the design treatments considered in this research, providing planners, designers, operations engi- neers, and decision makers with a toolbox of possible options for addressing nonrecurrent congestion through design treat- ments. The Guide begins with an introduction to nonrecurrent congestion and reliability, a discussion of the six main causes of nonrecurrent congestion, and a basic explanation of how the reduction of delay and the improvement of reliability can be valued in economic terms. Next, the design treatments that were considered in this research are presented with a decision tree that assists the user in narrowing the full list of design treatments to a shorter list that may be appropriate for further consideration and evaluation. Following the decision tree, the design treatments are cataloged, and relevant information is provided in the following categories: • Treatment description and objectives • Typical applications • Design criteria • How treatment reduces nonrecurrent congestion • Factors affecting treatment effectiveness • Factors affecting treatment cost • References The Design Guide’s final chapter includes examples of exist- ing implementations of many of the design treatments. These examples are brief and include information available from inter- net searches, interviews with agency staff, and the research team’s own experience through field visits to various treatment instal- lations. The intent of this chapter is to provide users with infor- mation about the cost, successes, and challenges experienced by agencies who have implemented a treatment in the past and to provide a starting point from which the user can seek additional information from the agency that implemented the treatment. The Design Guide, in its entirety, is meant to serve as a pri- mary reference for planners, designers, operations engineers, and decision makers interested in reducing nonrecurrent con- gestion and improving reliability on their freeways. The docu- ment does not have to be read completely; its main function is to serve as a catalog of nonrecurrent congestion treatments that the user can browse to find information about specific treatments of interest. It is anticipated that the Guide will help users identify a few treatments that may be applicable to a spe- cific roadway of interest and to investigate further by using the Analysis Tool discussed in the next section. analysis tool The Analysis Tool was developed to allow highway agencies to analyze and compare the effects of a range of design strategies on a given highway segment using the analytical procedures developed in this research. Analysts can input data about the highway (e.g., geometrics, volumes, crash totals), and the tool computes delay and reliability indicators resulting from vari- ous design treatments, further translating those results into life- cycle costs and benefits. The tool, shown in Figure 2.1, is a VBA interface overlaying a Microsoft-based Excel 2007 spreadsheet. The tool is designed to analyze a generally homogenous seg- ment of a freeway (typically between successive interchanges). Based on user-input data, the tool calculates base reliability conditions. The user can then analyze the effectiveness of a variety of treatments by providing fairly simple input data regarding the treatment effects and cost parameters. As out- puts, the tool predicts cumulative travel time index (TTI) curves for each hour of the day, from which other reliability variables are computed and displayed. The tool also calculates cost-effectiveness by assigning monetary values to delay and reliability improvements and comparing these benefits to the expected costs over the life of each treatment. The tool is inter- active, in that results are immediately updated and displayed as inputs are changed. The tool is designed to be used in conjunction with two companion documents: this report and the Project L07 Design Guide. It is also supported by an annotated user guide. The tool is the first of its kind, and reliability analysis is still in its infancy. Although this tool (and its successors) will become more sophisticated in the future, it is nevertheless a comprehensive approach to applying the principles developed in Project L07. The tool interface is divided into three parts, as shown in Figure 2.1: • Site inputs. The user enters data regarding location (e.g., segment name, length), geometry (e.g., number of lanes, lane widths, grade), demand (hourly demand, peak hour factors, and truck percentages for a typical 24-h day), spe- cial event information (hourly volume percentage increase and event frequency for up to nine events), work-zone information (work-zone feature and days active for up to nine work zones), precipitation data, and incidents (annual crash and incident totals by severity and type). • Treatment data and calculations. The user enters specific data regarding each selected treatment’s effects, including percentage of incidents reduced by type and the effects of the treatment on average incident duration. The user also enters treatment construction and annual maintenance costs. The tool calculates and displays the treatment’s ben- efits (operational and safety), and displays net present ben- efit and benefit–cost ratio as measures of cost-effectiveness. • Results. For each hour of the day, the tool graphs the five reliability variables that are inputs to the TTI prediction models (see Chapter 4), the treated and untreated cumu- lative TTI curves for each hour, and a series of reliability measures of effectiveness.

15 3. Deliver effective training for prospective implementers on how to use the products. 4. Offer a strategy for what target audiences should do with the information. The strategic dissemination of Project L07’s research results requires outreach to multiple stakeholder groups, with careful consideration of each group’s values and needs. The Dissemination Plan addresses the following: • Types of organizations that need to receive information about the research results • Types of individuals within those organizations who are the target audience for information dissemination • Types of media and materials that should be used to reach those individuals • Methods for managing and monitoring the success of the information dissemination effort The Dissemination Plan also accounts for the overarching activities of the SHRP 2 marketing program. The effective- ness of these activities is, however, dependent on their con- current implementation with the overall marketing efforts of SHRP 2. Dissemination plan From the initial development of the Project L07 scope of work, it was determined that a successful dissemination plan needed to be developed. Such a plan would provide a strate- gic approach to disseminating the results of the research. The objectives of the Dissemination Plan were the following: • Increase awareness of Project L07’s research findings, including the benefits and value of the Design Guide and Analysis Tool within the transportation community. • Spur the adoption and integration of the Design Guide and Analysis Tool into policies and standard practice within the transportation community. A dissemination plan has been developed and submitted to SHRP 2. The plan includes a four-pronged approach to disseminating the research results: 1. Provide clear and distinct messages outlining what the products are and how they add value to the target audience. 2. Engage partnerships to help reach a broader audience and add credibility to the research recommendations and products. Figure 2.1. Project L07 spreadsheet Analysis Tool: user interface.