Trade-Off Considerations in Highway Geometric Design (2011)

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.

8 CHAPTER TWO LITERATURE REVIEW This chapter presents the information gathered through the synthesis literature search. It starts with a focus on the con- ventional process of evaluating trade-offs using the Green Book, explores the concept of nominal and substantive safety, how the design exception process affects the analysis of trade-offs and examines the impact of flexibility in high- way design. Next, the concepts of CSS/CSD are explored, with discussions centered on the background of CSS/CSD, needed data and inputs to the CSS/CSD process, how the basic design controls are treated in CSS/CSD, the process for alternatives analysis, project documentation and decision processes, and identification of case studies. The following section discusses performance-based design and VE; the VE section provides a background on VE, expands on the rela- tion of VE to CSS/CSD, and outlines the VE process. Next, the process of choosing by advantages is outlined, followed by a section dealing with risk analysis and management. The section on risk examines the basic concept of risk and risk management, project management and risk and reliability analysis methods for dealing with risk, and the psychology of risk perception. The final section examines the influence of safety in the acceptance of trade-offs, with discussions focused on lessons from the field of organizational acci- dents, safety-conscious planning, and road safety audits, with overviews of the IHSDM, HSM, and the software tool SafetyAnalyst. CONVENTIONAL APPROACH The Green Book Design criteria, established through years of practice and research, form the basis on which highway designers strive to balance cost, safety, mobility, social and environmental impacts, and the needs of a wide variety of roadway users. The national policy for geometric design is presented in the AASHTO’s A Policy on Geometric Design of Highways and Streets (or Green Book) (1). It is important to note that the Green Book criteria are not all based on robust scien- tific based safety analysis. However, use of the Green Book design standards does promote design consistency, which has an impact on safety. Uniform application of these stan- dards creates roadways that conform to the driver’s expecta- tions by providing positive guidance through a variety of visual cues. At the time of this synthesis, the current edition of the Green Book is the 2004 edition, and it forms the basis for most of the design criteria utilized by STAs. The Green Book offers design policies and guidelines, not standards. For each design element, AASHTO typically provides a range of acceptable values, from the absolute minimum value to a more desirable target value. For an AASHTO guideline to become a standard, it must be adopted by an STA. Most states have adopted standards toward the middle or upper end of the AASHTO ranges (6). AASHTO, formed in 1914 as the American Association of State Highway Officials (AASHO) gave STAs a unified voice to improve road quality. Although the fundamental principles of geometric design were discussed in engineer- ing texts as early as 1921, it was not until 1938–1944 that AASHO published seven documents that formally outlined policies on certain aspects of geometric design. These seven documents were bound together as the Policies on Geo- metric Design (7) in 1950. The policies were amended in 1954 with the publication of A Policy on Geometric Design of Rural Highways (or Blue Book) (8), which was updated in 1965 (9). The urban environment was covered in the 1957 A Policy on Arterial Highways in Urban Areas (10) and updated in 1973 with A Policy on Design of Urban Highways and Arterial Streets (11). In addition, AASHTO developed Geometric Design Standards for Highways Other Than Freeways (12) in 1969. All of these publications were collected together with the first publication of the Green Book in 1984 (13), which was subsequently updated in 1990 (14), 1994 (15), 2001 (16), and 2004 (1). The design criteria created by AASHO and AASHTO have served as the basis for nearly a century of highway geometric design practice. In the foreword, the Green Book outlines the intent of the guidance. The intent of this policy is to provide guidance to the designer by referencing a recommended range of values for critical dimensions. It is not intended to be a detailed design manual that could supercede the need for the application of sound principles by the knowledgeable design professional. Sufficient flexibility is permitted to encourage independent designs tailored to particular situations. Minimum values are either given or implied by the lower value in a given range of values. The larger values within the ranges will normally be used where the social, economic and environmental impacts are not critical (1).

9 Updates to the Green Book have attempted to provide guidance regarding the necessity to consider the needs of nonhighway users and the environment as well during ben- efit-cost analysis. The authors acknowledge that this adds to the complexity of the analysis, but emphasize that this broader approach will allow for both the need for a given project and the relative priorities among various projects to be taken into account. In addition, this broader approach allows the goal of cost-effective design not merely to give priority to the most beneficial individual projects, but also to provide the most benefits to the highway system of which each is a part—a system solution. Trade-off Considerations The broad overview of trade-off considerations contained in the Green Book points out that the guidance is intended to pro- vide operational efficiency, comfort, safety, and convenience for the motorist, while taking into consideration environmen- tal quality. Further, the effects of the various environmental impacts can be mitigated by thoughtful design processes. This principle, coupled with that of aesthetic consistency with the surrounding terrain and urban setting, is intended to produce highways that are safe and efficient for users, acceptable to nonusers, and in harmony with the environment. The empha- sis is on obtaining a balance between all geometric elements, as far as economically practical, to provide safe, continuous operation at a speed likely to be observed under the normal conditions for a given roadway for a vast majority of motor- ists. However, its design guidelines may be overly conserva- tive for some conditions, often based on dated studies from a time when tires, braking systems, pavements, and vehicle dimensions were less forgiving than today (6). This is con- founded by the aging of the driver population with corre- spondingly reduced capabilities. The Green Book follows the philosophy that above-minimum design values be used where feasible. Further, there is a belief that a linear relationship exists between increased magnitude of these minimum design values and the project “quality” or benefits (17). However, in view of the numerous constraints often encountered, practical values need to be recognized and used as needed (18). The guidance speaks to the evaluation of trade-offs in con- ceptual terms and does not provide specific cause-and-effect examples. Further, as the guidance has developed, much of the background material regarding how standards have been developed has been removed, making it more difficult for practitioners to establish these relationships. Chapter three of the Green Book, which deals with elements of design, will be taken as an example to illustrate how the potential impact of trade-offs associated with the common elements of design of sight distance, superelevation, horizontal alignment, and vertical alignment are presented in the guidance. • Sight Distance. With regard to sight distance, it is important that the designer provide sight distance of sufficient length that drivers can control the operation of their vehicles to avoid striking an unexpected object in the traveled way. It points out that although greater lengths of visible roadway are desirable, the sight dis- tance at every point along a roadway should be at least that needed for a below-average driver or vehicle to stop. This establishes the minimum criteria that must be met with regards to sight distance, but does not pro- vide information on potential trade-offs. • Superelevation. The guidance points out that the design of roadway curves need to be based on an appro- priate relationship between design speed and curvature and on their joint relationships with superelevation and side friction. Again, no guidance is given on how to evaluate this appropriate relationship. • Horizontal Alignment. As with most sections, the guidance is focused on the impact of design deci- sions on safety. It points out that adjustments to the highway cross-section or alignment may be necessary in situations in which there are sight obstructions on the inside of curves or the inside of the median lane on divided highways. The guidance does identify a generalized approach to evaluating the trade-offs associated with providing horizontal sight distance on a curve. It acknowledges that because of the many variables in alignment, in cross section, and in the number, type, and location of potential obstructions, specific study is usually needed for each individual curve. The guidance further lists several general con- trols that need to be followed for horizontal alignment, but does not provide specific guidance regarding the importance of each. • Vertical Alignment. As with horizontal alignment, the guidance points out that the major control for safe operation on crest vertical curves is the provision of ample sight distances for the design speed. It explains that although research has shown that vertical curves with limited sight distance do not necessarily experi- ence safety problems, it is recommended that all verti- cal curves be designed to provide at least the stopping sight distance listed in the guidance. The guidance lists several general controls that could be followed for hori- zontal alignment, but does not provide specific guid- ance regarding the importance of each. • Combination of Horizontal and Vertical Alignment. The guidance lists general controls for combinations of horizontal and vertical alignment, but provides no guidance on how to evaluate trade-offs that may arise regarding them. An example of more specific recommendations for poten- tial deviation from desirable criteria is associated with lane widths. Although the guidance does not specifically address the evaluation of trade-offs, it does provide guidance under which conditions different design values can be used. The guidance states,

10 Although lane widths of 3.6 m [12 ft.] are desirable on both rural and urban facilities, there are circumstances where lanes less than 3.6 m [12 ft.] wide should be used. In urban areas where pedestrian crossings, right-of-way or existing development become stringent controls, the use of 3.3 m [11 ft.] lanes is acceptable. Lanes 3.0 m [10 ft.] wide are acceptable on low-speed facilities, and lanes of 2.7 m [9 ft.] wide are appropriate on low-volume roads in rural and residential areas. In some instances, on multilane facilities in urban areas, narrower inside lanes may be utilized to permit wider outside lanes for bicycle use. In this situation, 3.0 m to 3.3 m [10 ft. to 11 ft.] lanes are common on inside lanes with 3.6 m to 3.9 m [12 ft. to 13 ft.] lanes utilized on outside lanes (1). Trade-off on Criteria To evaluate the trade-offs associated with a design, the designer must understand the basic controls and criteria associated with each element of the design. Although the Green Book provides little guidance on evaluating these trade-offs, it does establish the framework from which most controls and criteria are derived. The following critical design controls affect the flexibility of the criteria and, thus, the ability to undertake trade-offs: • Highway functional classification • Design speed • Acceptable operational level of service of the facility • Physical characteristics of the design vehicle • Performance of the design vehicle • Capabilities of the typical driver • Existing and design traffic demand (18). During the past two decades, there has been considerable research on all aspects of geometric design affecting how roadways are designed, how they operate, and ultimately, their safety. A limitation to the potential application of this research is the sheer volume of new information; the follow- ing new publications became available in this period (19): • Older Driver Highway Design Handbook (FHWA 1995) • Highway Capacity Manual (TRB 2010) • Guide for the Development of Bicycle Facilities (AASHTO 1999) • Traffic Safety Toolbox: A Primer on Traffic Safety (ITE 1999) • Access Management Manual (TRB 2003) • Access Management Guidelines for Activity Centers (TRB 1992) • Impacts of Access Management Techniques (TRB 1999) • Driveway and Street Intersection Spacing (TRB 1996) • HOV Systems Manual (TRB 1998) • Design and Safety of Pedestrian Facilities (ITE 1998) • Building a True Community (U.S. Access Board 2001) • Interactive Highway Safety Design Model (FHWA 2010) • Highway Safety Manual (AASHTO 2010) • Guide for Achieving Flexibility in Highway Design (AASHTO 2004) • Flexibility in Highway Design (FHWA 1997) • Designing Walkable Urban Thoroughfares (ITE 2010). Additional new research into the basic building blocks of the Green Book such as stopping sight distance, super- elevation, and, most important, design speed may also sig- nificantly affect design criteria. Further, within the CSD community there is strong concern that the basic concept of functional roadway classifications has limitations (19). Nominal and Substantive Safety Many of the criteria called for by the Green Book are based on the concept of providing adequate safety through the use of design controls and criteria. To adequately evaluate trade- offs associated with deviations from these design standards requires an understanding of how safety is affected by a par- ticular control or criterion. However, the guidance presents safety as an absolute, not a continuum. To understand this concept, we will examine the concepts of nominal safety and substantive safety introduced by Hauer (20). Nominal safety refers to a design or alternative’s adherence to design control or criteria. A design that meets a design cri- terion is said to be nominally safe, whereas one that does not is nominally unsafe. It is important to note that these criteria are also based on the concept of providing a design that meets the needs of “most” drivers—that allows for most drivers to operate both legally and safely and is consistent with accepted design practices. Substantive safety, in contrast, refers to the actual performance of a highway or facility as measured by its crash experience (e.g., number of crashes per mile per year, consequences of those crashes as specified by injuries, fatalities, or property damage, etc.). A road or road segment is then determined to be substantively safe or unsafe based on its actual crash experience as compared with some rela- tive expectation. It is possible to have a road that is nominally safe (i.e., all the geometric features meet design criteria) but substantively unsafe (i.e., there is a known or demonstrated high crash problem). Similarly, not all roads that are nomi- nally unsafe are substantively unsafe. Thus, designs that do not meet the design criteria outlined in the Green Book, while nominally unsafe, may still be substantively safe. Knowledge of the safety effects of design aids design- ers and stakeholders in making reasoned decisions and trade-offs involving safety. Substantive safety is a contin- uum, whereas nominal safety is an absolute (see Figure 1). Incremental difference in a design dimension (e.g., radius of curve, width of road, offset to roadside object) can be expected to produce an incremental, not absolute change in crash frequency or severity. This concept differs from the idea that a nominally unsafe design will automatically result in a substantive safety problem (4).

11 FIGURE 2 Venn diagram relating crash causes. Source: Rumar (21). Design Exceptions Design criteria established through years of practice and research form the basis on which highway designers strive to balance competing needs for a roadway project. For many situations, the design criteria are flexible enough to achieve a balanced design and still meet minimum values. For various reasons, it is not always practical or desirable that a project meet each and every design criterion and standard. On occa- sion, designers encounter situations with especially difficult site constraints, and an appropriate solution may suggest the use of design values or dimensions outside the normal range established by a control or criterion. In such cases, a design exception may be considered. Before the final deci- sion is made to accept a design exception, design alternatives and their associated trade-offs are evaluated through a delib- erative process. Examining multiple alternatives provides a way to understand and evaluate the trade-offs. From the standpoint of risk management and minimizing tort liability, evaluating multiple alternatives demonstrates the complex, discretionary choices involved in highway design. For projects on NHS routes, FHWA requires that all exceptions from accepted guidelines and policies be justi- fied and documented in some manner, and requires formal approval for 13 specific controlling criteria. The process of justification and documentation is not required, but states that are exempted from FHWA oversight on non-NHS proj- ects can follow it as well. The FHWA Federal-Aid Policy Guide identifies the following 13 controlling criteria as requiring formal design exceptions: • Design speed • Lane width FIGURE 1 Comparison of nominal and substantive safety. Source: NCHRP Report 480: A Guide to Best Practices for Achieving Context-Sensitive Solutions (4). Much research has been performed over the past 30 years to uncover the relationship of substantive safety to design. Many of the models are relatively new, and few agencies have well- established procedures for exercising these models to under- stand the safety impacts of varying design criteria. It is an unintended consequence that many working-level staff believe that standards equal safety and that no compromises can be accepted. This view holds even with design values that clearly are not related to design exceptions or to substantive safety (21). The basis for many design criteria has been removed from the discussion presented in the Green Book. As such, practitio- ners must perform research to determine which design values directly support substantive safety and which do not. Not all crashes are the result of a geometric design or roadway issue. A study by Rumar (21) compared causes of crashes in both the United States and Great Britain with respect to the roadway, driver, and vehicle (see Figure 2). The results of the study show that only 3% of crashes are the result of the roadway environment alone, whereas 57% are related only to drivers and 2% only to vehicles. When interactions (roadway–driver, roadway–vehicle, and road- way–vehicle–driver) are accounted for, 34% of crashes are associated with road-related elements. However, 94% of crashes are associated with driver-related elements. To empower design staff to evaluate the trade-offs asso- ciated with being flexible within design criteria, staff need to become more knowledgeable in not just the criteria, but the reasons for them (which may include safety, operations, maintenance, constructability, and other issues) and will need to address the rigidity in current criteria and design manuals. With respect to the former point, FHWA has devel- oped a course to educate design staff on the functional basis of critical design criteria to enable informed decisions when applying engineering judgment and flexibility (see Geo- metric Design: Applying Flexibility and Risk Management, National Highway Institute, FHWA-NHI-380095).

12 However, there is no formal priority when examining the trade-offs associated with respect to nominal safety of each of the 13 controlling criteria. There is no clear, quantifiable means of determining which controlling criteria are most important and how crash frequency varies with variation in each controlling criterion. A study conducted by Stamatiadis et al. (24) examined crash exposure related to design exceptions filed in Kentucky from 1993 to 1998. Design exceptions were approved on 319 projects during the 8-year study period, an average of 40 per year. The majority of the projects associated with these design exceptions were bridge replacements (57%), roadway widening reconstruction (13%), and construction of a turn- ing lane (9%). A total of 562 design exceptions were filed (an average of 1.8 per project), with the most common associ- ated with design speed (34%), sight distance (12%), curve radius (12%), and shoulder width (11%). A crash analysis was performed to determine if there were any consequences associated with the design exceptions. This analysis showed that, with few exceptions, the use of design exceptions did not have a negative effect on highway safety. Flexibility in Highway Design The Green Book provides sound guidelines for many aspects of road design and construction. However, because the Green Book is a universally accepted roadway design guide, many of the guidelines it contains have come to be seen as rigid standards and its inherent flexibility has been neglected. To help overcome some of the limitations of the conventional approach to highway design as presented in the Green Book, FHWA produced Flexibility in Highway Design (18) in 1997 and AASHTO produced the Guide for Achieving Flexibil- ity in Highway Design (25) in 2004. These guides do not attempt to create new standards; rather, they build on the flexibility in the current standards to identify opportuni- ties to use flexible design as a tool to help sustain important community interests without compromising safety. This approach stresses the need to identify which sections of the standards are flexible and to understand the impacts of each standard on the effectiveness of the overall design. Further, the designer must become aware of local concerns of inter- ested organizations and citizens. The designer must then balance the need to provide a safe, efficient transportation system that is able to conserve and even enhance environ- mental, scenic, historic, and community resources. Flexibility in Highway Design was developed in conjunc- tion with several other agencies and interest groups, including AASHTO, Bicycle Federation of America, National Trust for Historic Preservation, and Scenic America. The guidance was written for a target audience of highway engineers and project managers who want to learn more about the flexibility avail- • Shoulder width • Bridge width • Structural capacity • Horizontal alignment • Vertical alignment • Grade • Stopping sight distance • Cross slope • Superelevation • Vertical clearance • Horizontal clearance (22). Individual STAs have established other controlling criteria that may also apply to decisions regarding design exceptions. Further, STAs can use design exceptions on non-NHS routes. This process establishes a clear understanding of the potential negative impacts of the decision to deviate from a design control or criteria. If the decision is made to go for- ward with a design exception, measures to mitigate or reduce the potential negative impacts need to be identified. Poten- tial mitigation strategies associated with each type of design exception are listed in the FHWA Mitigation Strategies for Design Exceptions (23). A design exception is a documented decision to design a highway element or segment of highway to design criteria that do not meet minimum values or ranges established for that highway or project. As discussed in the previous sec- tion, by definition, a design exception is the acceptance of a design that does not meet nominal safety for that criterion. Documentation of design exceptions is important to verify that sound engineering judgment and social/cultural impacts have been considered and that the proposed solution dem- onstrates an appropriate balance of these components. Fur- ther, a written design exception that illustrates that a flexible solution using sound engineering practices has protected the public’s interest as a whole, avoids undue risk in liability. A typical design exception request includes the following: • Description of existing highway conditions and pro- posed improvement project; • Thorough description of the substandard feature(s), providing specific data identifying the degree of deficiency; • Crash data for at least the latest 3-year period, indicat- ing frequency, rate, and severity of crashes; • Costs and adverse impacts that would result from meeting current design standards; • Safety enhancements that will be made by the project to mitigate the effects of the nonstandard feature; and • Discussion of the compatibility of the proposed improvement with adjacent roadway segments (4).

13 CSD is defined as the project development process (including geometric design) that attempts to address safety and efficiency while being responsive to or consistent with the road’s natural and human environment. It addresses the need for a more systematic and all-encompassing approach in project development that recognizes the interdependency of all stages and views them along a continuum. To achieve such balance, trade-offs among several factors are needed and are routinely made in most projects (27). The conventional approach to design does not empha- size an interdisciplinary approach, whereas CSS/CSD approaches do. Over the years, the conventional process has become more structured and formalized and has attempted to include the need to balance trade-offs between strictly transportation requirements and other goals. As the design process evolves, consideration given to issues that do not center on design criteria becomes increasingly impor- tant to determining the ultimate success of a design. This increases the need to identify trade-offs associated with design decisions accurately and completely and strike a balance between the competing factors in an interrelated decision-making process. Key steps in the CSS/CSD transportation decision-mak- ing process are as follows: • Comprehensive, upfront identification of context • Linked decision-making process to ensure that final solutions address the problems identified up front • Stakeholder involvement early and often • Use of multidisciplinary teams • Comprehensive documentation to ensure that commit- ments are communicated and implemented throughout the process (28) Figure 3 flowcharts North Carolina’s CSS/CSD decision- making process. FIGURE 3 North Carolina’s decision-making framework. Source: D’Ignazio and Hunkins (29). CSS/CSD focuses on identifying design problems in func- tional or performance terms and arriving at solutions that address them, instead of rote application of design standards. To develop these designs, the designer makes a series of trade- offs while balancing competing interests such as operational able to them when designing roads. It focuses on the flexibility already available within adopted state standards, often based on the Green Book, that allows designers to tailor their designs to the particular situations encountered in each highway project. It states that these standards often provide enough flexibility to achieve a design that both meets the objectives of the project and is sensitive to the surrounding environment. However, it points out that it is sometimes necessary for designers to look beyond the “givens” of a highway project and consider other options. This could be accomplished by utilizing the design exception process described previously, reevaluating planning decisions, or rethinking the appropriate design. The guidance points out that during the project planning, development, and design phases, designers and communities can work together to have the greatest impact on the final design features of the proj- ect. To facilitate understanding, the problems associated with the project are usually grouped into one or more of the follow- ing four categories: physical condition, operations, safety, and access. The guidance also notes that the flexibility available for highway design during the detailed design phase is limited by the decisions made at earlier stages of planning and project development, demonstrating the need to integrate the commu- nity into the process early in the project (18). The Guide for Achieving Flexibility in Highway Design was developed to aid the designer in better understanding the reasons behind design processes, design values, and design procedures commonly used. It emphasizes that flex- ible design does not represent a fundamentally new process; instead, it focuses on identifying ways to think flexibly and identify the many choices and options available. The guid- ance outlines processes, tools, and techniques to understand and incorporate community interests into project develop- ment and to select appropriate design criteria to support flexible design. It provides an overview of key geometric ele- ments, the models and assumptions used to develop design criteria, and a brief summary of the current knowledge regarding operational and safety effects of design. Finally, the guidance summarizes how designers can achieve flex- ibility while minimizing risk through open development and evaluation of multiple alternatives, assessment of trade-offs, and documentation of the final decisions (25). CONTEXT-SENSITVE SOLUTIONS/CONTEXT-SENSITIVE DESIGN FHWA defines CSS as a collaborative, interdisciplinary approach that involves all stakeholders to develop a trans- portation facility that fits its physical setting and preserves scenic, aesthetic, historic, and environmental resources while maintaining safety and mobility. CSS considers the total context within which a transportation improvement project will exist. CSS principles include early, continuous, and meaningful involvement of the public and all stakehold- ers throughout the project development process (26).

14 efficiency, safety, cost, serving multiple users, and achieving environmental sensitivity. CSS/CSD principles and practices provide a framework for designers to document the rationale for adjustments to guidelines or criteria to best satisfy the needs of the working environment. However, this creates a tension between the design consistency associated with the traditional Green Book approach and the flexible approach to roadway design associated with CSS/CSD. Because successful projects require a level of compromise and trade-off, CSS/CSD are excellent processes for provid- ing structure. In the end, key decisions will be documented, absolutely necessary design requirements will be met, appro- priate flexibility in design will be implemented, and guide- lines that can be applied for the betterment of other factors will be identified in a reasonable, defensible manner (4). An agency may be reluctant to deviate from standards in the face of tort concerns, limited safety research, and project plan modifications. In the Washington State Depart- ment of Transportation’s (WSDOT) early experiences, this was the case even when aesthetic, environmental, surround- ing community, or other benefits were quantitatively shown (30). Because the courts expect that the decisions made and actions taken will be reasonable under the circumstances (4), it is vital to have adequate documentation of the deci- sion-making process. To help implement CSS/CSD, the WSDOT Community Partnership Forum developed a best practices guidebook, Building Projects That Build Communities (31). This docu- ment focuses on effective community-based designs and collaborative decision making. It emphasizes that the key to CSS/CSD is to strive for balance among competing objectives. As such, projects must be supported by sound engineering practices and at the same time incorporate the needs of the jurisdictions involved. On the one hand, there is a need to respect the role of design standards in the development of a project. On the other hand, there is a need to balance application of these standards with other project elements, which may require deviations from the WSDOT guidebook. WSDOT discovered that as more experience is gained in community partnership projects that utilize CSS/CSD prin- ciples, it has become clear that design engineers on these projects must operate with more flexibility than they have in the past. Professional judgment to weigh the trade-offs inherent in urban planning and design is a critical skill. Furthermore, designers must be able to apply a “reasonable- ness” standard to ensure that safety, mobility, and local com- munity goals are met (32). CSS/CSD implementation means that transportation, community, and environmental goals are all on an equal footing. It is possible that transportation goals and tradi- tional engineering approaches may not be the primary driver for all of the final project decisions (32). Idaho’s Context- Sensitive Solutions Guide (33), for instance, identifies four vision principles associated with CSS/CSD (Figure 4) that emphasize aspects of the CSS/CSD approach. FIGURE 4 Context-sensitive solutions vision principles. Source: Context Sensitive Solutions Guide (33). The nature of CSS/CSD design is balancing the desired design elements of the roadway to achieve the most effective design. For example, in situations with constrained ROWs, design elements must be prioritized to ensure that elements that best help the design meet the stated purpose and need of the project are incorporated. Lower priority design ele- ments that do not support the purpose and need may then be adjusted or eliminated. NCHRP Report 642: Quantifying the Benefits of Context Sensitive Solutions (34) documents the findings of recent research work completed to establish a procedure for identi- fying and measuring the benefits of applying CSS/CSD prin- ciples. The study identified benefits that are strongly related to each CSS/CSD principle (see Table 1). The relationship between the benefit and CSS/CSD principle was determined to be fundamental, primary, secondary, or tertiary. The document then lays out a methodology that clearly demon- strates the metrics to be used to measure the benefits from CSS/CSD projects. This methodology can be used to assess the effectiveness of CSS/CSD implementation for a specific project or program, or to develop lessons learned to improve actions and outcomes on future projects. Context-Based Design Context-based design is closely related to CSS/CSD. How- ever, context-based design implies that the street or road is designed to be fully compatible with its context. In contrast, CSS/CSD takes context into account but does not neces- sarily consider it a governing factor in the design. Garrick

15 tation functions (for entertainment, retail, public gath- ering, and recreation). • Step 3: Select the road typology. The four most important factors governing the typology of the road are its physical components, the arrangement of these components, definition of the network, and the desired speed. • Step 4: Determine the design details. These details include both engineering and aesthetic factors that contribute to the proper performance of the roadway, such as the alignment design; the cross-section design; the choice and placement of trees and street furniture; and the relationship of the road to the surrounding buildings, land use, or natural environment (35). Needed Data and Input FHWA describes the Purpose & Need (P&N) statement as the foundation of the decision-making process, influencing the rest of the project development process, including the range of alternatives studied and, ultimately, the alternative selected (36). The generally accepted characteristics of an effective P&N statement are as follows: • The statement is concise, easy to read, and readily understandable. • It focuses on essential needs for the project, which gen- erally relate to physical condition, operational perfor- mance, safety, and access. and Wang (35) outlined basic process for context-based design (Figure 5) to produce good highway design solu- tions. Although this is an oversimplification of a complex design process, it provides a framework for understanding how decisions regarding design controls and criteria affect the trade-offs that are made. This process attempts to tie together both the urban (or place) function and the mobility function of streets and highways and take into account the full context for the design, including multimodal accommo- dation and full integration into the surrounding context. • Step 1: Define the context. Context refers not only to the transportation issues but also to the project’s social, physical, fiscal, ecological, and political background. Understanding the transportation context includes looking at all the modes of travel that exist in the area and understanding how the facility will fit into the full transportation network, not just the highway network. The importance of a network solution rather than a road-by-road solution needs particular emphasis, as this aspect of design has been neglected over the years. It is also the stage that the needs of the local communi- ties must be assessed. • Step 2: Characterize the function. It is important to explicitly consider all the various functions that the given street or road might serve and to design to accommodate these functions. These functions must explicitly include both transportation (for pedestrians, cyclists, private vehicles, and transit) and nontranspor- TABLE 1 CSS/CSD PRINCIPLES AND ASSOCIATED BENEFITS Source: NCHRP Report 642: Quantifying the Benefits of Context Sensitive Solutions (34).

16 Communities (37). This guidance identifies techniques that can be used to achieve a solid understanding of a given proj- ect (Table 2). The techniques range from simple approaches that could be used with projects such as routine maintenance and system preservation to complex approaches for use on more significant projects. Treatment of Basic Design Controls Flexibility in the application of design criteria requires a fundamental understanding of the basis for these criteria (i.e., basic design controls) and the impacts of changing the dimensions of a criteria or adding/eliminating design elements. It is critical that trade-offs associated with these decisions be fully understood to preserve the integrity of the resultant design. AASHTO emphasizes this requirement in A Guide for Achieving Flexibility in Highway Design by stat- ing that “Only by understanding the actual functional basis of the criteria and design values can designers and transpor- tation agencies recognize where, to what extent and under what conditions a design value outside the typical range can be accepted as reasonably safe and appropriate for the site- specific context” (38). Traditional basic design controls for roadway design proj- ects—design speed, functional classification, and context • It delineates other desirable elements (environmental protection, scenic improvements) as separate from the purpose and need. • It is supported by data that justify the need. • It focuses on the problems that need to be addressed and for which the proposed project is being considered and is not written in a way that focuses on the solution or too narrowly constrains the range of alternatives. The issues defined in the P&N document specifically identify what the proposed project is going to address. It is important that the P&N document reflect a full range of public values identified through scoping and public involve- ment, including community issues and constraints, sensitive environmental resources, and appropriate consideration of other factors. The P&N should be based on input from all interested parties, including STA and regulatory agency staff, consultants, and citizens. Any trade-offs associated with a project must meet the needs identified in the P&N, and measures of effectiveness (MOEs) must be developed with this end in mind. A collaborative effort between the Pennsylvania and New Jersey departments of transportation (DOTs) has produced Smart Transportation Guidebook: Planning and Designing Highways and Streets that Support Sustainable and Livable FIGURE 5 Four-step model of highway design process. Source: Garrick and Wang (35).

17 alignment are also by default defined. In addition, the func- tional classification essentially establishes the basic road- way cross section in terms of lane width, shoulder width, type and width of median area, and other major design fea- tures (18). The outcome of this mobility-focused process influences the rest of the design process, from working with stakeholders to the final design. A predetermined outcome can be a source of conflict with stakeholders that delays or even stops projects because the thoroughfare design may not be considered compatible with the surroundings or does not address the critical concerns of the community (40). A European Commission study of road design in nine Euro- pean countries identified similar problems associated with the use of functional classification in most of the countries studied. It notes that the conventional classification in most of these countries is really “roadway classification” and not “functional classification,” suggesting that functional classification should take into account all the functions of the thoroughfare and not only its vehicle-moving (or road) function (41). In addition to functional classification, speed, and con- text, the Green Book presents the following basic design controls and criteria for its recommended design guidance: • Design vehicle • Vehicle performance (acceleration and deceleration) (urban or rural)—are viewed by some as potentially limiting. For example, the flexibility available to a highway designer is considerably limited once a particular functional classifi- cation has been established. Once the functional classifica- tion of a particular roadway has been established, so has the allowable range of design speed and often the required level of service. Although the functional classification system establishes a hierarchy for street networks, it remains silent on the size and scale of the various roadways in each clas- sification by leaving that decision to a capacity-based needs calculation. The result is an emphasis on roadway capacity in transportation decision making, which may conflict with community objectives other than accommodating motor vehicle traffic (39). To address this limitation, Context Sensitive Solutions in Designing Major Urban Thoroughfares for Walkable Communities (40) identifies multiple arterial and collector thoroughfare types. The guidance presents general design parameters under varying contextual conditions for arterials (see Table 3) and collectors. It provides a range of recom- mended dimensions and practices for key design criteria. This increased focus on context provides greater flexibility in the approach to other design variables. With functional class defined, the principal limiting design parameters associated with horizontal and vertical TABLE 2 CSS/CSD PRINCIPLES AND ASSOCIATED BENEFITS Source: NCHRP Report 642: Quantifying the Benefits of Context Sensitive Solutions (34).

18 • Driver performance (age, reaction time, driving task, guidance, etc.) • Traffic characteristics (volume and composition) • Capacity and vehicular level of service • Access control and management • Pedestrians and bicyclists • Safety However, the following four design controls are used dif- ferently in the application of CSS/CSD principles than in the conventional design process: • Design speed • Location • Design vehicle • Functional classification In conventional design practice, design speed has been encouraged to be as high as practical—the designer begins with the highest value and then works down through the range. In CSS/CSD practice, the designer begins by consid- ering contextual factors, resulting in design speed typically taking on a greater range. This requires an understanding of TABLE 3 GENERAL PARAMETERS FOR ARTERIAL THOROUGHFARES Source: Context Sensitive Solutions in Designing Major Urban Thoroughfares for Walkable Communities (40).

19 TABLE 4 COMPARISON OF DESIGN SPEED RANGES BETWEEN THE 1997 AND 2006 DESIGN GUIDES Roadway Type (Based on 1977) 1997 Manual 2006 Guidebook Rural Arterial (Level Terrain) 60 to 75 mph 40 to 60 mph Urban Arterial 30 to 60 mph 25 to 50 mph Rural Collector (Level Terrain) 60 mph 30 to 60 mph Urban Collector 30 mph (minimum) 25 to 40 mph Sources: Project Development and Design Guide (42) and Highway Design Manual (44). WSDOT’s Understanding Flexibility in Transportation Design—Washington (45) outlines the potential impacts that can result when the feature listed is changed in the manner indicated and all other features are held constant (Table 5). This type of guidance is extremely helpful in understanding the trade-offs associated with changes to basic design elements. TABLE 5 POTENTIAL IMPACTS FROM CHANGES IN DESIGN PARAMETERS Feature Change Potential Impacts Design Speed Increase • Shorter travel times (depends on LOS) • Reduced opportunity to view features and services adjacent to roadway • Decrease in safety Decrease • Increased opportunity to view features and services adjacent to roadway • Improved pedestrian/bicyclist environment • Increase in safety Lane Width Increase • Additional room for vehicles to maneuver • Higher operating speeds • Increased impervious surface • Increased capacity • Longer pedestrian crossing distances— greater risk • Can provide room for turning movements at intersections • Can provide room for additional lanes • More room for bicyclists Decrease • Reduced room for vehicles to maneuver • Reduced capacity • Reduced vehicle speeds • Shorter pedestrian crossing distances • Decrease in safety for pedestrians Shoulder Width Increase • Increased space for errant and disabled vehicles • Increased space for bicycles • Increased impervious surface • Increased impervious area to be mitigated • Longer pedestrian crossing distances Decrease • Reduced area for errant or disabled vehicles • Reduced area for bicycles and pedestrians • Reduced impervious area to be mitigated • Shorter pedestrian crossing distances Source: Milton and St. Martin (45). the trade-offs associated with the selected design speed—the functional classification, roadway type and context, and sur- rounding land use characteristics (e.g., predominantly resi- dential or commercial). In urban areas, higher design speed is not a prerequisite for higher capacity, as under interrupted flow conditions intersection operations and delay have a much greater impact on capacity than does design speed. Once selected, design speed then becomes the primary con- trol for determining the following design values: • Minimum intersection sight distance • Minimum sight distance on horizontal and vertical curves • Horizontal and vertical curvature Like design speed, location also takes into consideration the context and surrounding land use characteristics. This includes the level of activity and location of pedestrians, bicyclists, and transit, as well as the types and intensity of surrounding land uses. The design vehicle selected directly influences the selection of design criteria for lane width and curb return radii. Under conventional design, often the designer will select the largest design vehicle (e.g., WB 50 or WB 65) that could use the road- way, regardless of the frequency of use by that vehicle. How- ever, it is not always practical or desirable to choose the largest design vehicle that might use the facility because the effects on pedestrian crossing distances, speed of turning vehicles, and the like may be inconsistent with the community vision and goals. CSS/CSD emphasizes an analytical approach in the selection of design vehicle, including evaluation of the trade- offs involved in selecting one design vehicle over another. Using this approach, after the evaluation of trade-offs, the designer selects the largest vehicle that will use the facility with considerable frequency as a design vehicle. The designer further selects a control vehicle, which is a vehicle that will use the facility infrequently but must be accommodated. It is permissible for the control vehicle to encroach onto opposing lanes or the roadside or be forced to make multipoint turns. When Massachusetts created its new Project Develop- ment and Design Guide (42), it identified a broader range of basic design controls that better respond to the context of Massachusetts communities and to the purpose and need of typical road and bridge projects in the commonwealth in the 21st century (43). These include roadway context, roadway users, transportation demand, MOEs, speed, and sight distance. For example, the new design controls include expanded ranges of design speed, with lower acceptable values for all types of roadways. This approach of utiliz- ing increased and more flexible design controls allows for greater variation in potential solutions and a corresponding greater need for adequate evaluation of the trade-offs asso- ciated with each alternative design. Table 4 compares the guidance in the 1997 and 2006 manuals.

20 – Most important measures needing to be balanced are usually “apples and oranges” and are impossible to collapse to a single common measure. Although disparate measures cannot be directly compared in common terms, simply computing and comparing them represents an improvement. The “apples and oranges” dilemma is not a fault of the process, but more likely an indication that a meaningful set of evaluation measures has been included. • Avoid weighting and scoring schemes – These schemes are likely to be cumbersome and contentious. At this nearly final stage in the planning process, participants’ energy is far better directed toward arriving at a solu- tion that addresses the wide range of project needs and objectives, rather than creating numerical weighting schemes for disparate measures of success that do not lend themselves to such treatment. • Collaborate, do not vote, on a recommended solu- tion – Avoid putting the decision on a recommended solution to a vote, regardless of how representative the study group is of broad community viewpoints. Rather, informed consent or negotiated recommenda- tion could be reached through a collaborative process. At this point, a third-party facilitator, skilled in con- sensus building, may be a valuable input (37). The Massachusetts Department of Transportation’s (MassHighway) new Project Development and Design Guide suggests that alternatives should be developed to comparable levels and presented in an evaluation matrix (for an example matrix, see Figure 6). The evaluation matrix visually presents the alternatives in a manner that facilitates comparison and helps ensure that the impacts of each alternative are consis- tently considered when selecting the best option (42). Alternative Analysis The goal of an alternatives evaluation is to provide an objec- tive and balanced assessment of impacts, trade-offs, and benefits of each alternative. This requires careful selection of, and stakeholder agreement on, MOEs to be used as eval- uation criteria. The MOEs need to reflect community and environmental objectives as well as transportation. Process In Smart Transportation Guidebook: Planning and Design- ing Highways and Streets that Support Sustainable and Livable Communities (37), the Pennsylvania and New Jer- sey DOTs have outlined a process to assess a full range of alternatives. In this process, MOEs are “balanced” against one another to determine the best solution to meet project purpose and need. This process also portrays the trade-offs between measures, such as a reduced traffic level of service balanced against a corresponding increase in civic value associated with on-street parking. The guidance recom- mends that following process steps: • Summarize the assessment – Collapse the assessment to simple and appealing summary products such as charts, tables, matrixes, and spreadsheets. Illustrations (photo- graphs, sketches, or abstract computer graphics) could be used for those measures best described graphically. • Understand important trade-offs – Illustrate the bal- ance (“trade-off”) between important competing mea- sures. One criterion might offset another, such as pairing vehicular traffic service and pedestrian level of service. Successful designs address these trade-offs and achieve a balance of values that can gain community consensus. FIGURE 6 Route 110 & 113 Methuen Rotary Interchange Study Evaluation Criteria Summary Matrix. Source: Route 110 & 113 Methuen Rotary Interchange Study (46).

21 The ITE Context Sensitive Solution in Designing Major Urban Thoroughfares for Walkable Communities (40) out- lines the alternative screening process (Figure 7), show- ing that trade-offs are considered at multiple places in the planning process. This process begins with evaluating fatal flaws and progresses to trade-off evaluation in increasingly greater detail as alternatives are reduced, until finally a pre- ferred concept is selected. WSDOT had developed Understanding Flexibility in Transportation Design—Washington as a companion to WSDOT’s Design Manual. The guidance was created to present information centering on the rationale for decision making and the trade-offs associated with many elements included in transportation projects in CSS/CSD (46). It out- lines how to integrate CSS/CSD principles into all aspects of the project delivery process, and presents case studies of projects within Washington State that illustrate the use of the CSS/CSD approach. The document compiles the issues that are associated with transportation facility design, discusses the trade-off considerations related to each issue, prompts the user to think about how a particular consideration affects other factors related to highway design, and lists resources with each section. It covers several overarching topics, including the following. • Legal Liability Issues – The guidance stresses the importance of full documentation of the options consid- ered, the trade-offs identified, and the rationale behind the decisions made. It provides a historical perspective and outlines current legal responsibility and liability. • Consideration of Facility Users – The guidance points out that the CSS/CSD process will frequently entail making trade-offs in order to provide a safe and func- tional facility for all users. It discusses many of the trade-offs that need to be considered for pedestrians, bicyclists, transit, and motorized vehicles. • Environmental Considerations – The guidance iden- tifies a variety of environmental, scenic, aesthetic, historic, and natural resource values that might be con- sidered and addressed in the planning, design, and envi- ronmental review processes of project development in order to avoid, minimize, or otherwise mitigate project impacts. It covers the environmental issues of urban for- estry; urban streams, fish, wildlife, and plant resources; cultural and historic resources; air quality; noise; vibra- tion; night sky darkness; and, use of recycled materials. • Design Considerations – The guidance provides infor- mation regarding many of the available design alterna- tives for roadways and intersections and discusses the trade-offs associated with each alternative. The guid- ance suggests avoiding the tone of “good” and “bad,” and instead centering on the perceived benefits and draw- backs of features depending on the objective of those interested in the project. It covers facility purpose and characteristics, land use transitions, roadway, roadside, intersections, access, ROW management and utilities, parking, traffic calming, illumination, visual functions, streetscape amenities, and stormwater management. • Community Involvement and Project Development – The guidance presents trade-offs based on the needs FIGURE 7 Project screening process. Source: Context Sensitive Solutions in Designing Major Urban Thoroughfares for Walkable Communities (40).

22 Livable Communities (37), the Pennsylvania and New Jer- sey DOTs jointly developed a generalized checklist to help ensure that alternatives are inclusive (Table 6). In addition, the guidance provides checklists to help develop alternatives to address specific issues, including mainline congestion, resurfacing, intersection congestion, bridge deficiency, and intersection safety (Table 7). A feature of the Arizona NBIP is that citizen stake- holders play an active role in the planning, design, and construction of the corridor. On the State Route 179 cor- ridor in Sedona, Arizona, the team prepared a series of “fishbone” diagrams indicating how community members linked together cross-sectional ideas into segment-level concepts. This visual representation of the potential cross sections of the roadway helped ensure that the full range of alternatives was considered. Figure 9 shows proposed combinations of roadway design elements for one segment of the corridor (48). Often a red flag or fatal flaw analysis of alternatives is performed early in the alternative development process. The idea of this analysis is to examine key areas such as environ- mental, historic, social, ROW, utility, and engineering (e.g., geometric, geotechnical), and identify locations of concern. Many STAs have developed red flag checklists to aid in this process. Once a location of concern has been identified, it must be determined if the issue is a “fatal flaw.” Fatal flaws are typically associated with significant negative economic, environmental, or historic impacts. Alternatives that have fatal flaws are then removed from further consideration. However, this process must be used judiciously, as the integ- rity of the process requires that all reasonable alternatives receive a fair assessment. of safety and mobility associated with highway design, and of livability, natural environment, and aesthetics associated with community character and values. It suggests focusing the analysis to inform all parties of the needs and expectations of the other involved parties and of the benefits and drawbacks to many of the ele- ments frequently included in transportation projects. • Long-Term Impacts – The guidance discusses in detail the trade-offs associated with many design elements and the liability issues that may arise owing to inap- propriate designs, and encourages thoughtful consid- eration of the trade-offs associated with the varied treatments available to the designer (46). The Arizona Department of Transportation has developed a CSS/CSD approach called the needs-based implementation plan (NBIP), which consists of a coordinated, collaborative team effort to assess needs and develop solutions for a cor- ridor. The NBIP process takes a context-sensitive approach by balancing safety, mobility, and the preservation of scenic, aes- thetic, historic, environmental, and other community values. Figure 8 shows the three phases of the NBIP process: process definition, corridor-wide framework, and segment concept design. In the process definition phase, the project team defines specific tasks to be completed in the remaining phases. In the corridor-wide framework phase, the project team helps the community arrive at a preferred planning concept for each corridor segment. Finally, the segment concept design phase allows community representatives to help the project team select more detailed design elements for each segment (47). Develop a Full Range of Alternatives In Smart Transportation Guidebook: Planning and Design- ing Highways and Streets that Support Sustainable and FIGURE 8 Needs-based implementation plan three-phase process. Source: Rauch (48).

23 TABLE 6 CHECKLIST FOR EXPLORING ALTERNATIVES Source: Smart Transportation Guidebook: Planning and Designing Highways and Streets that Support Sustainable and Livable Communities (37). TABLE 7 RANGE OF SOLUTIONS FOR MAINLINE CONGESTION Source: Smart Transportation Guidebook: Planning and Designing Highways and Streets that Support Sustainable and Livable Communities (37). FIGURE 9 Fishbone diagrams of roadway cross-sectional elements. Source: Rauch (48).

24 Measures of Effectiveness It is important that MOEs be directly related to the stated project needs. These needs often focus on the overarching categories of physical condition, operational performance, safety, access, environment, and social and historical. When possible, the MOEs selected for use in alternative analysis are to be standard, widely accepted measures. For example, when dealing with operational performance, the use of level of service, hours of delay, or total travel time would be appro- priate. The key is for chosen MOEs to be transparent and easily conveyed to all stakeholders. Achieving CSS/CSD means generating project outcomes that reflect community values; are sensitive to scenic, aesthetic, historic, and natural resources; and are safe and financially feasible (49). It is critical to identify MOEs early in the process with direct input from stakeholders. Even if only a few measures Table 8 continued on p.25 TABLE 8 EXAMPLES OF SMART TRANSPORTATION MEASURES OF SUCCESS

25 are finally selected for project evaluation, consideration of a wide range of measures at the beginning of a project can help identify important community values that may otherwise be overlooked. Further, it is critical to develop MOEs before alternatives have been formulated to help prevent stakehold- ers from attempting to steer the analysis toward certain types of alternative solutions instead of focusing on establishing MOEs to capture what is most important. The Smart Trans- portation Guidebook: Planning and Designing Highways and Streets that Support Sustainable and Livable Communi- ties (37) presents a table listing potential MOEs by category (Table 8). Although not all-inclusive, referencing this sort of tool early in the project planning process is vital to ensure that sufficient and appropriate MOEs will be available to address project purpose and need. Evaluation criteria can be quantitative or qualitative, depending on the complexity of the problem, the expected level of controversy, the structure and scope of the public involvement process, and the preference of decision makers. In general, projects involving difficult trade-offs and high degrees of controversy benefit from the use of quantitative measures. The use of appropriate MOEs and defensible data can help focus stakeholders on outcomes, remove emotional bias from the discussion and provide definitive rationale to support trade-off decisions to help move stakeholders off of strongly held positions. Whether the criteria are qualitative or quantitative, they help to focus the data collection and the discussion on the relative merits of the alternatives in relation to critical issues, and on trade-offs that distinguish among the alternatives under consideration (4). Table 8 continued from p.24 Source: Smart Transportation Guidebook: Planning and Designing Highways and Streets that Support Sustainable and Livable Communities (37).

26 time savings for a representative trip may be used to better understand the return on a proposed investment (37). Figure 10 illustrates this concept. The x-axis represents total project costs such as capital costs; life-cycle costs; environmental, historical, and social impacts; and user costs. The y-axis represents project benefits such as travel time savings, crash reduction, emission reduction, and improved access. Alternatives are examined to see where they fall along the continuum, with the overall goal being to “right-size” a project. FIGURE 10 Value to price curve. Source: Smart Transportation Guidebook: Planning and Designing Highways and Streets that Support Sustainable and Livable Communities (37). Benefit/cost analysis needs to be used with caution when examining transportation projects. It can be difficult to adequately represent factors such as safety, equity, and eco- nomic development in a benefit/cost analysis. Project Documentation and Decision Process One of the critical aspects of trade-off decisions is ensuring that the decision is adequately documented and then that the documentation is retained, as both crashes and tort claims may occur many years after the decisions and construc- tion. In such cases, the agency’s actions may be defended by professionals who were not directly involved in the actual project execution. It is unfortunately the case that design agencies lose or settle claims not because their staff actions were inappropriate, but because the project files are incom- plete or missing key documentation, and staff responsible for the project are no longer available to explain what was done and why (4). Putting all aspects of a project in writing is important not only to improve communication among team members and the community, but also to protect against liti- gation and ease the transition when staff change. Pennsyl- vania Department of Transportation’s (PennDOT’s) CSS/ The MassHighway Project Development and Design Guide (42) departed from the traditional measure of vehicu- lar Level of Service owing to its limited way of measuring the benefits of a transportation project. To account for this, the guidance provides new basic design controls associ- ated with measures of effectiveness. These MOEs include levels of service for all users (pedestrians, bicyclists and motorists), facility condition, safety, resource preservation, aesthetics, accessibility, environmental justice and other factors. This allows the designer to better assess trade-offs, make more informed decisions, and include project elements in the proposed design that respond to the real needs of the community, beyond the simple (and not always desirable) need to move more cars through an area at faster speeds (43). This approach will be reflected in refinements to the Highway Capacity Manual for the 2010 update, which will provide procedures for determining multimodal quality of service and level of service. ITE’s Context Sensitive Solutions in Designing Major Urban Thoroughfares for Walkable Communities identifies several MOE categories: • Mobility: travel demand, roadway capacity, level of service, travel time, connectivity, circulation, access, truck movement, access to multiple travel modes, etc.; • Social and Economic Effects: socioeconomic and cul- tural environment (historic, cultural, and archaeologi- cal resources; residential and business displacement/ dislocation; socioeconomics and equity; neighborhood integrity and cohesion; economic development; place- making qualities); • Environmental Effects: positive and negative effects of natural environment (air quality, noise, energy con- sumption, water quality and quantity, vegetation, wild- life, soils, open space, park lands, ecologically significant areas, drainage/flooding aesthetics and visual quality); land use (residential patterns, compatible uses, develop- ment suitability according to community values, etc.); • Cost-effectiveness and Affordability: capital costs, operations and maintenance costs, achievement of ben- efits commensurate with resource commitment, suffi- ciency of revenues, etc.; and • Other Factors: compatibility with local and regional plans and policies, constructability, construction effects, etc. (40). Benefit-Cost Ratio Most projects offer a range of alternatives with different costs, corresponding to different levels of value. However, the importance of understanding alternatives based on a value- to-price ratio is often overlooked. Current guidance is fairly silent on this subject and does not direct projects toward the most effective value-to-price yield. Performance measures such as cost per existing trip, cost per new trip, and cost per

27 CSD training, for instance, includes a session by the attorney general’s office that discusses the need to document the use of sound engineering judgment. Such practices offer more protection than blind adherence to the maximum values in design manuals (50). As important as adequate documentation is the need for a clearly understood decision process. Projects are made up of a series of tasks, each of which may involve a series of trade- off decisions that result in a final project decision. As such, it is critical that the decision process identify what decisions will be made and by whom, and what analyses, processes, and documents will be produced to support the final out- come. A successful process for making trade-off decisions contains several crucial elements. Develop Decision Process The purpose of developing a decision process is to identify the problem completely and accurately, select the best alter- native, enhance agency credibility, and make efficient use of resources—in short, to make good transportation invest- ment decisions. A decision process incorporates the follow- ing elements: • The decision points in the process • Who will make each decision • Who will make recommendations for each decision • Who will be consulted on each decision • How recommendations and comments will be trans- mitted to decision makers. Decision Points The CSS/CSD Project Development Process includes a rec- ommended set of decision points. These basic steps will sup- port almost any planning process, but may need to be refined to suit a particular project. The focus of a decision process is often mistakenly placed on only the final decision, over- looking the many intermediate decisions. For example, in an alternative selection process, the alternative development and screening occurs before detailed alternative evaluation. Whether or not it is explicitly stated, the early steps involve decisions on compiling the list of potential alternatives, the manner and level of detail in which they will be outlined or described, the feasibility criteria to be used, and the list of feasible alternatives to be considered further. Breaking down larger decisions into their component pieces also helps to identify the differences in stakeholder involvement needed at various points in the process. It may be important for different stakeholders to be involved at vari- ous decision points, or for different parties to make differ- ent decisions. For example, some decisions require specific technical expertise, whereas others require broader partici- pation and perhaps less technically oriented input (4). Case Studies Several guidance documents, including the following, pro- vide CSS/CSD case studies: • Building Projects That Build Communities (31) • Understanding Flexibility in Transportation Design— Washington (45) • CSS National Dialog (51) • Context Sensitive Solutions in Designing Major Urban Thoroughfares for Walkable Communities (40) • A Guide to Best Practices for Achieving Context- Sensitive Solutions (4) • Context-Sensitive Design Around the Country: Some Examples (27) • Quantifying the Benefits of Context Sensitive Solutions (34) The CSS National Dialog is a collection of case studies that were submitted to the National Dialog and have been transferred to the CSS clearinghouse database (51). This database is searchable by keyword or name of project and can be accessed at www.contextsensitivesolutions.org. PRACTICAL SOLUTIONS/DESIGN Practical Solutions and Practical Design (Practical Solutions/ Design) refers to a design process that attempts to maximize the rate of return for the individual project while maximizing the rate of return for the complete system. This focus on sys- temwide optimization is not adequately addressed by CSS/ CSD, which focuses primarily on incorporating all relevant factors into the project development process to account for context. Practical Solutions/Design, in contrast, acknowl- edges that there are finite resources that can be expended on the transportation system as a whole. The intent of the Practical Solutions/Design process is not to optimize the individual project, but rather to allow increased optimization of the entire transportation system with concern for mobil- ity and safety. This approach results in “reasonable” solu- tions for individual projects while preserving limited funds to address additional problems elsewhere in the system. To achieve this, the following general principles are utilized to control the potential for overdesign of a project: • Targeted Goals in a P&N Statement – The P&N statement identifies specific targets for performance and avoids generalized statements, such as “improve mobility.” For example, a specific target could be to shorten intersection delay to less than 50 seconds per vehicle during the typical peak hour. • Meeting Anticipated Capacity Needs – Rather than using a broad level of service goals, more detailed quality of service targets need to be identified. • Safety Evaluation Against Existing Conditions – The safety of each alternative is to be compared as incre-

28 mental gains from the existing conditions. Simple selection of the alternative that has been judged to be “safer” misses the opportunity to evaluate the safety gains based on the marginal rate of return. • Maximize Rate of Return – Increased investment yields less than a proportional increase in overall value at the point of diminishing returns (see Figure 10) (17). Performance-Based Planning Performance-based planning has most often been applied at the organizational level to assess program conformance with a stated overall transportation plan or goals and objec- tives. However, the process is just as applicable for evaluat- ing trade-offs associated with alternatives, especially when put in the context of CSS/CSD. The use of outcome-based performance measures, as opposed to the traditional use of output-based performance measures, is especially apt. Furthermore, as CSS/CSD becomes the accepted practice for STAs, there is a need to gauge performance in meeting strategic goals and objectives. Figure 11 illustrates a general framework for applying performance-based planning. Output measures generally reflect the quantity of resources used, the scale or scope of activities performed by an organization, and the efficiency in converting those resources into some type of product. Output measures are most often used as indicators of organizational activity or performance, but stop short of identifying results as viewed by intended beneficiaries. Nonetheless, outcome measures reflect success in meeting stated goals and objectives, which can be drawn from the project P&N statement, and can be augmented with customer satisfaction measures that focus on the beneficiaries of the project. A Guidebook for Perfor- mance-Based Transportation Planning (52), for instance, includes a Performance Measures Library as Appendix A. Its purpose is to provide practitioners of performance-based planning with a concise guide to many of the measures in use around the United States today. Context-sensitive project solutions often appear decep- tively simple, yet the holistic, interdisciplinary, commu- nity-driven nature of CSS-based project delivery makes measurement challenging. Performance Measures for Con- text Sensitive Solutions: A Guidebook for State DOTs (53) presents a framework for organizing performance measures and discusses key focus areas. The guidance assumes that individual agencies will develop their own MOEs tailored their specific needs. It contains four major sections: • Guiding Concepts for CSS Performance Measurement Programs. This section offers DOTs a framework for organizing measures that addresses CSS-related pro- cesses and outcomes at the project level and organiza- tionwide, and provides an understanding of some basic principles for measurement of CSS performance. • Project-level Focus Areas. This section describes how agencies can assess performance of individual projects or groups of projects by targeting key focus areas, and gives pointers for potential performance measures in each focus area. • Organizationwide Focus Areas. This section describes focus areas that agencies need to target as they assess overall organizational performance, and gives pointers for potential performance measures in each focus area. • Tips for Getting Started. This section provides sugges- tions on creating and using a CSS performance mea- sures framework. FIGURE 11 Elements of a performance-based planning process. Source: NCHRP Report 446: A Guidebook for Performance-Based Planning (53).

29 • Establish project scope • Fast track project development • Improve interagency communications • Bridge institutional borders • Better balance the needs of road users and those of the community or the environment • Reach consensus on difficult issues (54). Relation to CSS/CSD It is important to understand how the CSS/CSD and VE pro- cesses are related. • Both are goal-driven processes with important com- mon objectives. • Both place a high value on customer satisfaction, stakeholder values, and delivery or maintenance of safe facilities. • Both stress efficient and effective use of resources and aligning investments to produce the “best” outcomes or most “quality” for the money. • Both are open to a broad interpretation of what is “best” and face challenges in measuring performance in more difficult to quantify areas. Further, the guidance contains an appendix with a variety of relevant performance measures. The customer-oriented focus of performance-based plan- ning can help project managers and their teams to do their jobs better by maintaining a focus on the whole range of cus- tomer needs for transportation projects. Performance-based planning can be used to evaluate the CSS/CSD process and outcomes at both the project and agency levels (Figure 12). VALUE ENGINEERING The value engineering (VE) process is a powerful decision- making process, as using the common language of functions enables an interdisciplinary team to communicate more effec- tively to arrive at a supportable decision. VE is extremely use- ful to define the project concept, as the process of identifying the functions associated with the project objectives ensures that all participants clearly understand how the project deci- sion-making process affects these objectives. This process can eliminate confusion and conflicting viewpoints among stakeholders and reduce the overall time to reach an optimal solution. VE is being used to engage stakeholders to— FIGURE 12 Integration of performance-based planning. Source: NCHRP Web Document 69: Performance Measures for Context Sensitive Solutions: A Guidebook for State DOTs (54).

30 attempts at this quantification. The Victoria Transportation Policy Institute has produced a number of publications that begin to monetize environmental and community quality of life factors related to transportation investments. The Utah DOT also includes “user impacts” in its definition of life- cycle cost. However, even without monetization of costs, the use of interdisciplinary teams with members of the public and resource agencies help ensure that qualitative CSS/CSD considerations are incorporated into the VE process (55). FIGURE 13 Value engineering methodology in relation to CSS/CSD inputs. Source: Osman et al. (56). Value Engineering Process The systematic process used in VE is called the job plan. The VE job plan is organized into three major compo- nents: prestudy, value study, and poststudy. Figure 14 illus- trates the job plan process flow. As can be seen, the value study is further broken down into six phases: information, function analysis, creative, evaluation, development, and presentation. FIGURE 14 Job plan process flow. Source: “Value Standard and Body of Knowledge” (57). The VE focus on extracting the functional requirements and ensuring that they are integrated into the design makes its methodology well suited to examine trade-offs in design. The VE methodology can be used to determine user func- • Both share an emphasis on transparency, interdisci- plinary teams and perspectives, and analytical evi- dence-based decision making. • Both employ processes that examine multiple alterna- tives with the end goal of achieving consensus on the ultimate decision. • Both seek to optimize functions that can be delivered for the cost. • Both seek to improve value and quality and take a broader “life cycle” view of an alternative. • Both examine the use of flexibility in design to increase community, aesthetic, and environmental qualities through the assessment of risk. • Both rely on interdisciplinary teams to achieve desired outcomes. • Both emphasize creative and innovative solutions to improve the final design (55). A comparison of value engineering steps with those under- taken in a CSS/CSD approach reveals similar steps in both processes (Table 9). Figure 13 illustrates how sections of the VE process can feed needed inputs to the CSS/CSD process. TABLE 9 COMPARISON OF VE AND CSS/CSD PROCESSES Value Engineering Steps Similar Steps in CSS/CSD Project (pre-study): Selection of the projects, processes, or elements for evaluation DOT may decide to apply CSS approach to all projects. Major proj- ects often get more extensive process. Investigation: Background infor- mation (context is one factor), function analysis, team focus (WSDOT includes stakeholders) Convene team, including stakeholders Investigate, learn about context, understand and discuss purpose and need, functions Speculation: Creative, brain- storming, alternative proposals Listening, brainstorming, alternative proposals Evaluation: Analysis of alterna- tives, what are life-cycle cost impacts, which delivers highest value overall? Understand tradeoffs. Reach con- sensus, if possible, on alternative delivering the most value to the public Development: Develop technical and economic supporting data Document decisions and why they were chosen Present recommendations/find- ings. Fair evaluation. Present arguments Implementation of VE recommendations Implementation of CSS recommen- dations—sometimes a commitment tracking system provides support Audit: Review of completed results, accomplishments, and awards. Audit: Review of completed results, accomplishments, and awards. (post-study) (post-study) Source: Venner et al. (55). One of the challenges in using VE principles to make trade-off decisions in design is associated with the diffi- culty in monetizing key factors under consideration (e.g., quality of life factors). However, some agencies have made

31 the purpose or mission of the subject under study. (Promote Street Life) • Objective or specifications. Particular parameters or restrictions that must be achieved to satisfy the highest order function in its operating environment. Although they are not themselves functions, they may influence the concept selected to best achieve the basic function and to satisfy the users’ requirements. (Provide Good Walking Environment) • Dependent functions. Starting with the first function to the right of the basic function, each successive function is considered “dependent” on the one to its immediate left. (Provide Safe Intersections) • Independent functions. These functions are not dependent on another function or method selected to perform that function. They are considered secondary with respects to the scope and critical path. (Provide Street Art) (57). Figure 17 presents an example of a FAST diagram that was developed for a recent CSS/CSD-focused value plan- ning study. CHOOSING BY ADVANTAGES Choosing by Advantages (CBA) was developed by the U.S. Forest Service in the early 1980s to assist decision makers in making informed choices on program expenditures. CBA dif- fers from other decision-making systems in that it concentrates only on the difference between the advantages of alternatives tional requirements, analyze these functional requirements from the abstract to the specific, and strike a balance between user functional requirements and safety constraints. The Function Analysis System Technique (FAST) cre- ated by Charles Bytheway is an evolution of the VE process. FAST permits people with different technical backgrounds to effectively communicate and resolve issues that require mul- tidisciplined considerations. FAST links simply expressed verb-noun functions to describe complex systems. FAST diagrams (Figure 15) are built from left to right, starting with the higher order functions that are then decomposed to functions of lower order as the diagram evolves to the right. The vertical lines delimit the scope of the VE study. Figure 16 is an example of a transportation-oriented FAST diagram. The FAST diagram is made up of several key compo- nents. For each of these components, a representative value presented in the example FAST diagram (see Figure 16) is also provided. • Highest order function. Appears outside the leftmost scope line and represents the objective or output of the basic function or subject under study. (Enhance Social Ties) • Lowest order function. Appears outside the rightmost scope line and represents the function that initiates the study. (Reshape Transit) • Basic functions. The functions represented to the immediate right of the leftmost scope line represent FIGURE 15 The basic FAST diagram. Source: “Function Analysis Systems Technique—The Basics” (58).

32 FIGURE 16 FAST diagram for enhancing social ties. Source: Osman et al. (56). FIGURE 17 CSS/CSD-oriented FAST diagram. Source: Wilson (54).

33 being compared. Unfortunately, participants who have a hidden agenda have the opportunity to attempt to “game” the outcome of this decision-making process. For this reason, the use of an experienced, strong facilitator is recommended to emphasize the need for an open and fair treatment of all alternatives and to foster an environment of trust among participants. The CBA approach involves summarizing the attributes of each alternative, deciding the advantages of each alterna- tive, deciding the relative importance of each advantage, and developing incremental costs and incremental advantages (52). In the CBA vocabulary— • Factor is an element or a component of a decision and is a container for criteria, attributes, advantages and other types of data. • Criterion is any standard upon which a judgment is based. • Attribute is a characteristic or consequence of one alternative. • Advantage is a difference between the attributes of two alternatives. CBA focuses on the differences between alternatives and determines how important those differences really are. Ele- ments that are the same for each alternative will make no difference in the selection of the preferred alternative and therefore are not considered. This process allows the interdis- ciplinary team to focus discussion on the areas where there are truly differences among alternatives. At the final phase, cost is introduced to the evaluation process to establish an importance- to-cost ratio to determine which alternatives or components of larger plans provide the greatest benefit per dollar spent. CBA is a decision-making system based on the principle that a difference between two alternatives is an advantage for one alternative and a disadvantage for the other. To sim- plify the decision-making process and avoid repetition, only the advantages are considered. The use of CBA provides a logical, trackable linkage between the factors used to iden- tify the preferred alternative and the major trade-offs among the alternatives. In its guidance on General Management Planning (59), the National Park System outlines the CBA process and provides a simple example of how it works. In the exam- ple, a group is planning to go camping and has to choose between several available campsites using the CBA pro- cess. Table 10 outlines the factors, attributes, and advan- tages for the example. The CBA decision-making process has five basic steps. 1. Summarize the attributes of each alternative. 2. Decide the advantages of each alternative. 3. Decide the importance of each advantage. 4. Weigh costs with total importance of the advantages. 5. Summarize the decision. The following discussion demonstrates how the CBA analysis will help the camper make a campsite selection uti- lizing a CBA spreadsheet (Table 11). TABLE 10 AN EXAMPLE OF THE CBA PROCESS Source: General Management Plan Dynamic Sourcebook (59).

34 Step 1: Summarize the Attributes of Each Alternative The attributes for each alternative are identified in the spreadsheet by factor on the line titled “attributes.” No value judgment is made regarding these attributes. For example, for Site 8 and the factor addressing water, the attribute is that water is “60 feet away.” Step 2: Decide the Advantages of Each Alternative To determine the advantages, it is important that the group share an understanding of what attribute provides an advan- tage. In the example provided, everyone must first agree that being closer to water provides more advantage than being farther away. It is important to provide clear descriptions of the advantages, as they will be used later to summarize the rationale for the decision. The least preferred attribute is underlined for each fac- tor, and then the advantages of the other alternatives are described relative to the least preferred attribute. There is no advantage for the least preferred attribute, so leave it blank. For example, for Site 8 and the factor addressing water, the advantage is that water is “200 feet closer.” Step 3: Decide the Importance of Each Advantage There are four considerations for deciding importance: 1. The purpose and circumstances of the decision 2. The needs and preferences of the users and stakeholders 3. The magnitudes of the advantages 4. The magnitudes of the associated attributes. Using these four considerations, the most important advantage is determined for each factor and circled. In the example, for the factor addressing water, the fact that Site 8’s water is 200 feet closer was considered the most important advantage for that factor and circled. Next, the paramount advantage is selected from the important advantages (i.e., those just circled). It is critical to make the distinction that this is not the most important factor, but is the most important advantage (i.e., difference) between the alternatives. This paramount advantage will be used as the benchmark against which all other advantages will be compared and is assigned an importance score of 100. In the example, “much more privacy due to screening and remoteness” was selected as the paramount advantage and assigned an importance score of 100. It is often difficult to decide which advantage is para- mount. A useful technique to simplify this decision is to use the “defender/challenger” method. Two advantages are selected and the group determines which advantage is more important in the decision. The chosen advantage is then pit- ted against another advantage. This process continues until only one remains: the paramount advantage. Then, each of the remaining advantages will be assigned an importance score relative to the paramount advantage. The scale of importance has been established from 0 (not important) to 100 (of paramount importance). In the example, being “200 feet closer” to water was given an importance score of 40. Next, an importance score for each of the remaining advan- TABLE 11 CBA EXAMPLE SPREADSHEET Source: General Management Plan Dynamic Sourcebook (59).

35 tages is determined, keeping in mind that the score must be less than the score assigned to the most important advan- tage. The least important advantage (identified by the under- line) receives a 0. If advantages are identical, they receive the same score. In the example for the factor addressing a table, neither Site 8 nor Site 19 has a table. As this is the least preferred attribute for factor 3, both are given a score of 0. Once importance scores have been assigned for each advan- tage, it is important to cross-check the logic involved to ensure that the decisions were made consistently. In the example, an importance score of 30 was assigned to both Site 23 under fac- tor 1 and to Site 6 under factor 2. Are these really equal? Finally, the importance scores for each alternative are totaled. If costs were equal or were not an issue, the alter- native with the highest total importance score would be selected. In the example, Site 19 has a total importance score of 170, and thus has the greatest advantages. Step 4: Weigh Costs with Total Importance of Advantages If costs are not equal, then it must be determined whether the additional advantages justify the additional cost. In the example, Site 19 had the greatest total importance score of 170, whereas Site 8 had the lowest total score of 70; however, these advantages may not actually be worth six times the cost. Graphing the importance-to-cost data provides a visual way to assist in decision making. A steep slope upward indi- cates that there is a great increase in the total importance of advantages for not much more money, and hence it may be a good value. A shallow slope, no slope, or a decreasing slope indicates that although more money is being spent, there is no corresponding increase in the importance of advan- tages and therefore it is not a good value. Figure 18 shows an importance-to-cost graph for the example. FIGURE 18 Importance-to-Cost Graph Source: General Management Plan Dynamic Sourcebook, Version 2.1, U.S. Department of the Interior, National Park Service, March 2008 [Online]. Available: http://planning.nps.gov/GMPSourcebook/ GMPHome.htm. At this point, a decision is made regarding which alter- native to select. It is important to note that CBA does not provide a mechanism for making this choice; the CBA pro- cess informs the decision maker. The detailed examination of advantages also provides an opportunity to improve the preferred alternative by incorporating key advantages from alternatives not selected. Step 5: Summarize the Decision Using the advantage statements developed during the CBA process and notes from the discussion, a summary of the reasoning for selecting the preferred alternative is docu- mented. Based on the example problem, Site 23 was selected because it has the following advantages: moderately more private, 110 feet closer to water, has a picnic table (other sites do not), and is the greatest value. RISK ANALYSIS AND MANAGEMENT Risk is perceived as the effect of uncertainty on a project or organizational objectives and represents exposure to mis- chance, hazards, and the possibility of adverse consequences (59). Risk tolerance is the varying degrees of risk organizations or stakeholders are willing to accept. A risk can be acceptable for inclusion on a project when the risk is within acceptable tolerance and is balanced by a desirable benefit. For example, it may be desirable to use 11-foot lanes on a project in order to free up enough room in the roadway cross section to provide bike lanes. A risk analysis may be used to support decision making regarding the trade-offs between various alternatives. One of the most common ways to account for risk and uncertainty in engineering design is through a factor of safety. However, this approach does not provide any quan- titative idea of the risk in a particular situation or trade-off, can result in unnecessary overdesign, and still leaves uncer- tainty for decision makers. Ideally, such decisions are to be taken in an environment of total certainty, wherein all the necessary information is available for making the right deci- sion and the outcome can be predicted with a high degree of confidence. In reality, most decisions are taken without complete information, and therefore give rise to some degree of uncertainty in the outcome (Figure 19). FIGURE 19 The uncertainty spectrum. Source: Wideman (60).

36 Project risk analysis is a four-step process (Figure 20). First, the project objectives are determined; second, risks associated with the project are identified; third, the risks are prioritized; and fourth, control measures are identified to deal with high-priority risks. Often, a fifth step is included to document and communicate the risk determination. FIGURE 20 Project risk analysis. Source: Well- Stam et al. (62). Determine the Objective The P&N typically describes the objective(s) of the project. Identify the Risks This phase consists of identifying all of the possible risks that may significantly affect the success of the project. Conceptu- ally, risks may range from high impact/high probability to low impact/low probability. Note that combinations of risk may together pose a greater threat than each risk individually. A risk breakdown structure (RBS) (Figure 21) identifies categories and subcategories of risk. Once an RBS has been developed for a type of project, it can be reused on similar projects to remind participants in a risk identification exer- cise of the potential sources of risk. Similar to RBS, a risk identification checklist can be developed based on historical project information and expert knowledge (61). Another input to the risk identification task can be dia- gramming techniques, intended to help identify where risks may affect a project. Several types of diagramming are useful: • Cause and Effect Diagram. Also known as Ishikawa or fishbone diagrams, these are useful for identifying the causes of risk. (Figure 22). • System or Process Flow Chart. These show how vari- ous elements of the project interrelate. • Influence Diagram. These graphical representations show causal influences, time ordering of events, and other relationships among variables and outcomes (61). Risk elements that tend to attract or determine the response attitudes of decision makers include the following: • Potential frequency of loss • Amount and reliability of information available • Potential severity of loss • Manageability of the risk • Vividness of the consequences • Potential for (adverse) publicity • Ability to measure the consequences • Whose money it is (60) The goals of risk analysis and management are to— • Increase the understanding of the project; • Identify the alternatives available; • Ensure that uncertainties and risks are adequately considered in a structured and systematic way, which allows them to be incorporated into the planning and project development process; and • Establish the implications of risk on all other aspects of the project through direct examination of project uncertainties (53). The consequences of risk can be shared by multiple par- ties or shifted from one party to another. When associated with transportation projects, these concepts relate to the idea of making design trade-offs that result in changes to the risk exposure of different user groups. For example, narrowing travel lanes to provide a roadside buffer for a multiuse path is an example of shifting some risk from the pedestrian and bicyclist users of the multiuse path to the motorized users of the roadway. When coupled with the concept of Practical Solutions/Design, the risk can be shifted or shared from one project to another or between a project and the system as a whole. This section will examine three different treatments of risk, from the project management field, the risk and reli- ability field, and the risk management field. Project Risk Analysis Project risk is the cumulative effect of the chances of uncer- tain occurrences adversely affecting project objectives. In other words, it is the degree of exposure to negative events, and their probable consequences affecting project objectives, as expressed in terms of scope, quality, time, and cost (60). The Project Management Institute (PMI) has challenged the “negative” view of risk as being too restrictive and incom- plete. PMI believes that risk is essentially “neutral” and rep- resents both opportunities and threats. Under this approach, accepting risk leads to either a positive or a negative outcome, depending on the application of risk management (61).

37 Several risk identification techniques have been devel- oped for project planning: • Brainstorming. An interdisciplinary team of experts who are not part of the project team generate a com- prehensive list of potential project risks under the leadership of a facilitator. This is a simple but effec- tive approach to help participants think creatively in a group setting without fear of criticism. • Delphi Technique. Project risk experts participate in this exercise anonymously. A trained facilitator uses a questionnaire to solicit ideas about project risks. The responses are then summarized and recirculated to the experts to elicit further comment. This process is repeated until consensus is reached. • Interviewing. A risk analyst interviews key project participants, stakeholders, and outside subject matter experts to develop a list of project risks. • Root Cause Analysis. This technique identifies a prob- lem, determines the underlying cause, and prepares a preventative action (61). • SWOT Analysis. This process evaluates a project with regard to its strengths, weaknesses, opportunities, and threats (SWOT) and provides a framework for reviewing a project or specific decision. These fac- tors are typically identified through brainstorming, and the structure of the analysis requires participants to focus on proactive thinking, rather than relying on habitual or instinctive reactions. A SWOT analysis starts with the identification of a projects goals and objectives. In transportation planning, these have typically been stated in the P&N document created as part of the NEPA process. Strengths and weakness are related to factors internal to the project, whereas opportunities and threats are related to factors exter- nal to the project. FIGURE 21 Example of a simple risk breakdown structure (RBS). Source: A Guide to the Project Management Body of Knowledge (PMBOK Guide) (61). FIGURE 22 Example cause and effect diagram. Source: A Guide to the Project Management Body of Knowledge (PMBOK Guide) (61).

38 ensures that all participants address risk management from a common perspective (63). The two most common methods to assess risk in a qualitative manner are assigning points to the most significant risks, and assessing probability and consequence separately using numbers. The assigning points method has some ground rules to prevent one person from exerting a high degree of influence on the total by assigning all of the points to a single risk or diluting the process by assigning equal points to each of the risks. Figure 23 shows a sample risk matrix created using this method. In this example, each participant was given 100 points to assign to 10 different identified risks, with points to be assigned to a minimum of 3 risks and a maximum of 7 risks. To determine the prioritization of the risks identified, the points are added up for each risk and the risks are ranked by the number of points assigned to them. In the assessing probability method, the risk is divided into two concepts: probability and consequence. Risk equals the probability assigned, multiplied by the consequence. This approach does not use an absolute quantification of probability and consequence; rather, the assessment uses a simple classification scheme represented by a numeric scale. It is important to make sure the scale contains an even num- ber of values to force a choice, as no neutral or midpoint value is available. In the risk matrix shown in Figure 24, both the probability and consequence are assessed on a scale of 1 to 4. The probability classifications are 1—unlikely (5%), 2—possible (25%), 3—likely (75%), and 4—nearly certain (95%). The consequence classifications are 1—insignificant, 2—minor, 3—moderate, and 4—major. The risk with the highest risk score, in this case 12, is ranked as the first pri- ority. A scale of 1–10 for the classification schemes is also common, depending on the complexity of the project. • Sensitivity Analysis. This analysis seeks to place a value on the effect of changing a single variable within a project by analyzing that effect on the overall project. Uncertainty and risk are reflected by defining a likely range of variation for each component of the alterna- tive. The impacts of these variations can then be under- stood in the overall context of the project (60). When using a process that does not control for participant bias, such as the Delphi Technique, the bias of the partici- pants must be taken into consideration to make sure there are no blind spots in the process. Regardless of the risk identification technique used, it is also vitally important to document the assumptions made regarding the project and identified risks. The validity of these assumptions needs to be revisited at key points during the project development process. Project risk analysis is not a static event. Risks can either be overcome or decrease as a result of measures being imple- mented. As such, the risks associated with trade-offs must be reexamined at key points in a project to ensure that the analysis that underlies the decision is still valid after a par- ticular trade-off has been accepted or a mitigation measure put into place. Evaluation of risk is a continuing, integrative function of the project life cycle. Prioritize the Risk During the identification process, a number of risks may emerge. It is unproductive to focus attention on all of the risks that have been identified; priority must be given to the most significant risks in order to evaluate the trade-offs associated with them. A risk management matrix provides a systematic approach to analyzing and addressing risks and FIGURE 23 Sample risk matrix using the assigning points method. Source: Well- Stam et al. (62). FIGURE 24 Sample risk matrix using the assessing probability and consequence method. Source: Well-Stam et al. (62).

39 Each of the methods for qualitatively prioritizing risk has advantages and disadvantages, as summarized in Table 12. A further drawback to the assessing probability and conse- quence method is that it equates risk of a high-probability, low-damage event with that of a low-probability, high-dam- age event. Clearly, in real life these two events may not amount to the same risk (64). TABLE 12 ADVANTAGES AND DISADVANTAGES OF THE DIFFERENT METHODS Assigning Points Probability and Consequence Classes Advantages Advantages Quick; it is not necessary to assess all of the risks individually Provides a good first insight into priority Forces one to make a distinction between probability and consequence Less room for implicit inclusion of factors other than probability and consequence Disadvantages Disadvantages Room for implicit inclusion of fac- tors other than probability or consequence Very time-consuming; each risk must be individually assessed Source: Well-Stam et al. (62). Identify Mitigation The process of identifying mitigation strategies deals directly with risk response. These risk mitigation strategies form the response component of a risk management strategy (62). Document and Communicate Risk Final documentation is a vital part of project risk analysis so that appropriate trade-off decisions can be made with full knowledge of the apparent risks involved. Either the residual risks must be accepted or the alternative must be abandoned. Further, the purpose is to build a base of reliable data for the continuing evaluation of risk on the current project, as well as for improving the data for all subsequent projects (e.g., developing an improved RBS). Risk and Reliability Analysis Figure 25 illustrates the overall philosophy of decision mak- ing. This philosophy presents decision making through a triangle whose three vertices are benefit-cost theory, deci- sion theory, and sustainability theory. As shown, risk and reliability analysis occupies a central place in the interaction of the concepts of certainty and uncertainty, efficiency and equity, and single and collective decision making. The rela- tive importance of the vertices can change based on what a society values. The greater focus on sustainability and equity today results in decision making becoming participa- tory. To emphasize the importance of these principles, they are placed at the apex of the decision triangle. The ultimate goal of risk and reliability analysis is to reduce uncertainty and thereby reduce risk (64). FIGURE 25 Decision triangle. Source: Singh et al. (64). Engineering projects are always subject to a risk of failure to achieve their objectives. There are various definitions of risk for different purposes, including the probability of failure, the reciprocal of the expected length of time before failure, the expected cost of failure, and the actual cost of failure. Risks are possibilities that human activities or natural events lead to consequences that affect the intended objectives. In general terms, risk can be defined as the potential loss resulting from the combination of hazard and vulnerability. Reliability, the complement of risk, is defined as the probability of nonfailure. One of the dilemmas in risk assessment is defining “acceptable risk.” The acceptability depends on the context in which the risk occurs and the availability of financial resources. Logically, a decision maker will choose the opti- mum mixture of risk, cost, and benefit, and might be willing to take a higher risk only if it is associated with either less cost or more benefit. Risk and reliability analysis is a four-step process: hazard identification, risk assessment, risk management, and risk communication (Figure 26). FIGURE 26 Risk and reliability analysis. Source: Singh et al. (64).

40 Hazard Identification Hazard identification depends on knowledge, experience, forecasting, engineering judgment, and imagination to ade- quately identify potential hazards. The key is to include all hazards, however unlikely they may be. Hazard identification encompasses a compilation of pos- sible failures, each with a set of parameters for consequence modeling. The types of hazards to be managed are— 1. Natural hazards (from the physical environment) 2. Technological hazards (from human-created technology) 3. Social hazards (from within human society) The events that produce hazards are characterized by their magnitude, and an event is termed hazardous when the magnitude crosses over a threat-producing threshold (64). Risk Assessment Risk assessment is a systematic, analytical method used to determine the probability of adverse effects. After the hazard-producing events have been identified and their data have been collected, the risk can be assessed. In this step, the consequence of a particular risk is evaluated with regard to its impact on the overall project objectives. Fundamental to risk modeling is an assessment of uncertainties, both quan- tifiable and unquantifiable. It is simply not possible to deal with every kind of risk. As such, only the kinds of risk that can be estimated quantitatively, at least in principle, need to be considered. Risk Management Risk management is a systematic process of making deci- sions to accept a known or assumed risk and the implemen- tation of actions to reduce the harmful consequences or probability of occurrence. As part of risk management, these risks must be assessed and a proper response for mitigating them developed. Risk mitigation is the action phase of risk management wherein the best strategy is selected. Risk Communication Risk communication is intended to communicate all of the consequences associated with a trade-off regarding the existence and nature of a threat, the seriousness of risk, and details of the steps that can be taken to mitigate its effects. For risk communication to be successful, all parties must be open to the information presented. However, if the action required to mitigate risk will cost money or require people to change habits, the information is likely to be rejected. Com- mon ways of communicating risk are through mass media, public meetings, and written material. Risk Management An agency’s management structure and project development processes, including use of design criteria, design decision making, and documentation practices, are all important aspects of risk management. Full application of the CSS/ CSD design processes discussed here supports risk manage- ment, as demonstrated in the following: • Consider Multiple Alternatives – Good risk manage- ment practices involve thorough consideration of mul- tiple alternatives, including explanation for why a full standard design may not be possible or desirable and what the alternatives are. This practice highlights the concept of design as representing discretionary choices. • Evaluate and Document Design Decisions – Design reports document the expected operational and safety performance of the proposal. Stakeholder engagement, including developing, evaluating, and discussing dif- ferent alternatives, requires documentation. All such documentation should be readily available to place in project files for later reference. Special care is to be taken when a new or creative concept is proposed, such as a diverging diamond interchange or traffic-calming feature. If a design exception is needed, documenta- tion must be complete, including a full description of the need for the exception based on adverse effects on community values, the environment, and so on. • Maintain Control Over Design Decision Making – The owning agency must stay in control of decisions regarding basic design features or elements. Active stakeholder involvement and input does not translate to abrogating the agency’s responsibility of to make fundamental design decisions. • Demonstrate a Commitment to Mitigate Safety Concerns – When a design exception or unusual solution is pro- posed, plan completion could focus on mitigation. Decisions to maintain trees along the roadside, for exam- ple, may be accompanied by special efforts to delineate the edgeline or trees, implement shoulder rumble strips, or provide guardrails or other roadside barriers. • Monitor Design Exceptions to Improve Decision Making – A few states make a special effort to keep a record of design exceptions by location, committing to review their safety performance over time. The intent is not to second-guess a decision, but to build on and improve a knowledge base for future decisions regard- ing design exceptions (4). Psychology of Risk Perception It is vital to understand how the public perceives risk in order to understand how to communicate regarding the analysis

41 and mitigation of risk. Research into the psychology of risk perceptions by U.S. psychologists Paul Slovic and Vince Covello indicates— 1. People do not demand zero risk. They consciously and subconsciously take risks every day. 2. People’s judgments of degrees of risk are not coin- cident with most methodologies for measuring risk statistically. The public may greatly underestimate familiar risks while greatly overestimating unfamil- iar risks. 3. A variety of emotional, not logical, factors control typical risk perceptions. 4. Once established, risk perceptions are extremely hard to change. New information may be absorbed by the intel- lect, but it is not readily absorbed at an emotional level. 5. Risk perceptions reside fundamentally at an emo- tional level (60). These insights suggest that the traditional approach to risk communication, which relies on providing rational, statisti- cally based information alone, may miss the mark. Because the fundamentals of risk perception are emotional and not rational, the primary focus of project risk management com- munications should be to establish trust in the organization, rather than to educate the public about the engineering fun- damentals underlying a decision. Rather than try to change the public’s risk perception, which the research states is extremely hard to do, this approach seeks to change public attitudes toward those who are being held responsible for creating and managing the risk (60). SAFETY Promoting safety and safe travel is at the core of all trans- portation planning and design, where safety can be under- stood as a measure of the freedom from unacceptable risks of personal harm. The basic principles and guidelines that influence much of what happens in project development are founded on professional principles of encouraging safe design. TEA-21 increased emphasis on safety by identify- ing safety and security as one of seven key planning factors that must be considered in statewide and metropolitan plan- ning processes, which was continued with the next authori- zation bill, SAFETEA-LU. As such, special attention must be paid to trade-offs that may influence the potential of an alternative. Several key areas provide insight into how to address these trade-offs: the field of organizational acci- dent analysis and prediction, safety-conscious planning, the interactive highway safety design manual, and the new Highway Safety Manual. Organizational Accidents The field of organizational accident analysis and prediction cen- ters on the identification and mitigation of risk. A paramount consideration when dealing with the evaluation of trade-offs in roadway design is the preservation of safety in the resultant design. As such, some background on the theory of accidents is appropriate to understand how they occur, what must be done to prevent them, and how an organization must react to create an environment that promotes safety-conscious decisions. Theory The pioneering work done by Reason (65) serves as the underlying basis for much of the modern work dealing with organizational accidents. Reason’s paper “The Contribu- tion of Latent Human Failures to the Breakdown of Com- plex Systems” followed on the heels of several high-profile disasters in a wide range of endeavors: nuclear power plants, chemical installations, spacecraft, roll-on-roll-off ferries, aircraft, offshore oil platforms, and railway networks. Rea- son surmised that even though these disasters may appear unrelated, they share several important factors: 1. The accidents occurred in a complex environment that possessed elaborate safety devices and protections. 2. The accidents were caused not by a single failure, but by a conjunction of several diverse sequences, each necessary but not sufficient to cause the event by itself. 3. Human failures, not technical failures, were typically the root cause. Weick (66) expanded on this idea of linked events, pos- tulating that to anticipate and forestall disasters is to under- stand regularities in the ways small events can combine to have disproportionately large effects. The central question of accident investigation is how the defenses were breached (Figure 27). Reason (67) identifies three sets of factors that are responsible: human, technological, and organizational. FIGURE 27 The relationship between hazards, defenses, and losses. Source: Reason (67). Reason (65) further observed that human operators are increasingly remote from the processes that they nominally govern, and that most of the time these operators are sim-

42 ply monitoring the system to ensure that it functions within acceptable limits. By comparison, as much of the underly- ing rationale for design standards is removed from design manuals, designers who do not understand the fundamentals underlying the criteria are acting in a similar manner. Reason (67) identified two primary ways in which humans caused breakdowns in complex systems: active and latent failures. Active failures involve errors or violations that have an immediate effect, whereas latent failures are tied to deci- sions or actions that later result in a breach of the system’s defenses. Figure 28 depicts how both active and latent failures can combine to “holes” in even a layered defense. When the correct sequence of events lines up with these holes, an acci- dent trajectory can pierce the defenses, resulting in an acci- dent. This is often referred to as the “Swiss cheese” model. FIGURE 28 Accident trajectory passing through corresponding holes in layers of defenses, barriers, and safeguards. Source: Reason (67). Application to Road Design Salmon and Lenne (68) have directly applied the theory of organizational accidents to the field of traffic safety and road- way design. They found that Australia has been using a “sys- tems approach” to safety over the past two decades, resulting in significant safety gains. Under this approach, safety is a product of the overall system. They point to the Australian National Road Safety Strategy 2001–2021, the Swedish Vision Zero, and the Netherlands’ Sustainable Safety approaches as evidence that this approach is gaining popularity. These pro- grams advocate a shared responsibility for safety, an appre- ciation of the limits of human performance and tolerance, and a forgiving road transport system. They stress that in a road safety context, elements of the system beyond road users, such as vehicle design and condition, road design and condi- tion, and road policies, all shape driver behavior. Reason’s systems perspective model of human error and accident causation, more commonly known as the “Swiss cheese” model, is used as a basis to examine the relationship of problems with the transportation network and safety. Rea- son’s model holds that weaknesses in the system’s defenses, created by inappropriate or inadequate decisions and actions by actors at all levels of the system, allow accident trajec- tories to breach defenses and cause accidents. The Human Factors Analysis and Classification System (HFACS) was developed for the aviation safety domain, but has been suc- cessfully applied to other safety domains as well. It consid- ers both the errors at the “sharp end” of the system operation and also the latent conditions involved in a particular inci- dent or accident (Figure 29). HFACS identifies four layers of defenses onto which the active and latent failures are mapped: unsafe acts, preconditions for unsafe acts, unsafe supervision, and organizational influences. Many factors contribute to the latent and active failures in each layer of the defense. By mapping the data typically available for traffic crashes onto this model, it can be seen that the data available cover mainly the road user and environmental, equipment, and context (e.g., time of day) factors (Figure 30). Data cov- ering higher level latent errors, such as poor roadway design, poor maintenance, or poor operations, are not contained in a typical crash record, so a complete picture of the accident trajectory is not available with current traffic safety data. As such, to link these factors to a particular crash, additional data mining is necessary. Organization Types Reason (67) identifies three types of organizations in rela- tion to how they approach the idea of reforms to their pro- cesses and procedures with respect to safety: • Pathological organizations – Use inadequate safety measures. • Calculative organizations – Take a “by-the-book” approach to safety. • Generative organizations – Set safety targets beyond ordinary expectations and are willing to use unconven- tional means to achieve them. Westrum (69) emphasized that for an organization to avoid losses, it must have what he termed “requisite imagi- nation.” This is the diversity of thinking and imagining that is required to identify possible failure scenarios. This req- uisite imagination is typically present only in generative organizations, and its absence can leave blind spots in an organization’s defenses that can result in disaster. Until the last 20 years, most STAs would fall into the cat- egory of calculative organizations. However, with the advent of CSS/CSD, a gradual transition to generative organizations has begun. Sweden’s Vision Zero, begun in 1997, aims to achieve a highway system with zero fatalities or serious inju- ries and is an example of a generative organization goal that requires an agency to revamp its approach to engineering.

43 Overview One of the key takeaways from the work done in the orga- nizational accident area is that many disasters occur when a combination of active and latent failures align to create an accident. When dealing with trade-offs in roadway design, the designer and decision makers must be careful to step back and look at the trade-off decisions holistically once they are made, as individual trade-off decisions that may be perfectly safe can in combination create conditions that com- promise safety. Reason (67) identifies that a change in one system parameter must be compensated for by changes in other parameters. The current interest in three-dimensional design verification is an example of this holistic evaluation. FIGURE 29 HFACS mapped onto Reason’s Swiss cheese model. Source: Salmon and Lenne (68). FIGURE 30 Fatal road traffic accident data mapped onto Reason’s Swiss Cheese Accident Causation Model. Source: Salmon and Lenne (68).

44 Safety-Conscious Planning Although safety lies at the core of all transportation planning, the utility and role of safety measures in the planning process has been difficult to manifest in subsequent analysis, evalua- tion, prioritization, and system performance monitoring. The concept of incorporating safety issues into the transportation planning process in a more comprehensive way arose around the same time that CSS/CSD began to gain increasing consid- eration. The concept of safety-conscious planning integrates safety into all aspects of transportation planning, including setting the policy and planning context for eventual project development. Safety-conscious planning is comprehensive in the sense that it considers all aspects of transportation safety—not only infrastructure-related improvements but also enforcement and education strategies as well as enhanc- ing emergency service response to incidents (70). Substantive safety effects are currently described in two primary ways—safety performance functions (SPFs) and accident modification factors, now called crash modification factors (CMFs). • SPFs describe the expected crash frequency for a con- dition or element as a function of traffic volume and other fundamental values. SPFs are usually expressed as an equation or mathematical function. The follow- ing is an example of a SPF for a roadway segment on a rural two-lane highway: NSPF rs = (AADT) x (L) x (365) x 10(-6) x e(-0.4865) Where: NSPF rs = estimate of predicted average crash frequency for SPF base conditions for a rural two-lane two-way road- way segment (crashes/year); AADT = average annual daily traffic volume (vehicles/ day) on roadway segment; and L = length of roadway segment (miles) (71). • CMFs describe the expected change in crash frequency (total or particular crash type) associated with an incre- mental change in design dimension. CMFs may be shown in tabular form or in some cases as a simple func- tion. They are expressed as a decimal, with a CMF less than 1.0 indicating a lower crash frequency and a CMF greater than 1.0 indicating a higher frequency (23). The availability of new tools to quantify safety effects opens up the possibility of moving to performance-based design. Although the conventional approach provides design consistency and uniformity, code-based design and applica- tions may decrease in importance, yielding to performance- based design. Under performance-based design, designers will apply analytical procedures to quantify or estimate the possible trade-offs and to consider changes and variations in physical dimensions (72). Road Safety Audits Road safety audits (RSAs) are formal procedures for assess- ing the accident potential and safety performance of new and existing highways and streets by an independent audit team. This team considers the safety of all road users (e.g., drivers, pedestrians, bicyclists, elderly, children), considers all envi- ronmental conditions (e.g., day, night, inclement weather), and qualitatively estimates and reports on road safety issues and opportunities for improvement with the existing facility or proposed design (73). The objective of the RSA is to ensure that all new highways operate as safely as practicable. This objective means that safety needs to be considered throughout the design process by bringing an improved understanding of crash causes and countermeasures to bear in a proactive manner. RSAs can help produce designs that limit the num- ber and severity of crashes, promote awareness of safe design practices, and identify and correct safety issues before proj- ects are built (74). Experience with RSAs in the United States shows that the RSA team often uncovers safety concerns that a traditional safety review would have missed. An RSA and a safety review are two different processes (see Table 13). Some key differences are that a, RSA is con- ducted by an independent, interdisciplinary team and gener- ates a formal response report (75). TABLE 13 DIFFERENCES BETWEEN ROAD SAFETY AUDIT AND TRADITIONAL SAFETY REVIEW Source: FHWA Road Safety Audit Guidelines (75). FHWA has developed the FHWA Road Safety Audit Guidelines to assist agencies in developing policies and pro-

45 cedures for conducting RSAs. The guide outlines the typi- cal steps associated with undertaking an RSA (see Figure 31). To support this approach, FHWA has developed an RSA training course through the National Highway Insti- tute (FHWA-NHI-380069 Road Safety Audits/Assessments), which can be accessed at http://www.nhi.fhwa.dot.gov. FIGURE 31 Typical RSA steps. Source: FHWA Road Safety Audit Guidelines (75). All RSAs must have the following key elements: • Formal Examination – Design components and associ- ated operational effects are formally examined from a safety perspective. • Team Review – RSAs conducted are by teams of at least three auditors. • Independent RSA Team – Audit team members are independent of the design team. • Qualified Team – Auditors have appropriate qualifi- cations. An understanding of the AASHTO Roadside Design Guide, positive guidance techniques, access management, and crash analysis is strongly suggested (76). • Focus on Road Safety – The principal focus of an RSA is to identify potential safety issues, not to serve as a compliance review. • Includes All Road Users – All appropriate vehicle types, modes of transportation, and road users are to be included in the RSA. • Proactive Nature – RSAs need to consider all potential safety issues, not only those that may be demonstrated by a crash history. • Qualitative Nature – Outputs are qualitative, not quan- titative, in nature. Outputs include lists of identified issues, assessments of relative risk, and suggested cor- rective measures. • Field Reviews – Day and night field reviews are con- ducted (75). During the planning and preliminary engineering phases of a project, an RSA team can review all of the options being considered and make recommendations such as changes in horizontal and vertical alignment, provision of a median, land and shoulder width, provision of bike lanes and sidewalks, and provision of channelization (75). These recommendations can be used to improve alternatives, and the safety issues identified can be used to compare alternatives in a qualitative manner. Prompt lists help auditors identify problem safety issues during an RSA and have been developed for the different stages of a project by FHWA. Figure 32 is an example of the prompt list provided for the planning stage audit. These lists help the RSA team identify potential safety issues and ensure that nothing is overlooked in the audit. However, care must be taken in the application of prompt lists to ensure that the RSA does not become a mechanistic exercise of check- ing the boxes instead of helping the auditors apply their knowledge and experience (67). There is a range in the format and approach used in developing prompt lists by agencies most experienced with RSAs. The Austroads RSA Guide implements a comprehen- sive approach, detailing every consideration at each stage of the process, including feasibility, preliminary design, final design, pre-opening, roadwork traffic schemes, existing roads, and land use development proposals. At the other end of the spectrum is the broader approach found in the Cana- dian RSA Guide. This approach utilizes simple prompt lists meant to challenge the user to think about various issues, such as geometric design, traffic operations, control devices, human factors, environmental and integration, that could be found at all stages of project development. The FHWA prompt lists lean toward the broader approach (77, 78). Additionally, FHWA has developed software to aid in the performance of RSAs. This software is intended to aid the auditor by providing guidance, providing a means for tracking data, generating prompt lists at appropriate levels of detail, and aiding in the preparation of the final RSA report.

46 Interactive Highway Safety Design Model The IHSDM is a suite of software analysis tools for evaluat- ing the safety and operational effects of geometric design decisions. Under its current configuration, it can assist designers in evaluating design alternative for two-lane rural highways against relevant design policy values and provide estimates of a design’s expected safety and operational per- formance. The IHSDM—HSM Predictive Method 2010 Release Crash Prediction Module includes additional capa- bilities to evaluate rural two-lane highways, rural multilane highways, and urban/suburban arterials. The IHSDM cur- rently includes five evaluation modules: crash prediction, design consistency, intersection review, policy review, and traffic analysis. • Crash Prediction Module. Estimates the frequency and severity of crashes that can be expected on a high- way based upon its geometric design and traffic char- acteristics. This module can help identify potential improvement projects on existing roadways, compare the relative safety performance of design alternatives, and assess the safety and cost-effectiveness of design decisions. • Design Consistency Module. Diagnoses safety con- cerns at horizontal curves by providing estimates of the magnitude of potential speed differentials. This module provides a diagnostic tool for existing align- ments and serves as a quality assurance check on new design alternatives. • Intersection Review Module. Evaluates an existing or proposed intersection geometric design to identify potential safety concerns and suggest possible treat- ments to mitigate those concerns. • Policy Review Module. Checks design elements for compliance with geometric design criteria. This module provides a diagnostic tool for existing alignments and serves as a “nominal safety” audit for proposed designs. • Traffic Analysis Module. Estimates traffic quality-of- service measures for an existing or proposed design under current or projected traffic. The IHSDM software may be downloaded free of charge through the IHSDM public software website at http://www. ihsdm.org/ (79). Highway Safety Manual The HSM is a resource that provides tools and safety knowledge in a format to help decision making regarding highway design based on safety performance. The HSM provides methodolo- gies that can be utilized to analyze trade-offs across a broad range of design activities, including planning, programming, project development, construction, and operations and mainte- nance. Before the release of the HSM, practitioners did not have a single national resource for reliable quantitative information about crash analysis and evaluation. Although the HSM is not currently applicable to all types of facilities, it does represent the state of the art for evaluating safety trade-offs and incorpo- rating them into the decision-making process in a technically sound and consistent manner. However, to gain full predictive benefit of the HSM procedures, the base models need to be cali- brated to local conditions every 2 to 3 years. The HSM does not represent a legal standard of care, but is meant to provide guidance for evaluating safety trade-offs (71). Figure 33 illus- trates the organization of the HSM and how each of the sections relates to the others. Figure 34 illustrates how planning, design, construction, and maintenance activities relate to the HSM. FIGURE 32 Planning stage audit prompt list. Source: FHWA Road Safety Audit Guidelines (75).

47 FIGURE 33 Organization of the Highway Safety Manual. Source: Highway Safety Manual (71). FIGURE 34 Relating the project development process to the HSM. Source: Highway Safety Manual (71). The HSM has developed SPFs in Volume 2 for three facil- ity types and for specific site types of each facility, as sum- marized in Table 14. In addition, a software package called HiSafe has been developed to support the predictive meth- ods in this section of the HSM. TABLE 14 FACILITY TYPES AND SITE TYPES INCLUDED IN PART C OF HSM Facility Type Undivided Roadway Segments Divided Roadway Segments Intersections Stop Control on Minor Leg(s) Signalized 3-Leg 4-Leg 3-Leg 4-Leg Rural Two- Lane Roads     Rural Mul- tilane Highways      Urban and Suburban Arterial Highways       Source: Highway Safety Manual (71). The HSM has also developed CMFs for different counter- measures, which are found in Volume 3. They are organized by roadway segments, intersections, interchanges, special facilities and geometric situations, and road networks. The SPFs and CMFs developed in the HSM use the same base conditions and therefore are compatible. SafetyAnalyst SafetyAnalyst provides a set of software tools to help design- ers perform site-specific highway safety analyses, including many of the procedures that are presented in HSM Part B. It allows the user to evaluate trade-offs in a decision-making process using several screening tools. More information on SafetyAnalyst can be found on its website at http://www. safetyanalyst.org. SafetyAnalyst can not only identify accident patterns at specific locations and determine whether those accident types are overrepresented, but also determine the frequency and per- centage of particular accident types systemwide or for specified portions of the system. This capability can be used to investi- gate the need for systemwide engineering improvements (e.g., shoulder rumble strips on freeways) and for enforcement and public education efforts that may be effective in situations in which engineering countermeasures are not. SafetyAnalyst provides six tools to assist designers in trade-off analysis for site-specific alternatives.

48 • Network Screening Tool. Identify sites with potential for safety improvements. • Diagnosis Tool. Diagnose the nature of safety prob- lems at specific sites. • Countermeasure Selection Tool. Select countermeasures to reduce accident frequency and severity at specific sites. • Economic Appraisal Tool. Conduct an economic appraisal of a specific countermeasure or several alter- native countermeasures for a specific site. • Priority Ranking Tool. Rank sites and proposed improvement projects based on the benefit and cost estimates determined by the economic appraisal tool. • Countermeasure Evaluation Tool. Conduct before/ after evaluations of implemented safety improvements (80).