Comprehensive Human Factors Guidelines for Road Systems (Web-Only Document) (2005)

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.

APPENDIX C HUMAN FACTORS GUIDELINES FOR ROAD SYSTEMS: DESIGN AND OPERATIONAL CONSIDERATIONS FOR THE ROAD USER DRAFT EXAMPLE CHAPTER CHAPTER 5 From Driver Reaction Time, Maneuver Time, and Speed to Design Distances: General Guidelines C-1

TABLE OF CONTENTS 5. FROM DRIVER REACTION TIME, MANEUVER TIME, AND SPEED TO DESIGN DISTANCES: GENERAL GUIDELINES ........................................................................ C-4 5.1 BACKGROUND ............................................................................................................. C-6 5.1.1 Organization of This Chapter.............................................................................. C-6 5.1.1.1 Structure of the Chapter................................................................................... C-6 5.1.1.2 Relationship to Other Chapters........................................................................ C-6 5.1.2 The Human Factors Basis of Sight Distance Design Requirements.................... C-7 5.1.2.1 Objective of This Chapter................................................................................ C-7 5.1.2.2 Component Determinants of Sight Distance Requirements ............................ C-8 5.1.2.3 Real-World Driver Behavior Versus Design Models...................................... C-9 5.1.3 The Relationship of This Chapter to Other Key Reference Documents ............ C-15 5.2 DESIGN SIGHT DISTANCES ........................................................................................ C-16 5.2.1 Stopping Sight Distance .................................................................................... C-17 5.2.1.1 Definition: SSD [AASHTO 2001 Ch 3]....................................................... C-17 5.2.1.2 High Priority Considerations: SSD................................................................ C-18 5.2.1.3 Guideline for SSD ......................................................................................... C-18 5.2.1.4 Basis/Rationale for SSD Guideline ............................................................... C-20 5.2.2 Intersection Sight Distance................................................................................ C-23 5.2.2.1 Definition: ISD .............................................................................................. C-23 5.2.2.2 High Priority Considerations: ISD PRT ........................................................ C-24 5.2.2.3 Guideline for ISD .......................................................................................... C-25 5.2.2.4 Basis/Rationale ISD PRT, MT and Critical Gap ........................................... C-29 5.2.3 Decision Sight Distance .................................................................................... C-33 5.2.3.1 Definition DSD.............................................................................................. C-33 5.2.3.2 High Priority Considerations ......................................................................... C-34 5.2.3.3 Guideline for DSD MANUEVERS C, D & E............................................... C-34 5.2.3.4 Basis/Rationale for DSD Avoidance Maneuvers C, D & E Guideline.......... C-37 5.2.4 Passing Sight Distance ...................................................................................... C-41 5.2.4.1 Definition....................................................................................................... C-41 5.2.4.2 Related Design/Operational Issue ................................................................. C-42 5.2.4.3 Guideline ....................................................................................................... C-42 5.2.4.4 Basis/Rationale for Guideline........................................................................ C-43 5.3 INFLUENCE OF DESIGN ON SPEED.............................................................................. C-47 5.3.1 Background........................................................................................................ C-47 5.3.2 Scope ................................................................................................................. C-47 5.3.3 Speed and Lane Width ....................................................................................... C-47 5.3.3.1 Guideline: Speed and Lane Width................................................................. C-47 5.3.3.2 Basis/Rationale for Speed and Lane Width ................................................... C-47 5.3.4 Speed and Alignment ......................................................................................... C-48 5.3.4.1 Guideline: Speed and Alignment................................................................... C-48 5.3.4.2 Basis/Rationale for Speed and Alignment..................................................... C-48 5.3.5 Speed and Pavement Surface............................................................................. C-49 5.3.5.1 Guideline: Speed and Pavement Surface....................................................... C-49 5.3.5.2 Basis/Rationale for Speed and Pavement Surface ......................................... C-49 5.3.6 Speed and Side Friction..................................................................................... C-49 5.3.6.1 Guideline: Speed and Side Friction ............................................................... C-49 5.3.6.2 Basis/Rationale for Speed and Side Friction ................................................. C-49 C-2

5.4 DIAGNOSING SIGHT DISTANCE PROBLEMS ............................................................... C-50 5.4.1 The Six-step Process.......................................................................................... C-50 5.5 REFERENCES .............................................................................................................. C-67 ATTACHMENT A: EXAMPLE APPLICATION: SIGHT DISTANCE DIAGNOSTIC PROCEDURE .................................................................................................................. AT-1 C-3

5. From Driver Reaction Time, Maneuver Time, and Speed to Design Distances: General Guidelines This section of the document, preceding Section 5.1, provides an introduction to a draft sample chapter of the Human Factors Guidelines (“HFG”). It provides some information on what the sample chapter is intended to do and on how this document differs in some respects from what might be seen in a typical HFG chapter. This introductory section would not be present in an actual chapter. The actual sample chapter text begins with Section 5.1. This chapter addresses the human factors basis of sight distance requirements. It is written as a model chapter of a planned document tentatively titled Human Factors Guidelines for Road Systems: Design and Operational Considerations for the Road User (“HFG”). Table 1 presents a tentative high-level outline of the HFG, developed earlier in this project (Lerner, Llaneras, Hanscom, Smiley, Neuman, and Antonucci, 2002). As the table indicates, the HFG is comprised of four Parts. Part I is introductory. Part II presents basic human factors concepts and the user-centered design approach. It provides information and data on basic driver capabilities. The concepts, principles, and data in Part II will be related to many of the subsequent guideline statements in the HFG. Parts III and IV are the guidelines sections. Part III provides human factors guidance for roadway location elements, such as curves, transition zones, or intersections. Part IV provides human factors guidance for traffic engineering elements, such as signs, markings, and lighting. The present document corresponds to Chapter 5 of the tentative outline, which is in Part III. This document has two general purposes: • It is intended to serve as a model for subsequent chapters of the HFG. • It is intended to serve as a stand-alone document that provides guidance to the highway designer and traffic engineer regarding human factors considerations for sight distance. These objectives are somewhat incompatible, in that the HFG is viewed as an electronic document consisting of a highly integrated collection of interrelated chapters, with extensive cross-linking (Lerner et al., 2002). Development of the HFG is planned as a long-term incremental effort involving multiple authors, with input and review from a wide range of technical experts and stakeholders. Chapters in some sections of the HFG (Parts III and IV) will be highly focused on explicit guidance statements. Other chapters (Part II) provide general principles, explain basic concepts, or describe basic human factors data and procedures for highway safety. Because of the extensive linking, individual chapters of the HFG are not viewed as “stand alone” documents; important information will be contained in other sections and the reader can jump to those when desired. So, for example, guidance chapters will be relatively streamlined and focused on guidance statements, with discussion of background information, underlying concepts, supporting findings, and so forth, provided elsewhere, with the links highlighted. Or, some guidance relevant to signing for roadway curves might be located in the chapter on signs, with links in the appropriate places in the chapter on curves. The present document is intended to show what chapters may look like, yet also function as a useful, stand-alone document. For this reason, it has somewhat more background than is likely to be typical in actual guidance chapters (Parts III and IV) of the planned HFG. At the same time, it has some simulated “links” that point to planned chapters that will have more detail. These simulated links are shown in bold brackets [Section x.x]. In the actual HFG, there would only be underlining to indicate a hypertext link. There also are links to other key reference documents. C-4

Table 1. Proposed Parts and Chapters for the HFG (from Lerner et al., 2002) PART I: INTRODUCTION TO THE HFG Chapter 1. Why Have Human Factors Guidelines for Road Systems? Chapter 2. How to Use This Document PART II: BRINGING ROAD USER CAPABILITIES INTO HIGHWAY DESIGN AND TRAFFIC ENGINEERING PRACTICE Chapter 3. A System Approach to Highway Safety: Thinking Like a Road User Chapter 4. Basic Driver Capabilities PART III: HUMAN FACTORS GUIDANCE FOR ROADWAY LOCATION ELEMENTS Chapter 5. From Driver Reaction Time, Maneuver Time, and Speed to Design Distances: General Guidelines Chapter 6. Curves (Horizontal Alignment) Chapter 7. Grades (Vertical Alignment) Chapter 8. Tangent Sections and Roadside (Cross Section) Chapter 9. Transition Zones Between Varying Road Designs Chapter 10. Non-Signalized Intersections Chapter 11. Signalized Intersections Chapter 12. Interchanges Chapter 13. Construction and Work Zones Chapter 14. Rail-Highway Grade Crossings Chapter 15. Special Considerations for Urban Environments Chapter 16. Special Considerations for Rural Environments PART IV: HUMAN FACTORS GUIDANCE FOR TRAFFIC ENGINEERING ELEMENTS Chapter 17. Signing Chapter 18. Variable Message Signs Chapter 19. Markings Chapter 20. Lighting The topic of this chapter is also not entirely representative as a model for other guidance chapters. It was selected in part because it can function relatively well as a stand-alone document and because of current interest in the topic. However, it is in some ways unique as a chapter within the HFG. It is seen as the first “guidelines” chapter in the HFG. It differs from subsequent guidance chapters in that it is not specific to a particular roadway location element (e.g., curves) or traffic engineering element (e.g., signing). The guidance principles are therefore at a somewhat more general level than in subsequent chapters. Sight distance issues that are specific to a particular roadway location element or traffic engineering element will be treated in more detail in the appropriate chapters. Some differences among applications are dealt with here, but within this chapter the emphasis is on principles that are relevant to many design conditions. In this sense, Chapter 5 is something of a bridge chapter between Part II of the HFG, which provides basic human factors concepts, findings, and approaches, and Parts III and IV, which provide specific guidance statements for particular applications. C-5

Based on the preliminary outline of the planned HFG, this chapter on sight distance is labeled Chapter 5. Chapter numbering is likely to change as the actual HFG evolves. However, we retained the chapter number for this sample chapter to help place it in context and to use the tentative outline for purposes of the simulated “links.” 5.1 Background 5.1.1 Organization of This Chapter 5.1.1.1 STRUCTURE OF THE CHAPTER This chapter consists of five subsections. Section 5.1 is a background section that explains the scope and content of the chapter, describes fundamental sight distance concepts and their relation to human factors considerations, considers the features and limitations of sight distance design equations as models of driver behavior, and relates the chapter to other key reference sources. Sections 5.2 and 5.3 provide explicit guideline statements. Section 5.2 is organized around four primary sight distance applications: stopping sight distance, intersection sight distance, decision sight distance, and passing sight distance. For each of these applications there is a section on definition, a set of high- priority considerations, stated guidelines, and a discussion of the basis and rationale for the guideline. The primary focus of Section 5.2 is with the factors that influence perception-reaction time and maneuver time and distance. Section 5.3 deals with the influence of design on speed, as speed relates to sight distance requirements. Section 5.4 provides a systematic approach for diagnosing and addressing sight distance problems. It is a diagnostic tool that comprises a series of analytic steps. While Sections 5.2 and 5.3 provide guidance in the form of guidelines statements, Section 5.4 is a more procedural way of addressing many of the issues raised in Sections 5.2 and 5.3. Section 5.5 provides full reference citations for sources cited in the chapter. 5.1.1.2 RELATIONSHIP TO OTHER CHAPTERS While written as a stand-alone document, this document is also intended to be viewed as a chapter within the HFG. It has an important relationship to other sections of the HFG. Sight distance will be an important consideration for a number of roadway location elements that are the subjects of specific chapters in Part III of the HFG (see Table 1). We anticipate those chapters will link to the relevant portions of Chapter 5 where general sight distance considerations arise. We also anticipate that these chapters will have more application-specific guidance that goes beyond the general guidance of Chapter 5. The guidelines in Section 5.2 and 5.3 are somewhat limited in specificity because they are meant to be broadly applicable to many applications. Guidance in subsequent chapters of Part III will be narrower and more specific, or link back to Chapter 5. Likewise, Part IV chapters will link to Chapter 5 but will have more specific treatments of the relationship of sight distance issues to their topics. Speed control is a very fundamental aspect of highway design and traffic engineering and has many related human factors issues. Because vehicle speed is a key element of sight distance requirements, this chapter addresses driver speed selection in this context, most specifically in Section 5.3. Speed control will be treated in some detail for the individual roadway location elements of other Part III chapters (e.g., C-6

curves, transition zones, work zones). However, in dealing with driver speed selection in this chapter, it has become evident that there is a need for another cross-cutting chapter that deals with speed in much more detail. Specifically it needs to consider the human factors aspects of speed choice, speed perception, and speed control. No such chapter was proposed in the outline of the HFG (because of treatment within specific Part III chapters and discussion in Chapter 4 of Part II). However, this now appears to be an oversight. In developing this chapter, we saw the need for linking to much more detailed treatment of speed than is appropriate within this chapter. We suggest that a future chapter on driver speed selection and speed management be added to the HFG. This chapter also relates to the fundamental human factors concepts and driver attributes presented in Part II of the HFG. Issues such as perception-reaction time, driver expectancy, and driver attributes are essential considerations for understanding human factors concerns in sight distance. In the final version of the HFG, it may even be appropriate to move some of the material from this chapter to Part II and treat it through links. However, in order to make this chapter useful as a stand alone document, it contains enough basic human factors considerations to make it self-contained. 5.1.2 The Human Factors Basis of Sight Distance Design Requirements 5.1.2.1 OBJECTIVE OF THIS CHAPTER This chapter describes human factors considerations that influence sight distance requirements. Sight distance is the length of roadway ahead that is continuously visible to the driver. It is a central concept in roadway design because the driver must have enough preview of the roadway to safely accomplish various driving maneuvers. Different maneuvers – emergency braking, passing, making a left turn at an intersection, etc. – each have their own sight distance design requirements. Later sections of this chapter address sight distance for different maneuvers, including stopping sight distance [Section 5.2.1], intersection sight distance [Section 5.2.2], decision sight distance [Section 5.2.3], and passing sight distance [Section 5.2.4]. Although these design requirements are expressed as a design distance, from the driver’s perspective the critical aspect is time. It takes time to recognize a situation, understand its implications, decide on a reaction, and initiate the maneuver. While this process may seem almost instantaneous to us when driving, it can translate into hundreds of feet at highway speeds before a maneuver is even initiated. Speed is the factor that transforms road user time needs into distance requirements. Speed also can directly influence the requirements of the maneuver itself. This chapter addresses those human factors considerations that influence time and speed, and hence sight distance. Although the roadway designer and traffic engineer work with distances, sight distance requirements actually stem from driver time needs and speed choice. Therefore to understand, diagnose, and address sight distance concerns, one must address the human factors issues of time and speed. Sight distance is an important concern for many specific roadway location elements (e.g., curves) and traffic engineering elements (e.g., signs) addressed in the subsequent chapters of this guide. Those chapters provide specific guidance dealing directly with those elements. This chapter addresses human factors issues at a general level that is relevant to many sight distance applications. This chapter is intended to assist in situations where AASHTO sight distance standards [AASHTO 2001 Chapter 3] may be difficult to meet or may be less than optimal. Trade-offs among competing requirements sometimes require compromise decisions; in some cases, time requirements, based on measured driver behavior, may be less than those required by AASHTO. In other situations, conditions may make it desirable to meet or go beyond the standard requirement. The information and guidance provided in this chapter is intended to assist the engineer by recognizing the factors contributing to the behavioral components of the sight distance requirement. C-7

5.1.2.2 COMPONENT DETERMINANTS OF SIGHT DISTANCE REQUIREMENTS There are several major components that together determine the sight distance that the driver requires to safely execute a maneuver. Perception-reaction time (PRT). [Section 4.5] Before a driver can execute a maneuver, he or she must recognize there is a need for some action and decide what that action should be. Therefore this mental activity – detection, perception and cognition – precedes an overt vehicle control action and takes some amount of time. Perception-reaction time (sometimes also termed perception-response time) is typically defined as the period from the time the object or condition requiring a response becomes visible in the driver’s field to view to the moment of initiation of the vehicle maneuver (e.g., first contact with the brake pedal). Although a particular PRT value (e.g., 2.5 s) is used in deriving sight distance requirements for a given design situation, this “reaction time” value should not be viewed as a fixed human attribute. PRT can take on a wide range of values depending upon many factors. Section 5.2 deals with these in more detail. PRT is sometimes discussed as a sequence of stages. An example is the PIEV model (for Perception- Identification-Emotion-Volition), which is useful for illustration since it is cited in the MUTCD [http://mutcd.fhwa.dot.gov/pdfs/2003r1/Ch2C.pdf]. This model conceives of PRT as the sum of four stages: Perception (becoming aware of the presence of the object or event), Identification (understanding the object/event and its implications), Emotion (deciding what action to take), and Volition (translating the decision into overt action). Models of this sort are useful for pointing out the various perceptual and cognitive activities that must occur for a successful driver reaction. However, they over-simplify the process and the linear sequence of events is simply not a very complete or accurate description of driver cognitive activity. Section 5.1.2.3 provides further discussion of how the driver behavior assumptions of sight distance models differ from real-world driver behavior. Maneuver time (MT) and maneuver distance. Maneuver time is the interval from the initiation of the vehicle control response (i.e., end of the PRT) to the completion of the driving maneuver. “Completion” is variously defined for different maneuvers, such as braking, turning, or passing. Section 5.2 considers these various maneuvers and what influences the time it takes to complete them. The amount of distance needed for the safe and comfortable completion of the maneuver is dependent upon MT but also to other maneuver requirements. Maneuver distance (e.g., braking distance) is directly related to the physics of the situation (e.g., tire-pavement friction, grade), including vehicle performance capabilities. Maneuver time is also related to individual driver characteristics. Speed selection. [Section 4.6; Chapter on Speed Control] Vehicle speed is what translates time requirements into distance needs. The operating speed determines the distance traversed while the PRT and MT are happening. Depending on the sight distance situation under consideration, the relevant speed may be that of the driver’s own vehicle (as in stopping sight distance) or the speed of the approaching vehicle (as in stop-controlled intersection sight distance) or both (as in passing sight distance). In addition to providing the multiplier that converts PRT and MT to distances, speed can also affect distance requirements in several other ways. Maneuvers, such as braking, may require greater distances at higher speeds. Further, under some conditions speed can directly influence PRT, by altering how and where drivers allocate their attention. [Section 4.2] Finally, speed can influence the options the driver has and the difficulty and urgency of the decision. For sight distance considerations, then, an important question is, what factors influence a driver’s choice of speed? In considering this, it is important to consider not only conscious driver decisions, but also perceptual factors that might influence a person’s perception of speed [Section 4.6]. Section 5.3 deals with the influence of design factors on driver speed selection as related to sight distance. C-8

In summary, sight distance requirements are jointly determined by PRT, MT, and speed selection. All of these are sensitive to a range of human factors considerations. Sight distance equations are based on simplified assumptions about road user abilities and behavior; if these assumptions are inappropriate for a given application, the actual driver behavior may not match the predicted behavior and road design may not be adequate. ***************************************************** Figure 5.1 Animation showing the influence of human factors variables on stopping distance Animation: show two parallel animations, from driver’s eye view and diagram view. In both cases, child pedestrian enters road 300 feet ahead of vehicle on a 40 mph road. Run the animations in slow-motion, indicating point of detection, recognition, brake initiation, stopping. Left side: Expected Right side: Unexpected High contrast Low contrast Simple background Complex background Simple Tangent Geometric feature Alert driver Distracted No other traffic Vehicle in next lane Traveling at speed limit 10 mph over limit **************************************************** 5.1.2.3 REAL-WORLD DRIVER BEHAVIOR VERSUS DESIGN MODELS This section discusses how driver behavior, as represented in sight distance models, may differ from actual driver behavior. Design models use simplified concepts of how the driver thinks and acts. This simplification should not be viewed as a flaw or error in the sight distance equations. These models are a very effective way of bringing human factors data into design equations in a manner that makes them accessible and usable. After all, the intent of a sight distance equation is not to reflect the complexities of human behavior but to bring what we know about it into highway design in a practical way. However, like any behavioral model, models for deriving sight distance requirements are not precise predictors of every case and there may be some limitations to their generality. Therefore it is useful to understand certain basic principles of human behavior in driving situations to better interpret these models and how they may differ from the range of real-world driving situations. Driver’s-eye view of road Bird’s-eye view of road, including vehicle, child Bird’s-eye view of road, including vehicle, child Driver’s-eye view of road C-9

Sight distance formulas for various maneuvers (presented in Section 5.2) differ from one another, but they share a common simple behavioral model as part of the process. The model assumes that there is some time required for perception and reaction (PRT), followed by some time (MT) and/or distance required for executing the maneuver, and some vehicle speed in effect during these times. Sight distance equations for some maneuvers may contain additional elements or assumptions, however, all have this basic two- stage model somewhere at their core. The two equations below show two versions of the general two-stage model. In both cases, the first term shows the distance traveled during the PRT phase and the second term shows distance traveled while executing the maneuver. The difference is that the first equation shows a case where the distance traveled while executing the maneuver is based on the time it takes to make that maneuver (for example, the time to cross an intersection from a Stop). The second equation shows a case where the distance traveled while executing the maneuver is based directly on the distance required to complete the maneuver (for example, braking distance for a emergency stop). For both forms of this general equation, vehicle speed (v) influences the second (maneuver) component. The general form of the sight distance equation is: d = kVtprt + kVtman, where maneuver time is input or d = kVtprt + dmanV, where maneuver distance is input where: d = required sight distance V = velocity of the vehicle(s) tprt = PRT tman = MT dmanV = distance required to execute a maneuver at velocity V k = a constant to convert the solution to the desired units (feet, meters) This model shows that the sight distance requirement is composed of (at least) two distances: there is a distance traveled while the driver perceives and evaluates a situation (determined by PRT and vehicle speed) and a distance traveled while executing the maneuver (determined by maneuver time/distance and vehicle speed). Figure 5.2 shows this simple model diagrammatically. As the figure shows, the PRT component is itself viewed as a series of steps. These individual steps are not explicit in the design equation but are included in the assumptions that underlie the PRT value. [Section 4.5] Design equations and their assumptions for specific maneuvers are dealt with in subsequent sections of this chapter. The sequential model of driver behavior shown in Figure 5.2 is a shared common conceptual underpinning of various sight distance equations. C-10

Figure 5.2. Diagrammatic version of the basic sight distance model The figure shows the linear chain of steps that comprise the PRT, and after the PRT is complete, the execution of the selected driving maneuver. As a description of what actually happens during driving, we can treat this model as a “convenient fiction.” It is a simple, fixed, linear, mechanistic process. As such, it provides a useful basis for deriving approximate quantitative values for design requirements that work for many situations. Real human behavior is far more complex than this. However, the highway designer or traffic engineer needs to work with less complex models of human visual perception, attention, information processing, and motivation. What is important, however, is to appreciate those factors that may affect application of design sight distance models for particular situations. This will help to prevent, recognize, or deal with sight distance issues. For a particular situation, the standard sight distance design equation might either underestimate or overestimate the actual needs of a driver. Subsequent sections of this chapter deal with specific factors that affect the human factors of the driver response and provide guidance for working with them. Before looking at these specific applications, it is useful to have an appreciation of how the simple driver models that underlie sight distance requirements contrast with the real complexities of driver behavior. Some places where actual driver behavior contrasts with the underlying basic sight distance model include the following: • What happens prior to encountering the object/event? The model shown in Figure 5.2 is not sensitive to things that happen prior to the moment that the potentially hazardous object or event becomes visible to the driver. In reality, the readiness with which drivers react may be strongly influenced by what happens leading up to the event. For example, drivers traveling on a roadway with few access points and little traffic may be unprepared to stop for a slow moving vehicle ahead. In contrast, if drivers had been encountering numerous commercial driveways and intersections, with entering truck traffic, they might more readily react to the vehicle. Roadway design and operational features in advance of the situation therefore are important influences that are not explicit in the basic model. Figure 5.3 shows an expansion of the basic model to include as a stage what happens prior to the object/event. C-11

Figure 5.3 Added components to basic sight distance behavioral model In the figure, two additional components to the model are shown prior to the event becoming visible. One component is labeled “cognitive preparation.” This is a general term to encompass various active mental activities that can influence response times and decisions. These include such things as driver expectancies [Section 4.4], situational awareness [Section 4.2], a general sense of caution [Section 4.7], and where attention is being directed [Section 4.2]. Part II (“Bringing Road User Capabilities into Highway Design and Traffic Engineering Practice”) of this manual [Section 2.0] provides some further explanation of these factors. As the arrows in the figure show, the driver’s cognitive preparation as he or she encounters the object/event can influence the speed of detection, the speed and accuracy of recognizing the situation, and the speed and type of decision made about how to respond. The critical point is that PRT at some point on the road is influenced by what drivers encounter as they approach the site. The other component in Figure 5.3 that occurs prior to the visible hazardous object or event is speed selection [Section 4.6, Chapter X on Speed Control]. As discussed earlier, speed can have perceptual effects, influencing how easily a target object is detected or how accurately gaps are judged. Speed may affect the driver’s sense of urgency, which can influence what maneuver options are considered and their relative appeal. Speed also may directly effect the difficulty, as well as the required time or distance, of the maneuver. Therefore whatever influences speed choice prior to the event may influence the driver decision process, as well as impacting the time available for the driver response. The basic sight distance behavioral model (Figure 5.2) makes assumptions about driver cognitive state and speed choice as the hazardous event is encountered. In reality, the driver does not arrive at the situation as a “blank slate.” The locus of a sight distance problem, or its solution, therefore may turn out to be in advance of the problem site itself. C-12

• Information processing [Section 4.3]. The behavioral model shows a chain of mental and physical events occurring in a sequential fashion. A key bit of information becomes visible; the presence of this event is detected; once detected, it eventually becomes recognized and understood; then a decision is made about what maneuver is needed; then that action gets initiated; and once initiated, the maneuver runs off. Each step takes some amount of time, and one step does not begin until the previous step is complete. This assumed “serial processing” model is one way a driver could respond, but it is not typical. For example, if drivers see some vague object ahead that might or might not be in the roadway, they may begin to brake even before the object is recognized. Once the object is recognized, the maneuver may be reconsidered. The mental processes shown by the various boxes in Figure 5.2 may actually occur in parallel, in a different sequence, and with modifications (feedback loops) as the process continues. The assumed linear response sequence is therefore really a special case used for design purposes. It should not be viewed as a necessarily realistic representation of the more complex perceptual and cognitive activity in complex driving situations. • Smooth driving versus a series of episodes. Related to the point above, the model underlying driver sight distance requirements could be described as “episodic.” Some object or event occurs, then some driver reaction to it takes place. Then another object or event occurs, and another reaction takes place. Real driving is normally smooth and continuous; it is not a jerky sequence of little episodes. Yet for ease of analysis, we often break driver behavior into little stimulus/response events, or treat the roadway as a succession of discrete segments or zones. To the driver, the roadway and the driving task are generally smooth and continuous. Real drivers do not just react; they plan and predict and manage and adapt to events as they go along. This view of driver performance is much more difficult to model and quantify, especially in a manner that easily will generate a simple design parameter. From a human factors perspective, sight distance models are based on a little bit of behavior that describes how a driver might react, and not on how drivers typically behave. However, this is generally reasonable from a design perspective, because it is somewhat conservative: those drivers who encounter a situation without planning or anticipation are those likely to be most in need of the full sight distance requirement. • The hazard. For each sight distance design application, the analysis is based around some object, event, or roadway feature that must be responded to with a driving maneuver. That cue might be debris in the roadway, braking by a vehicle ahead, an approaching vehicle on a conflicting path, a freeway lane drop, a change in signal phase, a pedestrian entering the road, a railroad gate, an animal, a vehicle entering from a driveway, or many other things. The PRT process begins with the potentially hazardous object or event (the “visual target”) becoming visible to the driver [Section 4.1], followed by some time to visually detect and recognize that target. Design equations have to include some estimate of when something becomes visible and how long driver reaction will take. The examples of various hazards suggest just how different these may be as visual targets, so making a single assumption is an obvious simplification. A target object may be large or small, bright or dull, familiar or unfamiliar, moving or stationary, or have other attributes that affect the speed of detection and recognition. Explicitly or implicitly, design equations have to make some assumption about the characteristics of the visual target. Furthermore, visibility conditions may vary with weather, glare, light condition, roadway lighting, and intervening traffic (especially truck traffic). Again, design equations must be based on some assumption about visibility conditions. A PRT model requires the user to be able to specify the point in time or space that the hazard becomes visible to the driver. This, too, is sometimes an over-simplification. For example, there is no sharp threshold where an object in the road suddenly goes from being invisible to visible. Some hazards do not occur all at once, but evolve over some time, such as a vehicle moving into C-13

a lane in front of a driver. Some events might have a preview, such as a vehicle positioned in a driveway prior to its pulling out, or children playing near the road prior to entering the road. Some events might have multiple cues. For example, a freeway lane drop has signing, an initial taper, lane markings, and the point where the lane finally disappears. Sometimes the important visual target is not the hazard object or event itself but a cue about the hazard. For example, brake lights on a vehicle ahead may be a warning about a sudden severe deceleration, but they may also reflect a minor tap on the brake. Drivers cannot respond to the brake light in the same way they respond to recognition of the actual deceleration. • The response. The behavioral components of sight distance models are based around some very specific maneuver in response to the object/event, with fixed assumptions about response parameters. For example, for responding to an unexpected need to stop, AASHTO (2001) [AASHTO chapter 3] assumes a braking maneuver with a deceleration of 3.4 m/s2 (11.2 ft/s2). Braking may be a reasonable response to assume, and 3.4 m/s2 may be a reasonable deceleration to assume, but this certainly does not mean that braking at this rate is the driver response to an unexpected hazard. The maneuver time and maneuver distance components of sight distance models are in many cases based on good empirical research and human factors considerations and work well for most applications. Still, the use of a single standard value is a convenient simplification. Actual maneuvers can be influenced by various factors. The perceived urgency of the situation (based on available time/distance, driver/vehicle capabilities) determines options and shapes the way drivers’ respond, and often there are multiple options. For example, for an unanticipated stop, a driver may brake severely, or brake gradually and steer around, or swerve sharply. The surrounding physical, traffic, and social environment will affect these options: is there a lane or shoulder to steer around, are there adjacent or following vehicles, is the obstacle a piece of debris or a child, is there a passenger in the vehicle? Drivers also make a trade-off of speed versus control when executing maneuvers. The AASHTO deceleration value of 3.4 m/s2 represents an estimate of a “comfortable deceleration” with which almost all drivers can maintain good vehicle control. In this sense it is appropriate for general design, but does not necessarily describe what drivers can do or actually do. Furthermore, once a driver initially selects and begins to execute a particular maneuver, that action does not simply reel off in a fixed manner. As Figure 5.3 illustrates, the situation is monitored and the maneuver is re-evaluated as it is being executed. The response may be refined or modified as it progresses. Drivers may not respond to a situation with a maximum response, but rather may initiate a more controlled action and monitor the situation before committing to a more extreme action. For instance, they may begin gradual braking and check their mirrors for following traffic before decelerating more sharply or swerving. • The driver. The diverse driving population ranges widely in capabilities and behaviors [Section 4.0]. Drivers vary in visual acuity, useful field of view, eye height, information processing rate, tolerance for deceleration, and other factors related to PRT and MT. A design equation is based around a design driver with some assumed set of attributes. For conservatism, the assumptions usually do not represent a typical driver, but rather poorer performing individuals (e.g., 15th percentile in terms of some attribute). Assumptions are made about the state of the driver as well. For example, data are generally based on drivers who are sober and alert. Yet impaired or fatigued drivers [Section 4.9] may represent a large part of the crash risk. Alcohol, drugs, medication, and fatigue can have dramatic effects on the psychological processes that underlie PRT and maneuver execution. Driver distraction by activity within the vehicle is also a common occurrence that is not reflected in the design model. In-vehicle technologies, such as cell phones, navigation systems, and infotainment systems are increasingly common. The “multitasking” driver [Section 4.2] is an increasing concern, but PRT models do not reflect this possibility. C-14

• Empirical findings. The values used in design equations may or may not derive from good empirical sources. In some cases (e.g., brake reaction time) there are numerous empirical studies and reasonably good agreement among them. In other cases, empirical data are very limited. The numbers that come from empirical studies are sometimes questionable on a number of grounds: the sample of drivers is small or unrepresentative; the situations evaluated are limited and may not generalize well; the research may be out of date (given changes in roadways, traffic, vehicles, traffic control devices, driver norms); the research setting (test track, simulator, laboratory) may lack validity; and there may be conflicting results with other studies. It would be wrong to assume that sight distance design equations are necessarily based on a strong empirical foundation that readily generalizes to all cases. General design equations based on simple behavioral models cannot incorporate site-specific considerations. Empirical observations made at the site may be at variance with the predicted behaviors. Even when design equations are based on “good” data, the generality of the models suggest that credence be given to any empirical data that can be collected at the site itself. In summary, sight distance requirements are based on a highly simplified and mechanistic model of driver behavior and capabilities. This is a reasonable and generally successful approach. The general assumptions often work well enough to approximate the needs of most drivers. But it is important to recognize that this simple model has a number of limitations as a description of actual driver performance. When diagnosing or addressing difficult sight distance problems, it may be useful to recognize how design models simplify driver actions and to acknowledge realities of more complex driver perception and behavior. 5.1.3 The Relationship of This Chapter to Other Key Reference Documents This sight distance chapter is related to the following key references: AASHTO Policy on Geometric Design of Highways and Streets (2001) • Chapter 2, Design Controls and Criteria: discusses driver reaction time and related issues in Driver Performance subhead • Chapter 3, Elements of Design: section on sight distance, with subsections on stopping sight distance, decision sight distance, passing sight distance, sight distance for multilane highways • Chapters 5 (Local Roads and Streets), 6 (Collector Roads and Streets), 7 (Rural and Urban Arterials), and 9 (Intersections) all have specific subsections on sight distance Manual on Uniform Traffic Control Devices (2003) • The MUTCD has several tables relating minimum sight distance to speed. These include Table 3B-1 (for passing sight distance), Table 4D-1 (for traffic control signal sight distance), Table 6C- 2 (for work zone tapers), and Table 6E-1 (for work zone flagger stations) • Section 2C.05, Placement of Warning Signs, describes a PRT model known as the PIEV (Perception-Identification-Emotion-Volition) model. Tables 2C-4 (metric units) and 2C-5 (English units) show advance warning sign placement as a function of speed based on PIEV time requirements. ITE Traffic Engineering Handbook (1999) • Chapter 2, Road Users, has sections on perception-reaction time and sight distance • Chapter 11, Geometric Design of Highways, has a section on sight distance, with subsections on stopping sight distance, passing sight distance, decision sight distance, and intersection sight distance C-15

ITE Traffic Control Devices Handbook (2001) • Chapter 2, Human Factors, has sections on driver perception reaction time, maneuver time • Chapter 11, Highway-Rail Grade Crossings, contains discussion of sight distance requirements for at-grade crossings Highway Safety Manual (under development) • Planned section 2.4, Fundamentals: Human Factors in Road Safety will include perception- reaction time and related human factors issues • Sight distance may be expected to included in various places in Part II – Knowledge, but these chapters have not yet been developed Highway Design Handbook for Older Drivers and Pedestrians (2001) • Rationale and Supporting Evidence section contains evaluation of perception-reaction time and sight distance requirements for older drivers • Various design recommendations to support older drivers include consideration of older driver perception-reaction time and sight distance needs 5.2 Design Sight Distances Design sight distances that depend on driver PRT and MT are as follows: • Stopping sight distance • Intersection sight distance • Decision sight distance • Passing sight distance • Railroad-highway grade crossing sight distance In this general chapter, only two of the eleven cases of intersection sight distance defined in AASHTO (2001, Chapter 9) are considered. The remainder are considered in the chapters on intersections. [Chapter 10, 11] Similarly, railroad-highway grade crossing sight distance is not considered here, but dealt with in the railroad-crossing chapter. [Chapter 14] The appropriate values for PRT and MT depend on the specific driving task underlying the sight distance requirement. They also depend on the degree of urgency. Drivers responding to a child running into the road are likely to have much shorter PRT and MT than might be the case for debris on the road. Drivers who see cues to an upcoming lane drop well back from the gore will have longer PRTs and MTs than drivers who are near the gore when they suddenly realize they must change lanes. Ideally the highway should be designed to allow for comfortable, less stressful responses. Although there are many studies of PRT and MT, very few determine how comfortable the driver found the driving task for a given PRT and MT. The following sections provide, for each type of sight distance: • Definition • High Priority Considerations • Guideline for PRT o Under baseline conditions C-16

o Under unfavorable conditions • Guideline for MT o Under baseline conditions o Under unfavorable conditions • Rationale for Guideline • Summary Table Definition: Each sight distance definition is taken from the AASHTO Policy on Geometric Design of Highways and Streets (2001). High Priority Considerations: AASHTO sight distances should always be provided. However, if a given sight distance is below standard at a number of locations, or if design tradeoffs must be made at a given location, then priorities must be set. High priority considerations give guidance in making design tradeoffs and setting priorities with respect to driver needs. Baseline Conditions: PRT and MT values given are generally 85th percentile or more values, since these include the majority of the driving population. Not all studies provide such values and in some cases, 85th percentile values are calculated from means and standard deviations. Older, novice, and unfamiliar drivers in passenger vehicles are the assumed baseline in addition to any other factors specifically noted. Unfavorable Conditions: Design (e.g. unusual geometric layout), environmental (e.g. nighttime) and operational conditions can increase driver requirements in some circumstances. Guidance is provided for those conditions for which data are available or for which the direction of effect on PRT or MT for a given variable (e.g. nighttime, unusual geometric layout, higher workload driving task) can be predicted. Driver characteristics including age, impairment (fatigue, alcohol, medical conditions), and familiarity affect PRT and MT, and where appropriate these will be discussed (see Chapter 2 [Chapter 2] for an extensive discussion of driver characteristics). Rationale: The rationale section provides a short summary of the studies used to develop the PRT and MT guidelines. Summary Table: This table provides a summary of the driver, operational and geometric factors that affect PRT and MT, guideline PRT and MT values, guideline SD values and AASHTO values for comparison. 5.2.1 Stopping Sight Distance 5.2.1.1 DEFINITION: SSD [AASHTO 2001 Ch 3] Stopping sight distance, as referred to by AASHTO (2001) “should be sufficiently long to enable a vehicle traveling at or near design speed to stop before reaching a stationary object in its path.” C-17

SSD is defined as follows: Metric US Customary V2 SSD = 0.278 VtPRT + 0.039 A V2 SSD = 1.47 VtPRT + 1.075 A where: tPRT = brake reaction time, 2.5 s V = design speed km/h A = deceleration rate, m/s2 where: tPRT = brake reaction time, 2.5 s V = design speed, mph A = deceleration rate, ft/s2 The current AASHTO value for PRT is 2.5 seconds. MT assumes that drivers are 100% efficient in braking, i.e., locked wheel braking, and that pavement friction is very poor. 5.2.1.2 HIGH PRIORITY CONSIDERATIONS: SSD Stopping sight distance should always be provided because any road location can become a hazard. A study (Fambro et al., 1996) found that the most common objects hit on sight-restricted curves were large animals and parked cars, the presence of which can create a hazard on any road section. If stopping sight distance is below standard at a number of locations then priorities must be set. Examples of hazards and conditions which are high priority with respect to the need for stopping sight distance are: • Change in lane width • Reduction in lateral clearance • Beginning of hazardous fill slope • Crest vertical curve • Horizontal curve • Driveway • Narrow Bridge • Roadside hazards – e.g., boulder markers at driveways • Unmarked crossovers on high-speed rural arterials • Unlit pedestrian crosswalks • High volume pedestrian crosswalks • Frequent presence of parked vehicles very near or intruding into the through lane • Slow moving vehicle • Frequent pedestrian or bicycle presence 5.2.1.3 GUIDELINE FOR SSD Guidelines for SSD PRT and MT are shown on the next two pages. C-18

STOPPING SIGHT DISTANCE: PERCEPTION-REACTION TIME GUIDELINE Under baseline conditions: Most reasonably alert drivers (95%) will be able to initiate braking within PRT of 1.6 s. • Daytime o Hazard clearly visible and directly in driver’s line of sight • Nighttime o Self-illuminated or retro-reflectorized hazard, with a lighting configuration that is immediately recognizable, near driver’s line of sight Under unfavorable conditions: Once the object is detectable, PRTs in unfavorable conditions can be 5 s or more. • Daytime o Hazard camouflaged by background and initially off line of sight • Nighttime o Hazard unreflectorized and not self-illuminated o Hazard self-illuminated or retro-reflectorized but lighting configuration is unfamiliar to the driver o Low beam headlights with or without streetlighting o Hazard off line of sight o Glare from oncoming vehicles or commercial lighting PRT does not start until drivers can see and, to some degree, recognize the hazard. The distance at which drivers can see an unilluminated, unreflectorized hazard depends on their headlights, their sensitivity to contrast and on their expectation of seeing the hazard. When drivers are not expecting a particular low contrast hazard (e.g. unreflectorized jersey barrier), their seeing distance is one half that that would pertain if the object were expected. At speeds of 60 km/h and greater, using low beam headlights, most drivers will be too close to an unexpected, unreflectorized hazard at the point they can detect it in time to stop. A very low contrast hazard may not even be detected in time to start braking. Therefore objects blocking the road path, such as traffic islands or jersey barriers in a construction zone must be reflectorized. Drivers confronted with an unusual lighting configuration (e.g. a white worklight on the rear of a tractor, or a flat bed trailer’s single amber light in the middle of a lane) may not begin the PRT until they can determine what the light is attached to. PRT can be increased by the following driver factors: • High workload (e.g. traffic merging, several signs to be read) • Fatigue and impairment C-19

5.2.1.4 BASIS/RATIONALE FOR SSD GUIDELINE SSD PRT Stopping sight distance PRT has been addressed in a variety of experimental studies. The principal studies include the following: • Daytime PRT for clearly visible hazard placed on the road (Olson, Cleveland, Fancher, & Schneider, 1984) • Daytime PRT for clearly visible hazard emerging from the side of the road (Lerner, Huey, McGee, & Sullivan, 1995) • Nighttime studies for unilluminated, unreflectorized hazards placed on the road (Olson & Sivak, 1983; Fambro, Fitzpatrick, & Koppa, 1997) • Simulator studies of PRT for low contrast targets on the side of the road (Ranney, Masalonis, & Simmons, 1996) STOPPING SIGHT DISTANCE: MANEUVER TIME GUIDELINE Under baseline conditions: Based on Fambro et al. (1997), a study conducted on flat road sections, and dry pavements, the mean constant deceleration is about 0.55 g (69% of the pavement’s coefficient of friction), and the 85th percentile is 0.38g (60%). On dry pavements no difference between ABS and standard brakes is expected. • Tangent • Dry or wet pavement • No grade • Passenger vehicles • Unexpected object • Tires in good condition Under wet conditions, with standard brakes, the mean constant deceleration is about .43g (54% of the pavement’s coefficient of friction), and the 85th percentile is .38g (47%). On wet pavements with ABS, the mean constant deceleration is about 0.53g (66% of the pavement’s coefficient of friction), and the 85th percentile is about 0.45g (56%). Under unfavorable conditions: Slightly lower braking efficiencies (by 2 – 8%) are obtained on curves. Based on physics, downgrades increase MT. No human factors studies are available. The assumed decrease in MT is V*f*grade%. • Curve versus tangent • Downgrade MT can be increased by the following driver factors: • Age • Gender Older drivers and women will not apply as much braking force as younger drivers and males. C-20

• Nighttime on-road study comparing detection distance for alerted and unalerted drivers (Roper & Howard, 1938) In a daytime study conducted for the purposes of assessing the established AASHTO value for PRT in a stopping sight distance situation, Olson et al. (1984) measured PRT for a group of drivers who suddenly encountered objects of various sizes in the middle of their lane as they crested a hill in the daytime. PRT was defined as the time elapsed between when the object first became visible until the point at which it was detected by the driver. The 85th percentile PRT for 49 young subjects was 1.3 seconds, and for older drivers, very similar at 1.4 seconds. In a study conducted by Lerner et al., subjects encountered a yellow barrel released at a predetermined point (on average 3.4 sec away) which rolled to the edge of the lane, restrained by chains (Lerner et al. 1995). Data were collected from 30 subjects aged 20 – 40 years, 43 subjects aged 65 – 69 years and 43 subjects aged 70+. The 85th percentile PRT was 1.9 seconds, and this value was the same for the older and younger age groups. The longest observed PRT was 2.5 seconds. Fambro et al. (1997) conducted two studies involving unexpected hazards, the first being a barricade that popped up suddenly in front of the driver and the second, a barrel which rolled off a truck parked by the side of the road. In both cases the hazards would have sufficiently contrasted with the background against which they were seen to have been immediately detectable. Based on 22 younger (age 24 years or less), and 24 (age 55 years or more) older subjects, the 95th percentile PRT was 2.0 seconds. There was no difference between younger and older subjects in response to unexpected hazards. As the object to be detected becomes more difficult to see, because of low contrast and/or low light levels or glare, PRT lengthens. In a study conducted in a driving simulator with 8 middle-aged subjects (aged 38 to 62 years) Ranney et al. 1996 examined PRT in response to a low contrast pedestrian target at the side of the road. Once the target was detectable, the mean PRT was 2.8 seconds and, assuming a normal distribution, based on the reported standard deviation, the 85th percentile PRT was 3.9 seconds, when no glare was present. This increased to a mean PRT of 3.5 seconds and an 85th percentile PRT of 4.9 seconds when glare equivalent to that of an oncoming vehicle was present. In a study of nighttime detection of various hazards that might be encountered on a road (e.g. animal, tire, tree limb, etc.) Fambro et al. (1997) found that drivers do not have the visual capabilities to recognize objects that are less than 30 cm (11 inches) in height at or beyond the AASHTO minimum stopping sight distance of 128 m. (420 ft) at 90 km/h (56 mph). PRT cannot start until drivers detect and partially, at least, recognize the detected object. Two studies have shown that when drivers are not expecting to see an object, the distance at which they see it is considerably less when they know they are about to encounter it (Roper and Howard,1938; Shinar, 1985). Based on the Roper and Howard (1938) study, in which subjects were as unalert as it is probably possible to be in an experimental study, Olson (2002) estimated that at 35 km/h (22 mph) fewer than 10% of unalerted drivers would be able to see an unreflectorized target (a pedestrian in a dark coat) on the left hand side of the road, and less than half would be able to see the same target on the right hand side of the road, in time to stop (Olson, 2002). SSD MT Stopping sight distance MT from 88 km/h (55 mph) for typical drivers, as opposed to test drivers, was recorded by Fambro et al. (1997) in the studies involving unexpected hazards described above, on pavement with a coefficient of friction of 0.8 g in dry conditions on a flat surface. When research participants used their own vehicles and responded to an unexpected object, on dry pavements, the mean C-21

constant deceleration was 0.55 g, and the 85th percentile, 0.48 g. When research participants used test vehicles, they decelerated more rapidly than they did in their own vehicles, with an equivalent constant deceleration that was about 14% higher. In this study no differences were found in braking efficiency between ABS and standard brakes on dry pavements. Under wet conditions, with standard brakes, the mean constant deceleration was about 0.43 g, and the 85th percentile, 0.38 g. On wet pavements with ABS, the mean constant deceleration was about 0.53 g, and the 85th percentile was about 0.45 g. In the study in which drivers used test cars, slightly higher g values (by 2 - 9%) were obtained on tangents as compared to curves. In a study similar to the Fambro et al. (1997) study, in virtually every braking situation tested, drivers stopped rapidly, but not to the point of locked wheel braking (Lerner et al., 1995). In locked wheel braking drivers are 100% efficient in making use of the available pavement friction. Locked wheel braking is typical in accidents. Clearly urgency plays a major role in determining braking MT’s. At intersections, when time gaps are 10 seconds or less, Harwood et al. (1996) found that major road vehicles slow by an average rate of 0.68 m/sec2 (2.2 ft/sec2), to accommodate entering minor road vehicles. For traffic decelerating at traffic lights, Wortman and Matthais found an average rate of 3 m/sec2 (10 ft/sec2). Given the coefficient of friction for the pavement used in the Fambro et al. study, the average rate of braking was 4.7 m/sec2 (15.4 ft/sec2) on dry pavement (Wortman & Matthais, 1983). For design purposes, neither rapid nor locked wheel braking is a desirable driver response, given the risk of a rear-end crash when there is a following vehicle. Although the AASHTO model assumes locked wheel braking, it also assumes poor pavement and tire conditions, neither of which may be present, making the assumption of locked wheel braking less problematic. The AASHTO model also assumes constant deceleration throughout the braking maneuver, and Fambro et al. found that deceleration profiles are not linear. Maximum deceleration was generally not exhibited until the last part of the braking when the vehicle had slowed and come closer to the unexpected object. The mean maximum deceleration was about 75% of the pavement’s coefficient of friction. Under wet conditions, the 95th percentile value for equivalent constant deceleration without ABS was 0.29 g (equivalent to 2.8 m/sec2 (9.3 ft/sec2), and with ABS, 0.41 g (equivalent to 4 m/sec2 (13.2 ft/sec2). SUMMARY: SSD PRT Factors PRT MT Factors Mean Deceleration (g) AASHTO Driver workload 1.6 to 5+ sec Driver age 60% f dry 2.5 sec + 100%f Poor visibility Urgency 56% f wet Low contrast hazard 47% f wet ABS Hazard off line of sight Unfamiliar object f = pavement coefficient of friction C-22

5.2.2 Intersection Sight Distance 5.2.2.1 DEFINITION: ISD Intersection sight distances (ISD) are the minimum sight distances required for drivers to safely negotiate intersections, including those with no control, stop control and signals, and including those for drivers turning left, right and going straight through. Until the 2001 version of the AASHTO Policy, ISD values have been calculated using models that assume a serial process whereby PRT is completed while the driver is stopped at the stop bar, followed by an acceleration time. For the simplest form of ISD, involving crossing or turning from a stop control on a minor road, the equation form used was: Metric US Customary ISD = 0.278 Vmajor(J + ta) ISD = 1.47 Vmajor(J + ta) where: ISD = intersection sight distance (length of the leg of sight triangle along the major road (m) Vmajor = design speed of major road (km/h) J = PRT required to determine if an available gap or lag is acceptable (s) ta = MT to accelerate and traverse the major highway pavement (for a crossing maneuver) or to accelerate and reach 85% of the major highway design speed (for a turning maneuver (s)) where: ISD = intersection sight distance (length of the leg of sight triangle along the major road (ft) Vmajor = design speed of major road (mph) J = PRT required to determine if an available gap or lag is acceptable (s) ta = MT to accelerate and traverse the major highway pavement (for a crossing maneuver) or to accelerate and reach 85% of the major highway design speed (for a turning maneuver (s)) There were seven AASHTO model situations (Case 1-V with 3 variations for III) which dealt with through, left and right turning maneuvers at intersections with no control, stop control, yield control and signal control. The values used in the AASHTO equations were based on limited empirical data. In the 2001 AASHTO Policy, ISD is no longer based on the serial model assuming that PRT starts when the driver is stopped at the stop bar, is completed before leaving the stop bar, followed by an acceleration time. Instead ISD is based on a gap acceptance model, in which the time gaps accepted by drivers for the various maneuvers made at intersections are the basis. Although gap acceptance is an alternative means of conceptualizing driver requirements for ISD, this does not imply that the various elements of the traditional sight distance model (Figures 5.2 and 5.3) are not important at intersections. PRT is completed once drivers have decided to accept the gap, but before they move forward. The time gap accepted must be of sufficient length to accommodate their estimated MT, without requiring substantial braking from the oncoming driver. The new model uses an equation form as follows: C-23

Metric US Customary ISD = 0.278 Vmajortg ISD = 1.47 Vmajortg where: ISD = intersection sight distance (length of the leg of sight triangle along the major road (m) Vmajor = design speed of major road (km/h) tg = time gap for minor road vehicle to enter the major road(s) where: ISD = intersection sight distance (length of the leg of sight triangle along the major road (ft) Vmajor = design speed of major road (mph) tg = time gap for minor road vehicle to enter the major road(s) In these equations, tg is the gap in seconds accepted by drivers 50% of the time it is presented for crossing or turning maneuvers. In the 2001 AASHTO Policy, there are a total of 11 AASHTO model situations which deal with: through, left and right turning maneuvers at intersections with no control, 4 way stop control, 2 way stop control, yield control and signal control from the minor road. In addition ISD for a left turning maneuver from the major road is considered. The object height is considered to be equivalent to the driver’s eye height of 1.08 m (3.5 ft) above the surface of the intersecting road. From a driver behavior perspective, it should be noted that both the PRT-based ISD equations and the gap acceptance ISD equations contain an assumption of some cooperative behavior from the conflicting (major road) traffic. If approaching traffic does not slow to some degree, the equations may not work. AASHTO (2001) notes that the values given for sight distance (e.g., Exhibit 9-54) “provide sufficient time for the minor road vehicle to accelerate from a stop and complete a left turn without unduly interfering with major-road traffic operations.” [emphasis added] Further considering the values for the gap acceptance model, AASHTO states: “Observations have also shown that major-road drivers will reduce their speeds to some extent when minor-road vehicles turn onto the major road. Where the time gap acceptance values in Exhibit 9-54 are used to determine the length of the leg of the departure sight triangle, most major road drivers should not need to reduce speed to less than 70 percent of their initial speed.” The previous PRT-based models also contained assumptions that major-road traffic may have to slow from design speed (AASHTO, 1990). For example, left and right turning maneuvers from a stop are based on the time it takes for the turning vehicle to achieve 85% of design speed before being overtaken by vehicles on the major road “that are approaching the intersection from the (left or right) and are reducing their speed from the design speed to 85 percent of the design speed.” In the guideline below ISD is considered for turning and crossing maneuvers from a minor road with a stop control. Guidance for other ISD situations is considered in more detail in a later chapter on intersections. 5.2.2.2 HIGH PRIORITY CONSIDERATIONS: ISD PRT It is particularly important to provide adequate intersection sight distance wherever a significant level of visual clutter or overload exists, for example where there are: • High major road volumes • Complex signs (multiple destinations, route shield assemblies) C-24

• Complex pavement markings (multiple turn lanes) • Complex or atypical intersection geometry • Visual clutter in urban areas due to commercial lighting • A high percentage of older drivers. It is also important to provide adequate intersection sight distance wherever drivers are less likely to be expecting to respond to an intersection, for example: • Requirement to stop is unexpected due to right of way on previous road section • Stop or signal controlled isolated intersection • Intersections with high volume but signals not yet warranted In these situations, ISD is a minimum – it is preferable to provide DSD [Section 5.2.3]. 5.2.2.3 GUIDELINE FOR ISD Because of the recent change in AASHTO Policy ISD has been considered both from the perspective of the traditional model which considers PRT and MT separately as well as from the perspective of the 2001 AASHTO model, which is based on gap acceptance. The accepted time gap is measured from the moment of perceptible movement of the vehicle, that is, after the PRT is finished. Thus time gap measures do not include PRT. The ISD guideline below applies to crossing and turning maneuvers from a minor road (Cases IIIA, B, C AASHTO Policy 1994, Cases C1 and C2 AASHTO Policy 2001). Guidelines follow for ISD PRT, ISD MT and ISD Critical Gap. C-25

INTERSECTION SIGHT DISTANCE: PERCEPTION REACTION TIME GUIDELINES Under baseline conditions (based on Lerner et al., 1995) the median PRT is about 1.3 sec, and the 85th percentile PRT is about 2.0 sec. PRTs are longer for: • Younger drivers (by about 0.2 sec) • Female drivers (but difference is mainly in daytime) • Drivers using standard transmissions (by 0.06 to 0.38 sec depending on age) • Under daytime conditions Fewer night sessions than day sessions were run when it became apparent that day values were higher than night values. Under unfavorable conditions: PRT may be lengthened. • Turning right through the minor angle of skew intersection • Crossing or turning at an intersection on a horizontal curve where the main road curves behind the driver • Crossing at an offset intersection In the first case, drivers must turn their heads through a greater angle to assess the presence of oncoming vehicles. In the second case the assessment of the acceptability of the gap may take longer due to the greater complexity of the geometry. C-26

******************************************************* Figure. Static or dynamic illustrations showing turning through major and minor angles of skewed intersection ********************************************************** INTERSECTION SIGHT DISTANCE: MANEUVER TIME GUIDELINES Under baseline conditions: The 85th percentile value for the time from initiation of the maneuver to where the vehicle is oriented parallel with the roadway is about 6.3 sec. • Turning from right angle intersection • Turning through the major angle of skew intersection Under unfavorable conditions, such as the following, MT may be lengthened: • Turning right through the minor angle of skew intersection • Crossing or turning: o At an intersection on a horizontal curve where the main road curves behind the driver o On an upgrade o On wet or slippery pavement o In trucks • Crossing at an offset intersection The first condition is unfavorable because drivers must turn through a greater angle to fully complete the turn than is the case at a right-angled intersection. This is also true when a driver turns on a curve where the main road curves behind him or her. On an upgrade, on wet or slippery pavement, and for trucks, acceleration is likely to be slower, increasing MT. At an offset intersection, crossing includes two turns, increasing MT. MT can be affected slightly (< 0.4 sec) by the following driver factors: • Age and gender C-27

INTERSECTION SIGHT DISTANCE: TIME GAP GUIDELINE Under baseline conditions: The 85th percentile gap accepted by left turning passenger car drivers, including substantial numbers of older drivers, is 11 sec. This is more than a 50th percentile gap for single unit trucks and close to a 50th percentile gap for double unit trucks. • Turning left at right angle intersection • Turning through the major angle of skew intersection These values apply to posted or advisory speeds ranging from 56 to 72 km/h (35 to 55 mph), major road traffic volumes ranging from 1,750 to 13,500 and minor road volumes from 2000 to 6600 AADT. At higher volumes accepted gaps will be shorter as drivers feel pressured to turn when others are waiting. Under unfavorable conditions: accepted gaps may be longer. • Trucks turning • Turning right through the minor angle of skew intersection • Crossing or turning at an intersection on a horizontal curve • Crossing at an offset intersection Single unit trucks require on average 2.6 sec, and double unit trucks, 4.0 sec gaps for left turns than passenger car drivers. At intersections with “difficult” geometry (e.g. offset or curve), the best estimate on the basis of very limited data is as much as 1-2 sec addition gap required for passenger car drivers. When drivers must make a right turn through the minor angle of a skew intersection, their major search is to the left, which requires a more extensive head turn, and which would be expected to lengthen PRT and therefore the accepted gap. Passenger car critical gaps for right turns are approximately 1.7 sec shorter than for left turns. On a multilane situation, a 0.7 second adjustment per additional lane in the critical gap size should be made for right turns, 0.4 seconds for left turns and 0.5 seconds for crossing maneuvers. In other words, for a three-lane crossing and a right turning passenger car driver, the critical gap would be 11 sec (for 85th percentile left turning passenger car driver for a single lane in each direction) -1.7 sec (to account for right turn) -2 x 0.7 sec (to account for two additional lanes to cross). For intersections on a grade, critical gaps are longer by 0.1 second per percent grade for right turns, and 0.2 seconds per percent grade for left turns or crossing maneuvers. Accepted gaps for older drivers average about 1 sec longer than those for younger drivers. The accepted gap may also be lengthened by the following factors: • High workload (e.g. multiple lanes to cross and therefore more than one oncoming vehicle to consider, several signs to be read, entrances and exits in area of influence of the intersection) Because ISD based on time gaps includes assumptions about speed adjustments made by the major-road driver, additional distance may be required in situations where the approaching driver may not slow sufficiently. An approaching major-road driver may not slow sufficiently because recognition of the conflict is delayed or because of aggressive driving. Consider additional sight distance: o Where the major road driver is busy with complex signing, lane drops, or other high-workload demands prior to the intersection in question o Where traffic conditions or site history suggest aggressive driving and driver unwillingness to accommodate entering traffic C-28

Basis/Rationale ISD PRT, MT and Critical Gap The separate components of driver behavior on which ISD depends are difficult to define precisely because drivers generally start the search process while stopping at an intersection and continue their search as they move forward, ready to abandon the maneuver. Thus PRT overlaps MT; it is not a serial process when the driver is stopped as had been traditionally defined by AASHTO. A gap acceptance model, in which PRT and MT are considered as a whole, and search may begin before the driver had stopped, better matches the reality of driver behavior. The critical gap is that gap that drivers accept 50% of the time. Key studies of ISD include: • Naturalistic observations of gap-acceptance for truck and passenger car drivers at six intersections (Fitzpatrick, 1991) • Measurement of PRT and MT at 14 intersections for 96 drivers (33 aged 20-40, 35 aged 65-69 and 34 aged 70+); measurement of critical gap and lag that subjects estimated they would accept (52 aged 20-40, 39 aged 65-69 and 47 aged 70+) (Lerner et al., 1995) • Naturalistic observations of gap-acceptance for passenger car drivers at 44 intersections (Kyte et al., 1996) (in Harwood et al., 1996) • Naturalistic observations of gap-acceptance for truck and passenger car drivers at 13 stop- controlled intersections (Harwood, Mason, Brydia, Pietrucha, & Gittings, 1996) ISD PRT The Lerner et al. (1995) study involved 96 drivers (33 aged 20-40, 35 aged 65-69 and 34 aged 70+) at 14 sites, of which 11 were used for both day and nighttime data collection. The intersection sites varied in terms of cross-section, geometric layout (right-angle vs. skew) and posted speed, and in the maneuver required of the driver(left turn, right turn, through). Drivers were observed while using their own vehicles in an on-road study. Drivers were occupied by having to make a rating while stopped at the intersection, before crossing it, preventing them from starting the PRT process while they were stopping. (Other studies show that drivers typically start the search process within the last few seconds as they approach a stop sign.) PRT and MT were recorded in response to gaps that were accepted. Median PRT was about 1.3 sec, with an 85th percentile PRT of 2.0 sec. PRTs were longer for: • Younger drivers (by about 0.2 sec) • Female drivers (but difference is mainly in daytime) • With standard transmissions (by 0.06 to 0.38 sec depending on age) • Under daytime conditions Fewer night sessions than day sessions were run when it became apparent that day values were higher than night values. PRT may be lengthened at skew or offset intersections. In the first case, drivers must turn their heads through a greater angle to assess the presence of oncoming vehicles. In the second case the assessment of the acceptability of the gap may take longer due to the greater complexity of the geometry. No studies were found of this issue. C-29 5.2.2.3

ISD MT MT for turning movements was determined to have ended at the point where the driver’s vehicle was oriented parallel to the major roadway (as opposed to the AASHTO definition of the end of the maneuver being when the driver has reached 85% of the major road speed). The 85th percentile value for MT was 6.3 sec. The 50th percentile value was 5 sec. Longest maneuver times were for older females; the average of the 65-69 and 70+ females was about ¼ sec longer than the average of the 20-40 year old group. Overall there was little difference between daytime and nighttime. ISD TIME GAP The beginning of measurement of the accepted gap is the point at which drivers have completed their PRT and have decided to accept the gap and their vehicle can be perceived to be moving forward. When drivers detect gap of sufficient length to accommodate their estimated MT, without requiring substantial braking from the oncoming driver, they pull out. A number of studies have made measures of critical gaps. Half of drivers accept a gap of this length when it is presented, while half reject this gap size. Below we consider not only critical gap, but the 85th percentile gap, that is, the gap accepted by 85% of drivers when it is presented. An early study of gap acceptance by Ebbesen et al. involved 2000 observations of left turning vehicles at three different T-intersections, with mean velocities of 40 km/h (25 mph), 61 km/h (38 mph) and 72 km/h (45 mph) as well as at a T intersection where there was considerable variability in velocities of vehicles on the main road (Ebbesen, Parker, & Konecni, 1977). As found by Lerner et al. (1995) and Harwood et al. (1996) in later studies, the critical gap was the same no matter what the speed of vehicles on the cross- road. When three different intersections were compared, each with different speeds for the mainline traffic, the critical gap accepted by left turning traffic was the same - 7.25 seconds. Drivers should require longer gaps at higher speed intersections because they will take longer to bring their speed up to that of the traffic stream. However, it appears that drivers do not estimate their own time requirements well, thereby forcing the following driver to slow. The impact of this underestimation may well be greater at higher speeds. Kyte et al. (unpublished, cited by Harwood et al., 1996) measured critical gaps in the field for passenger cars at 44 two-way stop-controlled intersections. They determined that the critical gap for right turns from a minor road was 6.2 seconds, and for left turns, 7.1 seconds. They further determined that, in a multilane situation, a 0.7 second adjustment in the critical gap size should be made for right turns, 0.4 seconds for left turns and 0.5 seconds for crossing maneuvers. Finally, through statistical analysis, they determined that critical gaps were longer by 0.1 second per percent grade for right turns, and 0.2 seconds per percent grade for left turns or crossing maneuvers. Kyte’s results indicate that drivers are sensitive to the need to allow extra time due to crossing more lanes and due to slower acceleration on an upgrade. Lerner et al. (1995) assessed both lags and gaps. In the case of a gap, the waiting driver is making a judgment about the gap between two moving vehicles. In the case of a lag, the waiting driver is making a judgment about the arrival time on a single vehicle. Subjects in a vehicle stopped on a minor road used a button to indicate whether it was safe to pull out to make a specific maneuver (right turn, left turn, through maneuver in the presence of a gap or a lag). No actual maneuvers were made. For passenger vehicles, the average critical gap was 7 sec., and the 85th percentile was 11 sec. Longer gaps were accepted: • By older drivers (overall oldest drivers required 1.1 more seconds than youngest drivers) C-30

• By female drivers (accept gaps that are 1 sec longer than those accepted by male drivers, but difference is mainly in daytime) • For left and right maneuvers as compared with through maneuvers • Under daytime conditions (by about 1.5 sec) With respect to gaps and lags, Lerner et al. (1995) found that lags accepted were shorter than gaps accepted (5.3 sec on average, vs. 7 sec for a gap). The Harwood et al. (1996) study used naturalistic observation involving videotaping to measure gap- acceptance behavior of drivers at 13 stop-controlled intersections in 3 states as they made left turns and right turns. The study sites included 5 intersections with 3 legs and 8 intersections with 4 legs. Only right- angle intersections were considered. The major road approaches had posted speed limits or advisory speeds ranging from 56 to 72 km/h (35 to 55 mph). All study sites had good safety records. A total of 6243 acceptance/rejection decisions provided data on critical gap for right turn maneuvers; 3526 acceptance/rejection decisions provide data on critical gap for left turn maneuvers. The Table below shows the results indicating that drivers of trucks require longer gaps, in both left and right turn situations, to enter a major road as compared to passenger vehicles. Critical Gaps Derived from Field Data for Right and Left Turns on a Major Road Vehicle Type Critical gap (sec) Raff method Logistic regression RIGHT-TURN MANEUVERS Passenger car 6.3 6.5 Single-unit truck 8.4 9.5 Combination truck 10.7 11.3 LEFT-TURN MANEUVERS Passenger car 8.0 8.2a Single-unit truck 9.8 10.8 Combination truck 10.0 12.2 a Based on an average giving equal weight to each site. (from Harwood et al., 1996) The current AASHTO value of a 7.5 sec time gap for left turning and 6.5 sec for right turning drivers turning in front of passenger cars was developed based on the Harwood et al. study. As shown above, the critical gap for left turns by passenger cars was 8.0 sec. However, when drivers accept gaps less than 10 sec., the major road vehicle typically slows to accommodate the entering vehicle. The median speed reduction of major road vehicles was 31%. This means that an 8.0 sec gap is equivalent to a 7.5 sec gap at the initial speed. The 50th percentile gap, rather than a higher percentile, was used on the basis it is the responsibility of the major road vehicle to accommodate the entering vehicle, and the field studies showing that major road drivers can do so by reducing their speeds by “ 15 to 50% using very modest deceleration rates.” However, the findings by Lerner et al. (1995) suggest that some groups of drivers (e.g. older drivers and female drivers in daytime, and possibly novice drivers, though no data are available on this, as well as drivers of vehicles with standard transmissions) will prefer longer gaps than the average driver. If sight C-31

distance is set at the value that is accepted 50% of the time by a large group of drivers, it may not be that desired by a group of older drivers, whose maneuver times are longer than those for young drivers, and who would willing have pull out onto a major road with a gap of only 7.5 sec. Furthermore, the critical gaps accepted by drivers of single or double unit left turning trucks are substantially longer (by 1.6-1.8 sec for single unit trucks, and by 2 sec for double unit trucks) than that accepted by passenger car drivers, because the larger vehicles requiring more maneuver time and much longer to reach traffic speeds. When trucks turn onto a rural highway, especially at night, the major road driver may not appreciate until they are too close, just how much more slowly the truck is moving. To accommodate these drivers, the gap accepted 85% of the time by passenger cars, that is, 11 sec, is proposed based on the Lerner et al. study. The Lerner et al. study is weighted more heavily towards older drivers than the Harwood et al. study would have been, based as it was on the drivers on the road at the time the study was conducted. Another argument for the 11 sec value, is that it exceeds slightly the critical gap for left-turning straight trucks, measured by the logistic regression method, and is approximately equal the critical gap for left-turning double unit trucks, when the value measured by the logistic regression (12.2 sec) and that measured by the Raff method (10.0 sec) are averaged. The use of an 85th percentile value is especially important on high volume roads where the chances of a vehicle appearing coincident with the driver pulling out are high and on roads that are used by trucks, especially those with trailers. Another issue which is important but has not yet been considered in the ISD Time Gap studies is the safety margin in relation to intersection sight distance. As will be discussed in Section 5.2.4.4, as available passing sight distance increased, there was no consistent effect on judgment time or on time in the opposing lane, but the time safety margin did increase in proportion to available sight distance. In those situations where gaps are hard to judge (see examples below) gaps accepted may be more variable, and longer sight distances would allow for errors in judgment by allowing a longer safety time margin once the gap was accepted. It should be noted though, that there are many reasons for the right-angle intersection crashes which result from errors in gap acceptance; more sight distance would not necessarily eliminate all or even most of such crashes. One study of critical gaps included a site with difficult geometry compared to the other five right angle intersections. This was an offset intersection (by 1.5 m. (5 ft)) on a horizontal curve (Fitzpatrick,1991). Depending on the method used to analyze the critical gap (Greenshields, Raff or logit), the critical gaps at this intersection were determined to be 1 to 2 seconds greater than the 6.5-second critical gap determined for passenger cars at other intersections. There may be two different contributing factors to the longer critical gap. First, while waiting to pull out from an intersection, drivers attempt to estimate the time of arrival of oncoming vehicles based on changes in the apparent size of the approaching vehicle. When an intersection is on a horizontal curve, the apparent size of the approaching vehicle changes, both due to decreasing distance to the intersection and due to the curving road path. This likely increases the difficulty of deciding whether or not a gap is acceptable and may increase accepted gap size. Second, the presence of the offset increases the difficulty of negotiating the intersection, which likely increases the critical gap. When drivers must make a right turn through the minor angle of a skew intersection, their major search is to the left, which requires a more extensive head turn, and which would be expected to lengthen PRT and therefore the accepted gap. No studies were found of critical gaps at skew intersections. C-32

SUMMARY: ISD Cases C1 and C2 PRT Factors PRT MT Factors MT CRIT GAP Factors 85th Accepted GAP AASHTO Driver age, gender 2.0 sec Driver age 6.3 sec Driver age 11 sec (left) 8.0 (left) Standard transmissions Vehicle type Vehicle type Day vs. night Clutter, complexity Intersection geometry No. of lanes Right vs. left turns Grade Intersection geometry No. of lanes Right vs. left turns Grade 5.2.3 Decision Sight Distance 5.2.3.1 DEFINITION DSD Decision sight distance is the sight distance that should allow drivers to detect an unexpected or difficult- to-perceive information source or condition, recognize the condition or its potential threat, select an appropriate speed and path, and initiate and complete the maneuver safely and efficiently (Alexander & Lunenfeld, 1975). Five maneuver types are defined by AASHTO (2001): • Avoidance Maneuver A: Stop on rural road: t = 3.0 s • Avoidance Maneuver B: Stop on urban road: t = 9.1 s • Avoidance Maneuver C: Speed/path/direction change on rural road: t varies between 10.2 and 11.2 s • Avoidance Maneuver D: Speed/path/direction change on suburban road: t varies between 12.1 and 12.9 s • Avoidance Maneuver E: Speed/path/direction change on urban road: t varies between 14.0 and 14.5 s The t values enumerated above are pre-maneuver values, that is pre-braking in the case of maneuvers A and B, and pre lane-changing in the case of maneuvers C, D and E. These t values are PRT values. More PRT is allotted on urban roads than on rural roads. Urban roads generally involve higher traffic levels and greater visual complexity of the driving environment. Avoidance maneuvers A and B involve the driver recognizing the roadway or traffic situation, identifying alternative maneuvers and comfortably braking to a stop. Avoidance maneuvers C, D and E involve the driver recognizing the roadway or traffic situation, identifying alternative maneuvers and making a lane change. Lane changes are assumed to require 3.5 to 4.5 sec, with decreasing time required at increasing speeds. In this chapter we consider only avoidance maneuvers C, D and E. Avoidance maneuvers A and B will be considered in Chapter X: Intersections. [Chapter X] C-33

The decision sight distance for avoidance maneuvers C, D and E are determined as: Metric US Customary d = 0.278VtPRT+MT d = 1.47VtPRT+MT where: tPRT+MT = pre-speed/path/direction change maneuver time: 10.2 – 11.2 sec rural 12.1 – 12.9 sec suburban 14.0 – 14.5 sec urban V = design speed, km/h where: tPRT+MT = pre-speed/path/direction change maneuver time: 10.2 – 11.2 sec rural 12.1 – 12.9 sec suburban 14.0 – 14.5 sec urban V = design speed, mph In computing and measuring DSD the same 1.08 m (3.5 ft.) eye-height and 0.6 m (2.0 ft) object height are assumed as for SSD. 5.2.3.2 HIGH PRIORITY CONSIDERATIONS Examples of traffic control devices and road geometric elements which are high priority with respect to the need to apply or to consider decision sight distance so that drivers can change lanes comfortably include: • A guide sign • Lane markings indicating a change in cross-section • Overhead lane arrows • Traffic signals • The paved area of an intersection for: o First intersection in a sequence o Isolated rural intersections • A change in cross-section (2 lane to 4 lane, 4 lane to 2 lane, passing lane, climbing lane, lane drop, optional lane split, deceleration lane, channelization). The presence of visual complexity combined with any of the above elements increases the need for consideration of decision sight distance. In addition, the presence of truck traffic which can block the view of any of the above traffic control devices and road geometric elements may be compensated for by increased sight distance, which allows more opportunities for drivers to see the item of interest. 5.2.3.3 GUIDELINE FOR DSD MANUEVERS C, D & E DSD PRT MANUEVERS C, D & E With respect to roadway decision points such as lane drops, turning points or merges, PRT includes time to detect the roadway change, recognize the need to make a decision, make the decision and initiate the response. The response may include searching for a gap in traffic in order to make a lane change and/or C-34

speed reduction for turns. Where lane changes are required, PRT includes time for drivers to search for a gap in traffic. Depending on the site, there are a number of potential cues that a decision point is ahead: signs, markings, traffic patterns, parked vehicles etc. and site geometry (e.g. lane split). The physical feature that the driver must respond to is the gore at the lane drop or split. Signs generally are visible first, followed by markings, and then the physical gore itself. ******************************************************* Figure showing favorable and unfavorable conditions for DSD Animation illustrating examples of baseline and unfavorable conditions, including high (urban expressway, closely spaced exits and multiple guide signs at night, with dense but free-flow traffic) and low workload (rural highway, daytime, little traffic) situations. *************************************************************** C-35

DECISION SIGHT DISTANCE: PERCEPTION REACTION TIME – MANUEVERS C, D AND E GUIDELINES Under baseline conditions it can be assumed that the driver is responding to either signs or markings. Under baseline conditions most drivers (85%) will be able to determine that there is a decision point ahead, locate a suitable gap in traffic and initiate a lane change within a PRT of 7.8 sec. In baseline conditions PRT can be measured from the point at which the markings for the gore or turn lane are first visible at night. Baseline conditions assume: • Visually uncluttered environment • Conspicuous, easily understood signs, placed overhead or on the right • Conspicuous markings accompanied by PRPMs in the gore area for visibility in rain • Minimal view blockage of signs, markings and gore due to traffic • Unfamiliar driver Unfavorable conditions assume: • Poor marking and signing • Deceptive appearance of site • Features that violate driver expectancies (e.g. freeway left exit, add-drop lane, first signalized intersection) Under these conditions it is assumed that some or many drivers will miss the sign and marking cues and respond to the last available cue, which is the physical gore. Most drivers (85%) will be able to detect the decision point, locate a suitable gap in traffic and initiate a lane change within a PRT of 20 seconds measured from the physical gore. Situational variables that may affect PRT are: • High driver workload due to concurrent tasks (e.g. traffic merging, presence of guide, warning or regulatory signs unrelated to the lane drop) • Dense traffic • Truck traffic which intermittently blocks the view • Off roadway clutter which can distract drivers • Poor weather which increases driver workload and makes cues (especially markings) less conspicuous These variables can contribute to delayed recognition of signs, markings and the presence of the physical gore. In the worst case, the gore will be the cue, and drivers’ response will be sufficiently delayed that they are unable to complete a lane change before reaching it. DSD PRT has not yet been measured under all these conditions. The worst 85th percentile value PRT found in experimental studies at a poorly marked freeway site was 23 seconds. C-36

5.2.3.4 BASIS/RATIONALE FOR DSD AVOIDANCE MANEUVERS C, D & E GUIDELINE DSD PRT and MT are difficult to define exactly because drivers may respond to one of several cues (signs, markings or site geometry), and may find it difficult to report recognizing the situation before they start responding to it. Furthermore, studies of DSD have examined what drivers do in specific situations, but without determining whether the driver was able to respond comfortably, or was able to respond just in time. The longer the sight distance drivers have available, generally the longer their PRTs and MTs will be, because there is a lack of urgency. With shorter and shorter sight distances, drivers will respond more quickly, getting to the point where the lane change in no longer comfortable, and some drivers do not make it in time. This point has yet to be determined in an on-road study. Key studies of DSD include: • Measurement of PRT and MT for 19 drivers (5 aged 16-39, 12 aged 40-59, and 3 aged 60 or older) at 6 freeway sites, including 4 lane drop exits, a mainline lane drop, a lane split, as well as at 2 sites with lane reductions prior to intersections (McGee, Moore, Knapp, & Sanders, 1978) DECISION SIGHT DISTANCE: MANEUVER TIME – MANUEVERS C, D AND E GUIDELINES Under baseline conditions: When MT was measured from the gore, only for those drivers who did not start to respond until the gore was visible, 85th percentile MTs were 6.4 sec overall (urban and freeway sites combined), with longer MTs by 1.2 – 2.0 sec for urban sites. • A single lane change Data suggests that maneuver times are 0.5 sec longer for left lane as compared to right lane merges. Under unfavorable conditions: MT should be increased by 5 sec for each additional lane change in light traffic, and 7.2 sec for each additional lane change required in moderate or heavy traffic (726+ vehicles per lane per hour). • More than one lane change • Dense traffic MT can be decreased by the following driver factors: • Age • Urgency The closer to the physical gore at the point at which the driver realizes the need for a lane change, the more quickly it will be accomplished. In some situations older drivers have faster MT’s to compensate for delayed recognition of cues such as signs and markings. C-37

• Measurement of PRT and MT for 98 drivers (28 aged 20 – 40, 35 aged 65 – 69 and 35 aged 70+) at 11 sites: 2 freeway lane drop exits, 3 mainline lane drops, 4 arterial turn lanes, one arterial lane drop due to parking, and one complex intersection (Lerner et al., 1995). There is an important methodological difference between these studies which affects their interpretation. In the McGee et al. (1978) study, detection and recognition time was measured between the point that the physical gore was in view to the experimenter and the point that drivers indicated that they needed to change lanes or reduce speed to stay on course. Decision and response time was measured as the time elapsed between the driver recognition of the need to change lanes and the maneuver being initiated. In approximately half the trials, signs and markings allowed the driver to start responding before the physical gore was in view, leading to PRTs of 0 in response to the gore; these data were excluded from calculations of PRT and maneuver time. In contrast to the methodology of McGee et al., the Lerner et al. (1995) study first established the point at which an experimenter could first sight each potential cue (sign, marking and gore). Then PRT was measured from this point to the point at which drivers reported sighting of the first cue (sign, marking or gore) to the upcoming lane drop or arterial turn lane. Thus PRT is based more precisely on the particular cue used by the driver, but is not necessarily related to when the gore or decision point was visible. DSD PRT MANUEVERS C, D AND E Based only on subjects (approximately 50%) who signaled their detection of the need to change lanes after the gore became visible, McGee at al. (1978) found that mean PRT (detection and recognition plus decision and response initiation time) was 10.5 sec. Based on available information, the 85th percentile PRT value is estimated to be about 20 sec. Since the McGee et al. data are based only on drivers who identified the need to change lanes after the gore was visible, they are based on PRT with respect to the physical gore. Since only 50% of drivers responded this late, the mean value of 10.5 sec actually encompasses the majority (about 75%) of subjects. The site with the longest PRT, with an 85th percentile value of 22.5 sec, was a lane drop exit with poor marking and signing. At this site one-third of drivers did not realize they were in an exit lane and drove on the shoulder until they realized they had passed the actual gore. Despite the fact that the available DSD was equivalent to 24 sec., because of the poor marking and signing, this DSD was insufficient to comfortably allow PRT and MT. Lerner et al. (1995) found that at most sites, most drivers responded to signs. At freeway sites, the posted speed was 88 km/h (55 mph) and the signs were placed 214 m (687 ft) to 1600 m (5133 ft) from the physical gore. At the arterial sites, the posted speed was 64 km/h (40 mph) and signs were placed 92 m (295 ft) to 229 m (734 ft) from the intersection. At 3 out of 6 arterial sites (2 arterial turn lanes and 1 complex intersection) drivers responded only to markings. At 1 of the 5 freeway lane drop sites only 1/3 of the drivers responded to signs; the rest responded to markings. At this site, in contrast to the other freeway lane drops, the signs were placed on the left. Driver PRT depends in part on urgency. The longest PRTs were for the sites with signs farthest in advance of the lane drop (458 m [1500 ft] to 1600 m [5250 ft]), and the shortest for sites with signs closer to the lane drop (305 m [1000 ft], 214 m [700 ft]). Similarly PRTs were longer in daytime when cues could be seen further away as compared to at night. PRTs were also longer for young as compared to older drivers. This may have been because younger drivers were more likely than older drivers to use sign C-38

cues which are visible from a greater distance, as well as being further from the physical gore than are markings, allowing more time for the response. At freeway sites, the 85th percentile values for daytime PRTs were similar for all three age groups (7.8 for 20 – 40 year olds, 7.6 for 65 – 69 year olds, and 7.8 for 70+ year olds). At arterial sites, the 85th percentile values for daytime PRTs were considerably shorter for the younger group (4.2 sec) than for the older groups (7.6 and 7.1 sec). These values are all shorter than those measured by McGee et al. (1978). The reason is that the Lerner et al. values were measured in response to the cue actually used by the driver (sign, marking or gore) and not assumed to be in response to the gore. Sign and marking cues were visible before the gore. Baseline Conditions In baseline conditions, where signs and markings are conspicuous and easily understood, an 85th percentile PRT value of 7.8 sec is selected based on the longest PRTs, i.e., daytime values for freeway conditions for the oldest and youngest age groups (7.8 for 20 – 40 year olds, and 7.8 for 70+ year olds) in the Lerner et al. study. Most of these PRTs are relative to sign placement, however at some sites a large number of drivers used marking cues. Therefore a conservative approach is proposed whereby PRT is based on the visibility of the marking, rather than the sign cue. Unfavorable Conditions In unfavorable conditions, where signs and markings are inadequate (e.g. poor reflectivity) and the driver responds to the appearance of the physical gore, an 85th percentile PRT value of 15.4 sec is selected based on the 85th percentile values from approximately half of the McGee et al. subjects, who did not indicate detection of the upcoming lane change requirement until the physical gore was visible. For unfavorable conditions it is proposed that the PRT be based on the visibility of the physical gore. DSD MT Like PRT, driver MT depends in part on urgency. Lerner et al. (1995) found that the shortest MTs were at an arterial site where there were no marking or sign cues to the lane drop – only the presence of parked vehicles. MTs, like PRTs, were longer in daytime when cues could be seen further away as compared to at night. Unlike the case with PRTs, MTs were longer for older as compared to younger drivers. Lerner et al. (1995) do not provide 85th percentile values for MT, but rather for the combination of PRT and MT. Times were longer for daytime than for nighttime. If PRT values are subtracted from the combination, then at freeway sites, 85th percentile daytime MTs were 8.7 sec for younger subjects and 10 and 11 sec for the two older groups. At the arterial sites, the younger drivers also had shorter 85th percentile total time. If PRT values are subtracted from the combination, then the 85th percentile values for daytime MTs were 9.9 sec for the younger group, and 8.6 and 8.9 sec for the two older groups. The shorter maneuver times for the older groups (by 1 to 1.3 sec) may indicate a greater urgency by the time the maneuver was made. It must be noted that these times are not related to the distance from gore, but rather from the first cue identified by the driver, which might have been a sign, markings or in some cases the gore itself. Based on McGee et al., 1978, the overall mean for maneuver time at freeway and arterial sites was 4.6 sec (st.dev. 1.7 sec). Based on the standard deviation, the 85th percentile MT was 6.4 sec. MTs were longer at lower speeds, and longest at the two urban intersection sites, where means were 5.8 and 6.6 sec, 1.2 – 2.0 C-39

sec longer than the overall mean. MTs were measured separately for left (4.7 sec) and right (4.2 sec) lane merges at a lane split. The longer maneuver times measured by Lerner et al. may reflect less urgency than was the case for the McGee et al. subjects. This is because the only data used by McGee et al. to calculate MT are for subjects who did not indicate a need to respond until the gore was visible. This was likely many seconds after the signs and markings were visible. A study by McNees of lane changing distances indicates that the time and distance required to move from the left to the right side of a multi-lane highways can be considerable (McNees, 1982). The amount of time and distance was recorded for 20 subjects driving an instrumented vehicle who maneuvered from the far left lane to the far right lane on three and four lane freeways in light (725 vehicles per hour or less), medium (726 – 1225 vehicles per hour) and heavy (> 1225 vehicles per hour) traffic. Subjects were asked to keep to the posted speed limit which was 88 km/h (55 mph). Distance was calculated according to the speed traveled and the time taken from signaling to turn from the left-most lane until all four wheels had crossed into the right-most lane. The results are shown in the table below: Three lane maneuver Four lane maneuver Traffic Condition N Mean Distance m – (ft) 85th percentile distance m – (ft) N Mean Distance m - ft 85th percentile distance m – (ft) Light 56 282 (925) 367 (1204) 48 367 (1204) 488 (1600) Medium 56 307 (1007) 405 (1329) 57 464 (1522) 587 (1925) Heavy 59 305 (1001) 472 (1549) 63 419 (1375) 538 (1765) The longest distance was required for 3 lane changes (4 lane maneuver) in moderate traffic. In heavy traffic, speeds were lower (13-27 km/h [8-17 mph]), so although more time was required, the distance required was less. Assuming that subjects were traveling at the speed limit, and in light traffic, the 85th percentile time to complete three lane changes as compared to two lane changes was an additional 5 sec., and, in medium density traffic, 7.4 sec. SUMMARY: DSD Avoidance Maneuvers C, D & E PRT Factors PRT MT Factors MT AASHTO Driver workload 7.6-7.8 sec # lane changes 6.4 sec measured from gore 10.2 11.2 sec rural measured from gore Urgency – sign, marking or physical gore as cue Poor visibility Measured from point sign or marking is visible Required Left lane vs. right lane merges Urban vs. freeway Age urgency 12.1-12.9 sec sub-urban measured from gore 14.0-14.5 sec urban Expectancy violation 20 sec measured from gore measured from gore C-40

5.2.4 Passing Sight Distance 5.2.4.1 DEFINITION Passing sight distance is the length of the highway ahead necessary for one vehicle to pass another before meeting an opposing vehicle that might appear after the pass begins (ITS Traffic Engineering Handbook, Pline, 1999). The AASHTO model is based on field studies conducted before 1958 (Hassan, Easa, & Abd El Halim, 1996) and assumes that once drivers begin to pass, they have no opportunity to abort it. The MUTCD guidelines for markings, on the other hand, assume that drivers can abort the pass, and the assumed required passing sight distance is much shorter. PSD includes four components [AASHTO, 2001, CH3]: • d1, which is traversed during PRT and during the interval when the driver brings the vehicle from the trailing speed to the point of encroachment of the passing lane • d2, which is traversed while the passing vehicle occupies the passing lane • d3, which is the distance between the passing vehicle at the end of its maneuver and the opposing vehicle • d4, which is traversed by the opposing vehicle for two-thirds of the time the passing vehicle occupies the passing lane (i.e., 2/3 of d2) The distances for d1 and d2 are defined in AASHTO (2001) as shown below: Metric US Customary ati d1 = 0.278ti v – m + 2 ati d1 = 1.47ti v – m + 2 where: ti = time of initial maneuver, s a = average acceleration, km/h/s v = average speed of passing vehicle, km/h m = difference in speed of passed vehicle and passing vehicle, km/h where: ti = time of initial maneuver, s a = average acceleration, mph/s v = average speed of passing vehicle, mph m = difference in speed of passed vehicle and passing vehicle, mph Metric US Customary d2 = 0.278vt2 d2 = 1.47vt2 where: t2 = time passing vehicle occupies the left lane, s v = average speed of passing vehicle, km/h where: t2 = time passing vehicle occupies the left lane, s v = average speed of passing vehicle, mph C-41

5.2.4.2 RELATED DESIGN/OPERATIONAL ISSUE Required passing sight distance relates to vehicle characteristics, road grade, and vehicle speeds. It also relates to whether or not the pass is aborted. Required passing sight distance is shorter if consideration is given to the possibility that the pass can be aborted. When passes are aborted, shoulder characteristics are important, since one consequence, is a loss of control due to encountering a pavement edge drop-off. 5.2.4.3 GUIDELINE Just as a gap acceptance model is used for describing driver behavior for crossing intersections, such a model could be used for describing passing behavior. However studies have not yet been conducted using this approach. Consequently, PSD is considered with respect to PRT and MT, which assumes a serial process. PASSING SIGHT DISTANCE: PERCEPTION REACTION TIME GUIDELINES Mean PRTs to initiate a pass, and measured from when passing sight distance was available until when the right tire crossed the centerline, have been found to vary from 3.6 to 6.0 sec., depending on the particular site on two lane rural highways. No information is available on subject variability, but 85th percentile PRTs will certainly exceed mean PRTs. Just as ISD PRT is affected by age, gender, standard transmissions and day versus night conditions, PSD PRT may be as well. However no studies were found on this issue. PASSING SIGHT DISTANCE: MANEUVER TTIME GUIDELINES Under baseline conditions: • Passenger vehicle passing single passenger vehicle the 50th percentile time required for passing, and measured from when the right front tire crossed the centerline until the right rear tire crossed back into the driver’s lane, was found to be 5.2 to 7.3 sec. depending on the site, with the longest time, by 1.3 sec, found for the site with a 7% grade. In a study where the 50th percentile time required for passing was measured from when the left front tire crossed the centerline until the left rear tire crossed back into the driver’s lane, values ranged from 13 to 14.5 sec. 85th percentile times would exceed 50th percentile times. Under unfavorable conditions: • Passenger vehicle passing multiple vehicles • Passenger vehicle passing truck • Truck passing other vehicle • Passing occurring on an upgrade the time required for passing, once PRT is completed will be longer. MTs may be increased with driver age, however no data were found on this issue. C-42

5.2.4.4 BASIS/RATIONALE FOR GUIDELINE The AASHTO model for PSD is based on data for single passenger vehicles passing single passenger vehicles. It is further based on the assumption that once drivers begin to pass, they have no opportunity to abort the pass. The MUTCD guidelines for markings, on the other hand, assume that drivers can abort the pass, and the assumed required passing sight distance is much shorter (e.g. AASHTO (2000) Exhibit 3-6 indicates that when the speed of the passing vehicle is 40 km/h that the total passing sight distance is 160 m., compared to the MUTCD (2003) which indicates 140 m.). The discrepancy is much higher at higher speeds; when the speed of the passing vehicle is 120 km/h, the total passing sight distance calculated is 915 m as compared to the MUTCD (2000) minimum of 395 m. As reported in the Older Driver Design Guidelines (www.tfhrc.gov/humanfac/01103, “Weaver and Glennon (1972) reported that, in limited studies of short passing sections on main rural highways, most drivers do not complete a pass even within a 244 m. (800 ft) section; and the use of passing zones remains very low when their length is shorter than 274 m (900 ft).” (Weaver & Glennon, 1972). A concern about the marked end of the passing zone is anecdotal evidence from a workshop on traffic safety (Smiley, 2004) that indicates drivers are uncertain about whether this is the last point at which a pass can be started or the point at which passes must be completed. The MUTCD assumes that the change from the dashed to solid means “do not start a pass and get back into the right lane if you are in the left lane.” This may be another reason contributing to the low use of passing sections. A number of models of PSD have been developed that consider PSD requirements from the point of view of the minimum sight distance required at the critical point, namely that point where a driver requires as much sight distance to safely abort the maneuver as to complete it. Depending on the exact model assumptions, that point occurs when the two vehicles are abreast of one another A revised model which better matches field observations has been developed by Hassan et al. (1996). It should be noted that head-on crashes related to passing maneuvers, though serious, are rare. Only 4.6% of head-on fatalities are related to passing. It is interesting however that the majority occur in marked passing lanes (Federal Highway Administration, 1994), suggesting that the discrepancy between AASHTO and the MUTCD as well as an understanding of the nature of passing zone crashes requires attention. PASSING SIGHT DISTANCE: GAP TIME GUIDELINES Based on one study at five sites, the average passing time gaps accepted ranged from 15.7 to 22.4 sec, increasing linearly with available passing sight distance. All passes were made in the absence of oncoming traffic. Based on two studies, there is no relation between time spent in the opposing lane and passing sight distance. The linear relationship between average passing time gap and available sight distance is due to longer time margins at the end of the pass as available passing sight distance increases. These ranged from 4 sec (284 m or 929 ft.) to 10 sec (416 m or 1363 ft or longer). Limited passing opportunities may influence driver decision criteria. Drivers may accept smaller gaps and compensate with higher passing speeds, which could lead to vehicle control problems. Drivers have difficulty accurately judging the speed of approaching vehicles. Poor gap acceptance decisions related to misjudging high-speed vehicles is not a sight distance problem and may not be improved by increased sight distance. C-43

The primary cue that a driver uses to determine whether or not it is safe to initiate a pass is the size of the image of the oncoming vehicle. In a series of experiments on a road not open to the public, Farber and Silver examined judgments in an overtaking situation (Farber & Silver, 1967). They found that drivers could make reasonable estimates of the distance of an oncoming car but not of its speed. Judgments of distance were accurate within a 20% error or less, 95% of the time. Judgments of speed were much poorer. At the extremes of oncoming car speed used, the passing distance at 96 km/h (60 mph) was actually less than that at 48 km/h (30 mph), indicating that subjects were not at all able to discriminate between these extreme speeds. Staplin et al. found that this may be a more pronounced problem for older drivers (Staplin, Lococo, & Sim, 1993). In a field study where drivers indicated whether or not they would accept a gap for the purposes of turning left and the speed of the oncoming vehicle was 48 km/h or 96.5 km/h (30 mph and 60 mph), older drivers accepted gaps based on the distance at which the vehicle was seen rather than its speed. In contrast younger drivers accepted a gap that was 25 percent larger for the higher speed vehicle. Drivers’ inaccurate estimates cannot be compensated for by increasing sight distance, but the difficulty of speed perception [Section 4.6] can explain some crashes. Large vehicles may be especially susceptible to misjudgment. Crashes due to underestimating the available maneuver time when there is a high-speed approaching vehicle may be addressed through speed control measures or site factors that improve speed judgments [Chapter X, Speed]; it should not be assumed that greater sight distance will address this problem. Drivers who pass may approach a slower vehicle and pass immediately (a flying pass), or may adopt a short headway and wait for an opportunity (a delayed pass). In the second case, more time for acceleration is required. In either case drivers may adopt a short headway just prior to the pass. A study on two-lane highways found that 40% of drivers following at short headways (1/2 sec or less) were doing so in anticipation of passing (Rajalin, Hassel, & Summala, 1997). PSD PRT A single study was found of PRT in the passing situation (Hostetter & Seguin, 1969). This study involved five sites on a two-lane highway in Pennsylvania and observations of 1462 passes. Subjects were not aware that their behavior was being measured. Impedance by an experimental vehicle was established prior to the subject vehicle entering the passing zone. Available sight distance varied from 283 m to 497 m (930 to 1630 ft). Subject drivers were impeded over distances of 1, 3 and 5 miles, by an experimental vehicle which traveled at 10, 20 or 30% of the subject vehicle’s previously measured speed . Traffic volumes varied from 16 to 86 vehicle per hour. Observations of judgment time (time elapsed from availability of passing sight distance to front left wheels crossing the center line) were made. Opposing traffic was stopped out of view of the subject driver so that no opposing traffic was present during the passes. Mean judgment time was reported for each of five sites and varied from 3.6 to 6.0 sec. Standard deviations were not reported, however, based on studies of PRT in other situations, the 85th percentile PRT values would be expected to be on the order of 50% longer. Just as ISD PRT is affected by age, gender, standard transmissions and day versus night conditions, PSD PRT may be as well. However no studies were found on this issue. PSD MT In the Hostetter and Seguin (1969) study cited above, movement time was measured from the point at which the right front tire of the subject vehicle crossed the center line to the point at which the right front tire of the subject vehicle crossed the center line back into the lane. Mean movement times are reported for each of the five sites and varied from 5.2 to 7.3 seconds. There was not a linear relationship with sight distance. The longest value, by 1.3 sec, was found for the site that had an approach gradient of 7% and a slight upgrade over the entire passing zone. C-44

In a study at five sites on a recreational two-lane highway in Wisconsin, Kaub (1990) used field observers to record time in the opposing lane and type of pass (Kaub, 1990). Five types of passes were recorded: pass with no opposition (i.e. no opposition at the so-called “critical position” alongside the passed vehicle where the driver is assumed to make a pass/abort decision), pass with opposition: greater than 10 sec (i.e. at the point at the driver returned to his or her own lane there was greater than a 10 sec gap to the oncoming vehicle), between 5 and 10 sec, or less than 5 sec., pass with full abort, and multiple pass. Passing zone lengths varied from 549 m (1800 ft) to 2012 m (6600 ft) in length. Operating speed was approximately 96 km/h (60 mph). Observers recorded the time elapsed between the crossing of the centerline by the passing vehicle’s left front tire and the return of the vehicle’s left rear tire to the lane of origin, in other words to the first moment the opposing lane was encroached until the last. It should be noted that this definition of MT is different than that used by Hostetter and Seguin (1969), who measured from the crossing of the right tire. Given the definitions of MT, the MTs measured by Hostetter and Seguin (1969) would be expected to be a few seconds shorter than those measured by Kaub (1990), to allow for the time taken between the right and left tire crossing the centerline. A total of 4153 passing maneuvers were observed. Under low traffic volumes (200-250 vehicles/hr in the major direction and 85 to 175 vehicles/hr in the minor direction), 65-75% of passes were attempted in the face of opposing traffic, 25-35% of passes were attempted in the presence of oncoming traffic, and 0.8% of passes were aborted. At high volumes (330- 420 vehicles/hr in the major direction and 70 to 170 vehicles/hr in the minor direction), 51 to 76% of passes were made with no opposition, 26 – 50% of passes were in the presence of oncoming traffic and, 7.2 % of passes were aborted . The average time in the opposing lane was 12.2 sec under low-traffic conditions and 11.3 sec with high traffic volumes. No standard deviations were provided. Depending on site and direction, times varied from a low of 7.98 sec to a high of 12.87 sec. There was no clear association between length of available passing lane and time spent in the opposing lane. At a speed of 96 km/h (60 mph) the average times in the opposing lane are equivalent to distances of 325 m (1064 ft) for low-traffic and 301 m (986 ft.) for high traffic. This may be the reason for Weaver and Glennon’s (1972) observations that passing zones shorter than 274 m (900 ft) were seldom used. Length of time spent in the passing lane was clearly related to the size of the time gap. Drivers returning to their own lane with more than 10 sec to spare averaged 12 sec in the opposing lane. Drivers returning with 5 to 10 sec to spare, averaged 8.7 sec and those with less than 5 sec to spare, 6.8 sec. With respect to differences between older and younger drivers, studies find that preferred speed decreases and preferred headway increases with age (Evans & Wasielewski, 1983). Similarly accepted gaps in turning situations increase with age by about 1.2 sec (Lerner et al. 1995). Although no studies are available, it seems likely that passing time requirements for older drivers will be longer by virtue of both lower speeds and more conservative gap acceptance behavior. It also seems likely that older drivers are more likely to be driving the passed as opposed to the passing vehicle. The time from when the vehicle wheels first encroach the opposing lane and ends when they last do so. Since drivers cannot accelerate until they enter the opposing lane, this definition of MT encompasses almost the entire maneuver. On this basis MT can be assumed to average 12.2 sec under lower volume situations (major flow 200-250 vph, minor flow 85-175 vph) and 11.3 sec under higher volume situations (major flow 330-420 vehicles/hr, minor flow 70 to 170 vehicles/hr). These are average values. Kaub does not report standard deviations which would allow 85th percentile values to be determined. C-45

Multiple passes were found to occur during 6.4 to 21.4% of passes, depending on the direction and on the site. The likelihood of a multiple pass did not appear to be related to the length of the passing zone. In higher flow conditions, time in the opposing lane averaged 0.9 sec less, and number of passes completed with the minimum safety margin of 5 sec or less increased from 6.3% to 9.2% Aborted passes increased from 0.75% to 7.2%. MTs related to multiple passes and trucks will be longer than the times reported by Kaub (1990) which applied to single impeding passenger vehicles passed by other passenger vehicles. Drivers do not typically accelerate at the maximum level their vehicles are capable of. Whether drivers accelerate closer to the maximum level to compensate in situations where geometric design factors slow acceleration is unknown. MT’s may be longer in these situations. While driver factors would be expected to include age, given older driver preferences for lower speeds, they are more likely to be in the passed rather than the passing vehicle. However, as the population ages, increasingly older drivers will be passed by other older drivers who are likely to require longer MTs. No studies were found on this issue. PSD TIME GAP The Hostetter and Seguin (1969) study provided a measure of desired gap, in that the time safety margin when the pass was completed was also measured. The average passing time gaps accepted ranged from 15.7 to 22.4 sec, increasing linearly with available passing sight distance. All passes were made in the absence of oncoming traffic. Neither this study nor the Kaub (1990) study found any relation between time spent in the opposing lane and passing sight distance. The linear relationship between average passing time gap and available sight distance is due to longer time margins at the end of the pass as available passing sight distance increases. These safety time margins ranged from 4 sec (for passing sight distance of 284 m or 929 ft.) to 10 sec (for passing sight distance of 416 m or 1363 ft or longer). SUMMARY: PSD PRT Factors Average PRT MT Factors Average MT Average gap AASHTO Site geometry 3.6 – 6.0 sec Site geometry 5,2-7.3 sec measured from right tire in lane 15.7 – 22.4 sec 14.4 sec at 40 km/h 27.5 sec at 120 km/h 13 – 14 sec measured from left tire in lane C-46

5.3 Influence of Design on Speed 5.3.1 Background The design of a road affects drivers’ speeds through two major mechanisms. First, the design creates the driving task. Narrow lanes and sharp curves make the driving task more difficult and lead to reductions in speed. Secondly, drivers have expectations about the posted, and comfortable speeds, based on various combinations of design elements. Users of this guide should be aware that operating speeds may be very different from posted speed when the road message and the posted speed are at variance. Thus design sight distances may be more appropriately determined based on operating, not posted speed. Design elements that influence speed include the following: • Lane width • Alignment (horizontal and vertical) • Road Surface • Side Friction • Shoulder width 5.3.2 Scope This section is intended to address road features that influence driver speed choice and therefore impact required sight distances. There are a number of engineering studies which have used speed measurements made at numerous sites to develop models to predict speed based on road design. While it is not the intent of this chapter to critique these studies in detail, it does give an idea of the degree to which design features can affect driver speed choice. Operational features such as speed limit signs, lateral lane markings, post mounted delineators etc. may also affect speed. These are discussed in a later chapter on speed management [Chapter X, Speed]. For a more fundamental understanding of driver perception of speed the reader is referred to Chapter 4. [Section 4.6] 5.3.3 Speed and Lane Width 5.3.3.1 GUIDELINE: SPEED AND LANE WIDTH 5.3.3.2 BASIS/RATIONALE FOR SPEED AND LANE WIDTH Lane width influences speed because it influences the difficulty of the driving task. Narrower lanes require more frequent, smaller steering corrections (McLean & Hoffman, 1972), that is, more effort. Slowing down reduces the effort required. In a 1990 report entitled “Behavioral Adaptations to Changes in the Road Transport System”, an OECD Scientific Experts Group (OECD, 1990) reviewed impacts of lane width on driver behavior. Researchers consistently found a reduction in speed with decreases in lane width and vice versa. SPEED AND LANE WIDTH Increasing lane width from 3.3 to 3.8 m. is associated with an increase of 2.85 km/h in speed on high design standard two lane rural highways. C-47

A study of the effects of various geometric and environmental factors on the speeds for 2-lane rural highways (Yagar & Van Aerde, 1983) collected data for over 5000 5-minute periods at 35 locations. The most influential factors, in order of significance were as follows: legal speed limit, land use adjacent to the road, grade, access from other roads and lane width. Road curvature, the presence of an extra lane, sight distance, center line markings and lateral obstructions were not found to have a significant effect on speed. The lack of effect of some of these variables, especially road curvature, may be due to the fact that the roads examined had high design standards – gradients were less than 3% and radii of curvature were more than 1400 m. Other studies have found strong impacts of road curvature on speed, but that for curvature over 800 m. speeds on curves are essentially the same as those on tangents (Fitzpatrick, Carlson, Wooldridge, & Brewer, 2000). Increasing lane width over a range from 3.3 to 3.8 m. was associated with an increase of 2.85 km/h in speed (Yagar and Van Aerde, 1983). The finding of very modest changes in speed associated with lane width is corroborated by more recent work (Fitzpatrick et al. 2000). Although Yagar and Van Aerde found the legal speed limit had a strong effect, it must be remembered that legal speed limit is strongly associated with road design, and therefore the legal speed limit is going to be correlated with the presence of a specific bundle of road features. Another study by Parker looked at the effect of changing speed limits at 98 sites in 22 U.S. states (Parker, 1997). Depending on the site, speed limits were raised as much as 15 mph and lowered as much as 20 mph. At these sites it is important to note that the only change that was made was the speed limit sign. No other engineering or enforcement changes were made. The results showed minimal changes in speed. Furthermore the direction of the changes that did occur were not necessarily in the same direction as the speed change. 5.3.4 Speed and Alignment 5.3.4.1 GUIDELINE: SPEED AND ALIGNMENT 5.3.4.2 BASIS/RATIONALE FOR SPEED AND ALIGNMENT Speed is strongly related to radius of curvature. Lamm and Choueiri and Krammes et al. (1995) developed models predicting speed based on radius, deflection angle and curve length. These models account for more than 80% of the variance in speed. In a study of speeds in 176 curves on rural 2 lane highways with posted speeds of 75 – 115 km/h, Fitzpatrick et al. (2000) found that the 85th percentile velocity was most strongly related to radius, and related, but less so, to grade and sight distance (R values .58 to .92). Once the curve radius exceeded 800 m., curves had similar speeds to tangents. Speed on tangents is much more difficult to predict and is dependent on a wide array of road characteristics such as tangent length, radius of curve before and after the section, cross-section, grade, SPEED AND ALIGNMENT Speed on curves can be reasonably accurately predicted using models based on radius, curve deflection angle and curve length. Once the curve radius exceeds 800 m., curves have similar speeds to tangents. Speed on tangents is much more difficult to predict and is dependent on a wide array of road characteristics such as tangent length, radius of curve before and after the section, cross-section, grade, general terrain and sight distance. Posted speed is a better predictor of speed on urban arterial tangents than it is on highway tangents. C-48

general terrain and sight distance. Models to predict operating speeds on tangent sections of two-lane rural highways were developed by Polus et al. (1999) based on speed measures at 162 sites with posted speeds varying from 75 to 115 km/h ( equivalent to 50 to 70 mph). Models were developed for various combinations of radii and tangent length, and predicted between 20 and 84% of the variance. Studies on urban arterials find posted speed limit predicts 53% of the variance in speed. 5.3.5 Speed and Pavement Surface 5.3.5.1 GUIDELINE: SPEED AND PAVEMENT SURFACE 5.3.5.2 BASIS/RATIONALE FOR SPEED AND PAVEMENT SURFACE One of the cues drivers use to estimate their own speed is noise level. Evans (1970) showed that when sound cues were removed through the use of earmuffs, drivers underestimated their actual speeds by 6 to 10 km/hr. Cooper (1980) showed that re-surfacing a road resulted in a speed increase of 2 km/h. 5.3.6 Speed and Side Friction 5.3.6.1 GUIDELINE: SPEED AND SIDE FRICTION 5.3.6.2 BASIS/RATIONALE FOR SPEED AND SIDE FRICTION Side friction refers to elements close to the edge of the lane such as pedestrians, bicyclists, parked vehicles, foliage, etc., which can strongly affect speed. This is because one of the major cues used by drivers is the streaming of information in peripheral vision. Side friction increases the stimulus in peripheral vision. In one study, drivers were asked to drive at 60 mph (96 km/h) with the speedometer covered. In an open-road situation, the average speed was 57 mph (91 km/h). After the same instructions, but along a tree-lined route, the average speed was 53 mph (85 km/h) (Shinar, McDowell, & Rockwell, 1977). The trees, close by, provided peripheral stimulation, giving a sense of higher speed. The elements that create side friction, such as pedestrians, bicyclists, parked vehicles and landscaping also present various levels of hazard, likely influencing drivers to slow down to various degrees. In other words pedestrian presence close to the road edge is more likely to impact speed than landscaping close to the road edge. SPEED AND PAVEMENT SURFACE Re-surfacing is associated with no or small increases in speed. SPEED AND SIDE FRICTION Elements close to the edge of the lane contribute to a reduction in driver speed. Results of one study of road sections posted at 50 km/h (31 mph) showed that 85th percentile speeds were 12 km/h (7.5 mph) lower in road sections with side friction due to the presence of pedestrians, bicyclists, parked vehicles etc. C-49

Results of one study of 30 road sections posted at 50 km/h show that 85th percentile speeds were 62 km/h in road sections with little side friction (that is wide clear zones), but were 50 km/h in road sections with side friction due to the presence of pedestrians, bicyclists, parked vehicles etc. (Transport Canada, 1997). 5.4 Diagnosing Sight Distance Problems The foregoing sections of this chapter have provided explicit guideline statements regarding human factors aspects of various sight distance concepts. However, for users to implement these guidelines in a practical sense, it is desirable to provide a procedure for their operational application. Therefore, this section comprises a hands-on tool whereby practitioners apply human factors techniques to analyze sight distance requirements at a selected highway location. A starting point for development of the current procedure was a review of previously documented procedures for conducting on-site driving task analyses [Ontario Traffic Manual, Appendix C, Positive Guidance Tool Kit] that applied techniques such as commentary drive-thru procedures to generate check-list subjective scaled ratings of hazard severity and information load. The current in-situ sight distance diagnostic procedure includes application of previously available engineering tools, e.g., AASHTO analyses of geometric requirements and MUTCD traffic control device requirements, and augments these techniques with those sight distance concepts presented in Section 5.2 and 5.3 herein. This sight distance diagnostic procedure consists of a systematic on-site investigation technique to evaluate the highway environment to support the concepts of interest, i.e., SSD, PSD, ISD, and DSD. The highway location is surveyed, diagrammed, and divided into component sections based on specific driving demands (e.g., requirement to perform a maneuver). Then each section is analyzed in terms of its suitability to support the required task (e.g., information provided to driver, allotted time to the complete required task). This procedure enables the practitioner to compare the available sight distance with the required sight distance to safely perform the driving task. Appendix A provides an example application of the procedure described in this section. 5.4.1 The Six-step Process The procedure consists of six steps as follows: 1. Collect field data to describe roadway characteristics and other environmental factors affecting sight distance requirements and driver perception of a potential hazard. 2. Conduct engineering analyses applying traditional techniques, e.g., AASHTO design criteria and MUTCD compliance, to initially assess site characteristics or deficiencies. 3. Examine accident data and prepare collision diagram to seek possible association between safety and a sight distance problem. 4. Establish component roadway sections in which drivers respond to specific visual cues in order to avoid a hazard to initiate a maneuver. 5. Analyze driving task requirements (PRT and MT) and determine the adequacy of each component roadway section to support these requirements. 6. Develop engineering strategies for amelioration of sight distance deficiencies. A flow diagram overview of the process is shown on the next page. C-50

Step 1. Collect Field Data Prepare site diagram Collect speed data Observe erratic vehicle maneuvers Inventory traffic control devices Measure geometric sight distances Record features affecting flow speeds Record visual distractions at hazard and approach Add specified labels to site diagram Step 2. Conduct Preliminary Engineering Analyses Evaluate site in terms of AASHTO design criteria Evaluate site in terms of AASHTO DSD warrants Evaluate traffic control devices in terms of MUTCD criteria Step 3. Apply Accident Data Examine spatial distribution of accident types Assess suitability of accident sample Examine potential sight distance causation Step 4. Establish Sight Distance Roadway Segments Establish and plot driver action requirements Plot information sources and associated sight distance Step 5. Analyze Component Driving Task Requirements Determine relevant sight distance application Determine driving task requirements Quantify the applicable PRT and MT requirements Assess sight distance adequacy Step 6. Develop Engineering Strategies for Amelioration of Sight Distance Deficiencies Flow diagram of six-step diagnostic process C-51

Step #1: Collect Field Data This step involves making specific field measurements and observations. Data are to be gathered both at the location of a designated or possible hazard as well the approach roadway section immediately in advance of the hazard. Approach distances over which field measurements should be gathered are determined from Table 1 at the end of this step. Approach distances were derived from approximated perception-reaction and sign reading times applied to the designated operating speeds. Step # 1A Identify hazard and prepare site diagram Procedure Product/Application The specific hazard location under investigation is identified and the approach roadway is diagrammed. Example of hazards requiring sight distance consideration and the associated sight distance concepts are as follows. • A hidden intersection [SSD] • An exit from a shopping mall in a heavily lit ( or visually cluttered) setting [DSD] • A vehicle approaching an intersection [ISD] • An oncoming vehicle in a passing zone [PSD] Note distances from hazard to the following features: (1) traffic control devices, (2) intersecting driveway or roadways, and (3) sight distance obstructions. NOTE: An example sketch is shown in the example which follows. [Appendix A] Reference: Lunenfeld, H. and Alexander, G. J., A User’s Guide to Positive Guidance FHWA Report FHWA-SA-90- 017, Federal Highway Administration, Washington, DC 1990 Step # 1B Collect operating speed on approach Procedure Product/Application Spot speeds for randomly selected vehicles are to be observed at a sufficient advance distance upstream from the hazard beyond which slowing in response to the hazard is expected. Candidate speed collection techniques are radar/laser detection, automated speed recorders, and manual timing. References noted below describe appropriate procedures to ensure random vehicle selection and suitable sample sizes. In the event that the approach roadway section is characterized by horizontal or vertical curvature, speed collection points should be selected so as to represent operational speeds at these locations. The product of this step will be a statistical distribution of speeds from which means and/or percentile values will be applied to estimate vehicle speed for the approach roadway under study. References: Hanscom, F. R., Validation of a Non-automated Speed Data Collection Methodology. Transportation Research Record 1111. Transportation Research Board, National Research Council, Washington, D.C., 1987. Institute of Transportation Engineers, Manual of Transportation Engineering Studies, 2000 C-52

Step # 1C Observe erratic vehicle maneuvers on approach Procedure Product/Application Observations of vehicle movements should be considered in situations of sufficiently high traffic volumes to justify this type of study, e.g., 100 vph and above. Typical target vehicle behaviors indicative of a sight distance problem are sudden slowing (e.g., observable break light activation) and abrupt lane changes when these maneuvers are not induced by other vehicles in the traffic stream. A considerable literature base is available regarding the conduct and interpretation of “traffic conflicts” studies; however the reader is cautioned that traffic conflicts studies are limited to interactions between vehicles. A sight-distance induced erratic maneuver, on the other hand, can involve a single vehicle. Methodological literature addressing conflicts study is helpful with respect to observational techniques. The outcome of this step should be insightful with respect to possible sight-distance induced vehicle behaviors. References: Parker, M.R. and Zeeger, C.V. Traffic Conflicts for Safety and Operations. FHWA-IP-88-026 (Engineer’s Guide) and FHWA-IP-88-027 (Observer’s Guide) Federal Highway Administration, Washington, DC Taylor, J.I., and Thompson, H.T., Identification Of Hazardous Locations: A Users Manual, FHWA-RD- 77- 82, Federal Highway Administration, Washington, DC Step # 1D Inventory existing traffic control devices Procedure Product/Application Document existing signs, signals, and pavement markings along with their respective distances from the hazard under study. The letter heights on signs need to be recorded. The resulting device inventory will be subsequently applied in this diagnostic analysis to evaluate the suitability of provided information, as well as visual distractions and information processing demands on motorists as they approach the hazard under study. Step # 1E Measure existing geometric sight distances Procedure Product/Application Existing geometric sight distance limitations along the approach to the hazard must be measured in accordance with AASHTO criteria. Specifically, sight distance observations should be made from an elevation above the pavement which equals the design driver eye height, i.e., 3.5 feet, to a point ahead that is 2.0 feet above the pavement. This step will yield the length of specific roadway sub-sections along the approach in which drivers must observe and process available information, e.g., roadway features, other vehicles. Reference: (Pages 127 to 131) AASHTO, A Policy on Geometric Design of Highways and Streets, 2001 C-53

Step # 1F Note factors affecting flow speeds Procedure Product/Application Certain roadway environmental features are known to affect drivers’ selection of speed. Examples are pavement defects, narrow shoulder widths and protruding bridge piers, abutments, guardrail, median barriers, etc. Non- roadway features (e.g., pedestrians, parked vehicles) should also be noted. Documentation and general awareness of these factors are important due to the fact that subsequent minor highway improvement projects may result in higher highway speeds, thus producing increased sight distance requirements. Step # 1G Note visual distractions at hazard location Procedure Product/Application Certain environmental conditions are known to produce “visual clutter”, i.e., distractions which make hazards more difficult for drivers to perceive. Examples include: (1) off-roadway lighting, (2) commercial signing in driver field of view, (3) complex urban intersection designs, (4) high volumes of vehicular/pedestrian movement, and (5) proliferation of intersection traffic control devices. Observations should document drivers’ field of view at SSD from hazard, e.g., see page 111 of AASHTO, A Policy on Geometric Design of Highways and Streets, 2001 This inventory of visual distractions will be subsequently applied in a human factors analysis to determine the applicable sight distance criterion (e.g., Decision Sight Distance, to address driver perception and information-processing time requirements at the hazard location. Step # 1H Note visual distractions along approach roadway Procedure Product/Application As in Step 1G above, visual environmental conditions along the approach to the hazard may also produce driver distractions. These need to be included in the field data collection process. Observations should document drivers’ field of view at DSD from hazard, e.g., see page 116 of AASHTO, A Policy on Geometric Design of Highways and Streets, 2001 This inventory of visual distractions will be subsequently applied in a human factors analysis to determine the applicable sight distance criterion to address driver information processing time requirements on the approach to the hazard location. C-54

Step # 1I Label the diagram with specified symbols. Procedure Product/Application Σ∆ΗΑΖ - Sight distance to a potential hazard – The point at which a location or object is first detectable to an approaching motorist. Α - Point of required action – The location where an intended maneuver (e.g., hazard avoidance) is to be completed. Σ∆ΤΧ∆ - Sight distance to a traffic control device– The point at which the device is first detectable to an approaching motorist. ΤΧ∆- Location of traffic control device that warns of the hazard – measured as a distance from the location or object about which information is provided. The inclusion of uniform symbols on the site diagram will facilitate the subsequent sight distance analysis. A two-lane 55-mph roadway approaches a 35-mph curve. Example Symbol Diagram C-55

Approach Distance to Hazard, ft. Estimated Operational Speed, mph Visually Cluttered Environment (A) Visually Non- Cluttered Environment (B) Additional, when TCDs Present (C) 25 360 180 95 30 440 220 110 40 580 290 150 50 730 370 185 60 880 440 220 70 1030 520 260 (A) Allows 10-second approach PIEV, per MUTCD for high judgment requirement (B) Allows 5-second approach for 5-second visual scanning and PRT (2) Allows an addition 2.5-second PRT for sign comprehension Table 1 – Recommended approach distance to hazard for collection of field data. Step #2: Conduct Preliminary Engineering Analyses This step involves the application of traditional traffic engineering techniques, e.g., AASHTO Design Policy geometric design criteria and Decision Sight Distance warrants, as a preliminary determinant of site deficiencies. In addition, the placement of traffic control devices needs to be examined in terms of MUTCD requirements. Step #2A. Examine Hazard Location with respect to AASHTO Design Criteria Procedure Product/Application In order to ensure a valid engineering diagnosis of sight distance to a hazard, it is necessary to first assess whether the hazard location itself has any inherent design shortcomings. One geometric deficiency potentially associated with a hazard location might be roadside that fails to meet requirements of the AASHTO Roadside Design Guide. Other examples are (1) a high-accident intersection which may be deficient with respect to existing corner sight distance, (i.e., see pages 655 to 680 of AASHTO, A Policy on Geometric Design of Highways and Streets, 2001); and (2) in the case of a high incidence of run-off-road accidents, compare observed operational speeds (from Step 1A above) with the design speed based curve radius and superelevation and the curve under consideration, (e.g., see pages 131 to 168 of AASHTO, A Policy on Geometric Design of Highways and Streets, 2001) The resulting analytical steps ensure that the hazard location itself is free of any inherent design shortcomings that have the potential for confounding the intended sight distance diagnosis. C-56

Step #2B. Examine Approach with respect to AASHTO Design Criteria Procedure Product/Application As with the procedure noted in Step 2A, to ensure the integrity of the overall sight distance diagnosis, it is necessary to assess whether the approach to the hazard location has any inherent design shortcomings. (For example, a substandard lateral clearance to a roadside object along the approach may create a visual obstruction, thus producing an unintended sight distance limitation.) Likewise, crest vertical sight distances along the approach should be consistent with observed operational speeds gathered during Step 1B above. The resulting analytical steps ensure that the approach to the hazard is free of any inherent design shortcomings that have the potential for confounding the intended sight distance diagnosis. Step #2C. Examine Hazard Location with respect to possible DSD Warrants Procedure Product/Application AASHTO Design Policy (e.g., page 115, section on Decision Sight Distance) notes a distinction between typical stopping sight distances and those in which drivers are required to make complex decisions, i.e., in which drivers require perception response time beyond the design value (typically 2.5 s). The Decision Sight Distance criterion applies to a difficult-to-perceive information source in a roadway environment that may be visually cluttered. Therefore, the hazard location needs to be examined for conditions of “visual noise” from competing sources of information, e.g., roadway elements, traffic, TCDs, pedestrians, and advertising signs. Specific sources of visual clutter were also noted in Step 1E above. When DSD warranting conditions are found to exist, apply the sight distance requirements noted in Table 3- 3 (page 116, 2001 AASHTO Design Policy) rather than conventional stopping distances based on a 2.5-second perception response time. Step #2D. Examine Approach with respect to DSD Warrants Procedure Product/Application The approach to the hazard location must also be examined for conditions of visual clutter meeting requirements for DSD application. In particular, these could take the form of advertising signs and/or complex TCDs at intersections along the approach. Visual clutter along an approach to a hazard detracts from drivers’ perception of the hazard. When DSD warranting conditions are found to exist along an approach to a hazard, the distraction is sufficient such that available sight distance to the hazard must be restricted to that distance beyond the distraction. C-57

Step #2E. Examine Traffic Control Devices with respect to MUTCD Criteria Procedure Product/Application The Manual of Uniform Traffic Control Devices (MUTCD) prescribes device placement criteria for signs, signals, and markings. Devices at both the hazard location and along the approach need to be examined for MUTCD compliance. The output of this step will reveal whether inadequate traffic control device application, e.g., insufficient warning distance or inappropriate warning message, constitute possible sources of driver confusion. Inappropriate or inadequate TCD information can result in longer information processing times, thereby creating an artificial sight distance problem. Step #3: Apply Accident Data This step involves the integration of traffic accident data into the analysis. The objective is to locate specific accident-prone locations within the roadway segment which may be indicative of sight distance problems. The practitioner is cautioned that the absence of accidents does not rule out the existence of a sight distance problem, as accidents are probabilistic events and reporting requirements are variable. Step #3A. Establish Typologies and Frequency by Spot Locations Procedure Product/Application A review of accident data will reveal the occurrence of various types in close vicinity to the hazard under study. The associated pre- collision paths and their proximity to highway features may suggest the existence of a sight distance problem. Certain accident types are typically associated with specific sight distance problems, e.g.: • Run-off-road, Fixed object [SSD] • Side-swipe, rear-end [PSD] • Right angle, rear-end [ISD] A collision diagram is used to summarize accident types by location. For examples, see page 211, ITE Manual of Traffic Engineering Studies; and page 1-11 in Hostetter. R.S. and Lunenfeld, H. Planning and Field Data Collection, Report FHWA-TO-80-2, Federal Highway Administration, Washington, DC, 1982 Step #3B. Assess Suitability of Accident Sample Procedure Product/Application While well-documented procedures exist to statistically establish accident causation (see Accident Research Manual, Federal Highway Administration Report FHWA/RD-80-/016), this level of sophistication is not necessary for the diagnosis of a sight distance problem. It is desirable (to the extent possible based on available accident data) to establish causation inferences based on accident patterns and to rule out non-sight-distance causal effects. A reasonable level of confidence (albeit logic-based rather than statistically rigorous) regarding accident causation is possible based on the following: • Inferences based on accident patterns rather than a single event • Occurrences whereby non-sight-distance factors can be logically ruled out C-58

Step #3C. Examine Potential Sight Distance Causation Effect Procedure Product/Application Certain patterns of accident behaviors (i.e., pre- collision maneuver) are suggestive of sight distance problems. For example, single-vehicle or run-off-road occurrences with a fixed object, which may appear visible under some conditions, may not be easily detectable to drivers during conditions of more limited visibility (e.g., darkness). These patterns need to be examined to determine whether sight distance is a potential causal factor, i.e., adequate nighttime sight distance conveyed by TCDs. A collision diagram can be descriptive of the location and nature of a sight-distance hazard, thus supporting a hypothesis regarding the effect of a sight distance problem. Step #4: Establish Roadway Segments The user specifies component roadway approach segments in a manner to support the detailed human factors analysis in Step 5. Separate approach roadway segments are theoretically required for driver PRT and hazard avoidance maneuver functions. The product of this section is a series of driver task diagrams that depict the point where driver actions are required to avoid a potential hazard, information sources which warn of the hazard, and motorist’s available sight distances to perform the necessary information-processing and maneuver tasks. Step # 4A. Establish and plot action points along approach segment. Procedure Product/Application Identify and plot specific locations within the study roadway section requiring a driver action (e.g., maneuver). For example, the hazard under study is the key point where action (e.g., decelerating to the posted speed) is likely required. Where a maneuver (e.g., decelerating) is necessary prior to reaching the hazard, the “action point” is the point where the maneuver is initiated (e.g., end of the decision distance). In the event that the approach roadway section requires some intermediate action, e.g., merging from a dropped traffic lane, this action also needs to be identified and plotted. Action points on the site diagram prepared in Step 1 above should be indicated on the diagram by the symbol Α. A series of sequential action points may be designated as Α1, Α2 etc. The developed site diagram will indicate specific points where vehicle actions are required. Examples are as follows: 1. Approach maneuver (such as slowing) as required by the hazard under study 2. Any intermediate actions, e.g., required lane change, on the approach to the hazard under study. Note: Example plots of designated roadway segments (e.g., including appropriately labeled action points) are shown in the example diagnostic procedure application [Section 5.5]. C-59

Step # 4B. Establish and plot information sources and associated sight distances along approach segment. Procedure Product/Application Any driver action (e.g., hazard avoidance) must be based on information available to the driver. In this step, it is necessary to locate and document driver’s information sources providing information for an intended action. Information to the driver should be available from (1) detection of the hazard, and/or (2) traffic control devices pertaining to the hazard. The following information/detection sources were noted on the site diagram in Step 1-I. In this step, separate plots of component information-processing segments may be helpful. • Initial point of sight distance to the hazard by the symbol Σ∆ΗΑΖ. • Location of TCD providing information regarding the hazard by the symbol ΤΧ∆. • Initial point of sight distance to the applicable TCD by the symbol Σ∆ΤΧ∆ The developed site diagram will indicate specific points where information pertaining to the hazard is available to the driver. Examples are as follows. 1. Point of initial detection opportunity on an approaching of both the hazard and any traffic control device warning of the hazard. 2. Specific locations of any TCDs advising of the hazard. NOTE: In the event that the hazard under study is not detectable (i.e., defined in the visual field), the symbol Σ∆ΗΑΖ would not appear on the diagram. In such instances the required sight distance to action point (A) will be determined in Step 5. C-60

Step # 4C. Define component driver response sections within approach segment. Procedure Product/Application Distinctly different driver information- processing tasks are associated with each detection and maneuver activity. In this step, roadway sections will be designated and plotted to illustrate the required travel distances over which the driver would perform these varied information-processing and maneuver tasks. Depending upon physical characteristics of the roadway section under study, four distinct driver response cases are the following: Case 1. Direct line of sight to hazard Σ∆ΗΑΖ--------Æ Α Case 2. Intervening traffic control device, i.e., warning of hazard Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆--->TCD-----ÆΑ Case 3. Intervening, e.g., distracting, hazard (A2) within sight line of first hazard (A1) Σ∆ΗΑΖ1---Æ Σ∆ΗΑΖ2---ÆΑ2---ÆΑ1 Case 4. Intervening traffic control device and distracting hazard Σ∆ΤΧ∆ --ÆΛ∆ΤΧ∆-->TCD --Æ SDHAZ2 - ÆΑ2--ÆΑ1 The product of this step is a diagrammed set of roadway component sections, each corresponding to a specific information-processing and maneuver driver task. The distance over which a driver can react to a detectable hazard is the roadway section, SDHAZ−Α. In this roadway section the driver would detect the hazard and perform any required preparatory maneuver, e.g., decelerating. Likewise, the distance over which a driver reacts to an advance traffic control device is the roadway section SDTCD – TCD. In this roadway section the driver has the opportunity to detect the sign, and comprehend the sign’s message. The message becomes readable at the point, Λ∆ΤΧ∆ (i.e., the legibility distance from the sign), which will be computed and located during Step 5. In the final approach section to the hazard, TCD -- A, the driver would complete the decision-making and maneuver tasks. C-61

Step #5: Analyze Component Driving Task Requirements In this step the practitioner applies human factors principles, i.e., comprising information- processing and decision-making criteria, to ensure the adequacy (or to quantify the shortcoming) of the approach roadway to allow for time/distance hazard avoidance requirements. Step #5A. Determine the relevant geometric design sight distance application. Procedure Product/Application The analysis of driving task requirements involves application of the appropriate sight distance value for the given task. Sight distance requirements (to accommodate both the information processing and maneuver tasks) approaching action points(Α)will fall into one of the following categories (depending upon roadway environment condition) which were identified in Section 5.2. These are: • Stopping sight distance (SSD) [Section5.2.1] • Intersection sight distance (ISD) [Section5.2.2] • Decision sight distance (DSD) [Section5.2.3] • Passing sight distance (PSD) [Section5.2.4] The result of this task is the specification of the applicable procedure, e.g., engineering design formula, for the computation of Σ∆ΗΑΖ corresponding to each identified hazard or action point. The required sight distance based on application of the appropriate design formula is applied to determine the required length of the roadway segment under study. C-62

Step #5B. Determine driving task requirements within each component roadway segment. Procedure Driver information processing demands vary as a function of environmental factors, according to the four cases indicated below. Identify separate PRT and MT components of the driving task apply to each of the four cases [Section 5.2]. Specific values of PRT and MT will be determined subsequently determined. Case 1: Direct line of sight to hazard; no traffic control Σ∆ΗΑΖ--------Æ Α In this case, PRT and MT are determined from Section 5.2. Case 2: Intervening traffic control device, warning of hazard Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆--->TCD-----ÆΑ Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆ Driver must detect traffic control device. Λ∆ΤΧ∆--->TCD Driver must read or otherwise comprehend message and may begin decision process. (Legibility distance will be determined in Step 5C.) TCD-----ÆΑ Decision and maneuver must be completed. Case 3: Intervening, distracting hazard at A2 within sight line of first hazard at A1. Σ∆ΗΑΖ1---Æ Σ∆ΗΑΖ2---ÆΑ2---ÆΑ1 Σ∆ΗΑΖ1---Æ Σ∆ΗΑΖ2--ÆΑ Driver requires longer PRT due to complex visual scene ahead. Consider DSD application. Σ∆ΗΑΖ2---ÆΑ2---ÆΑ Driver may require longer MT due to complexity of maneuver and visual scene. Case 4: Intervening traffic control device and distracting hazard at A2 within sight line of first hazard A1. Σ∆ΤΧ∆ --ÆΛ∆ΤΧ∆-->TCD --Æ SDHAZ2 -ÆΑ2--ÆΑ1 Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆ Driver must detect traffic control device Λ∆ΤΧ∆--->TCD Driver must read or otherwise comprehend message and may begin decision process Σ∆ΗΑΖ2---ÆΑ2---ÆΑ Driver may require longer MT due to complexity of maneuver and visual scene. C-63

Step #5C. Quantify the applicable PRT and MT requirements for each driving task component. Procedure The general model to be applied for quantifying driver task requirements (i.e., required PRT and MT) is the following: Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆--->TCD-----ÆΑ. Driver task requirements are determined for each task as follows. No TCDs present: Σ∆ΗΑΖ--------Æ Α Apply applicable PRT and MT requirement corresponding to predetermined condition, i.e., SSD, ISD, DSD, or PSD as determined in Step 5A. TCDs present: Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆ Drivers should be able to detect a TCD prior to time required to comprehend its message. 2.5 s is desirable, although less time may be adequate, e.g., second, third, etc. in a sequence. Λ∆ΤΧ∆--->TCD Λ∆ΤΧ∆ is the “legibility distance”, i.e., the approach distance a traffic control device message is comprehended. A detailed discussion below addresses the Λ∆ΤΧ∆ for signs. In the case of pavement markings, it is the advance distance at which the marking is visually recognized. The Λ∆ΤΧ∆ a sign is the distance at which its legend is read or its symbol message is comprehended. PRT requirements [Ontario Traffic Manual] for signs consist of sign message legend and symbol reading times as follows: Reading Time = 1*(number of symbols) + 0.5*(no. of words and numbers) [secs] For messages exceeding 4 words, the sign requires multiple glances and the driver must look back to the road and at the sign again. Therefore, for every additional 4 words and numbers, or every 2 symbols, an additional ¾- second should be added to the reading time. TCDs present (Cont.): The minimum reading time is 1 second. If there are more than 4 words on a sign, a driver must glance at it more than once, and look back to the road and at the sign again. For every additional 4 words and numbers, or every 2 symbols, an additional 3/4 second should be added to the reading time. This segment must be sufficient in length to accommodate the reading time noted above. However, its length is constrained by letter height, i.e., limited to 40 feet for every inch of letter height. For example, a 4-inch letter-height sign must be read within a distance of 4 X 40 = 160 feet. On a 40 mph (58.8 fps) roadway, the driver is limited to a maximum of 160/58.8 or 2.7 seconds to read the sign. Moreover, the traffic engineer must consider that the driver can not be expected to fixate on the sign. Decision Time, i.e., to make a choice and imitate a maneuver if required. Considering the driver’s alerted state having read the sign, decision time can range from one second for commonplace maneuvers (e.g. stop, reduce speed) to 2.5 seconds or more when confronted with a complex highway geometric situation. Λ∆ΤΧ∆--->TCD-----ÆΑ While the required MT may be initiated prior to passing the TCD, it must be completed in the above- noted segment. MT values associated with designed sight distance considerations are treated herein [Section 5.2]. Additional literature sources of extensive maneuver time data are available [Lerner, N.D., Steinberg, G.V., Huey, R.W., and Hanscom, F.R., 1999] C-64

Step #5D. Assess the adequacy of the available sight distance components. Procedure Procedure (continued) Case 1: Direct line of sight to hazard; no traffic control Σ∆ΗΑΖ--------Æ Α Does the subsection length Σ∆ΗΑΖÆ Α allow sufficient time for the driver to perform any required hazard avoidance maneuver? Case 2: Intervening traffic control device, warning of hazard Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆--->TCD-----ÆΑ Does the subsection length, Σ∆ΤΧ∆ÆΛ∆ΤΧ∆ allow sufficient time (minimum1.5 seconds) for the driver to detect the traffic control device? Does the subsection length, Σ∆ΤΧ∆ÆΤΧ∆ allow sufficient time for the driver to detect and read the traffic control device? Does the subsection length, ΤΧ∆ÆA allow sufficient time for the driver to perform any required hazard avoidance maneuver? Case 3: Intervening, distracting hazard at A2 within sight line of first hazard at A1. Σ∆ΗΑΖ1---Æ Σ∆ΗΑΖ2---ÆΑ2---ÆΑ1 Does then subsection length Σ∆ΗΑΖ1Æ Α1 allow sufficient time for the driver to process and respond to the intervening distraction (i.e., apply DSD criteria) and perform any required hazard avoidance maneuver? Case 4: Intervening traffic control device and distracting hazard A2 within sight line of first hazard A1. Σ∆ΤΧ∆ --ÆΛ∆ΤΧ∆-->TCD --Æ SDHAZ2 -ÆΑ2--ÆΑ1 Does the subsection length, Σ∆ΤΧ∆ÆΛ∆ΤΧ∆ allow sufficient time (2.5 s desirable; minimum 1.0 to 1.5 s) for the driver to detect the traffic control device? Does the subsection length, Σ∆ΤΧ∆ÆΤΧ∆ allow sufficient time for the driver to detect and read the traffic control device? Does the subsection length, ΤΧ∆ÆA1 allow sufficient time for the driver to process and respond to the intervening distraction (i.e., apply DSD criteria) and perform any required hazard avoidance maneuver? C-65

Step #6: Develop Engineering Strategies for Amelioration of Sight Distance Deficiencies. In this final step the practitioner recommends improvement, e.g., traffic control device applications or minor design modifications to correct deficiencies. Step #6A. Apply traffic engineering and highway design principles to component sight distance deficiencies. Problem Solution In Case 1: Direct line of sight to hazard; no traffic control Σ∆ΗΑΖ--------Æ Α Available sight distance to hazard, Σ∆ΗΑΖ, is less than required based on Step 5B results. Add warning traffic control device, increasing warning distance as shown in case 2 below. In Case 2: Intervening traffic control device, i.e., warning of hazard Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆--->TCD-----ÆΑ Total available sight distance less than the required sight distance from Step 5C. If Λ∆ΤΧ∆--->TCD is inadequate, i.e., information overload. Apply “information spreading” by adding more devices, each with less information. Increase legibility distance, e.g., by increasing letter size. If Λ∆ΤΧ∆--->TCD-----ÆΑ is inadequate. Increase warning distance, Σ∆ΤΧ∆ÆΛ∆ΤΧ∆ via improving the TCD’s legibility distance. Apply larger device, increase letter size. In DSD condition, add conspicuity device, e.g., flashing beacon or consider ITS application. If Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆--->TCD is inadequate. Reduce information load on existing TCDs. Apply additional TCDs (e.g., delineation devices, advance supplemental devices) to convey essential information. Case 3: Σ∆ΗΑΖ1---Æ Σ∆ΗΑΖ2---ÆΑ2---ÆΑ1 Available sight distance to hazard, Σ∆ΗΑΖ, is less than required based on Step 5B results. Add warning traffic control device, achieving increased warning distance. Case 4: Σ∆ΤΧ∆ --ÆΛ∆ΤΧ∆-->TCD --Æ SDHAZ2 -ÆΑ2-- ÆΑ1 Total available sight distance less than the required sight distance from Step 5C. Apply combination of Case 2 solutions noted above. C-66

5.5 References AASHTO (2001). A Policy on Geometric Design of Highways and Streets. American Association of State Highway and Transportation Officials, Washington, DC., 127 - 131. Adebisi, O. and Sama, G. M. (1989) Influence of stopped delay on driver gap acceptance behavior. ASCE Journal of Transportation Engineering, 115(3). Alexander, G. and Lunenfeld, H. (1975) Positive guidance in traffic control. Federal Highway Administration, Washington, D.C. Ebbesen, E. B., Parker, S., and Konecni, V. J. (1977) Laboratory and field analysis of decisions involving risk. Journal of Experimental Psychology: Human Perception and Performance, 3(4), 576-589. Evans, L. and Wasielewski, P. (1983) Risky driving related to driver and vehicle characteristics. Accident Analysis & Prevention, 15, 121-136. Fambro, D. B., Fitzpatrick, K., and Koppa, R. J. (1997) Determination of stopping sight distances. Transportation Research Board NCHRP Report 400. Farber, E. and Silver, C. A. (1967) Knowledge of oncoming car speed as determiner of driver's passing behavior. Highway Research Record, 195, 52-65. Federal Highway Administration (1994) The magnitude and severity of passing accidents on two-lane rural roads. Rep. No. FHWA-RD-94-068, Highway Safety Information System, Washington, D.C. Fitzpatrick, K. (1991) Gaps accepted at stop-controlled intersections. Transportation Research Record, 1303, 103-112. Fitzpatrick, K., Carlson, P. J., Wooldridge, M. D., and Brewer, M. A. (2000) Design factors that affect driver speed on suburban arterials. Rep. No. FHWA Report No. FHWA/TX-001/1769-3, Federal Highway Administration, Washington, D.C. Hanscom, F. R., Validation of a Non-automated Speed Data Collection Methodology. Transportation Research Record 1111. Transportation Research Board, National Research Council, Washington, D.C., 1987. Harwood, D. W., Mason, J. M., Brydia, R. E., Pietrucha, M. T., and Gittings, G. L. (1996) Intersection sight distance. Rep. No. 383, NCHRP. Hassan, Y., Easa, S., and Abd El Halim, A. (1996) Passing sight distance on two-lane highways: Review and revision. Transportation Research, 30(6), 453-467. Hostetter, R. S. and Seguin, E. L. (1969) The effects of sight distance and controlled impedance on passing behavior. Highway Research Record. Hughes, W., Eccles, K., Harwood, D., Potts, I., and Hauer, E. (2004). Development of a Highway Safety Manual. NCHRP Web Document 62: Project 17-18(4): Contractor’s Final Report. Transportation Research Board, Washington, DC. Institute of Transportation Engineers, Manual of Transportation Engineering Studies, 2000. C-67

Kaub, A. R. (1990) Passing operations on a recreational two-lane, two-way highway. Transportation Research Record, 1280, 156-162. Krammes, R., Brackett, Q., Shafer, M., Ottesen, J., Anderson, I., Fink, K., Pendleton, O., and Messer, C. (1995). Horizontal alignment design consistency for rural two-lane highways. Rep. No. RD-94-034, Federal Highway Administration, Washington, D.C. Kyte et al. (1996) Capacity analysis of unsignalized intersections. In NCHRP Project 3-46. Unpublished, cited by Harwood et al., 1996. Lamm, R. and Choueiri, E. M. (1987). Recommendations for evaluating horizontal design consistency based on investigations in the State of New York. Geometric Design and Operational Effects - Transportation Research Record, 1122, 68-78. Lerner, N., Llaneras, R., Hanscom, F., Smiley, A., Neuman, T., and Antonucci, N. (2002). Revised Task 2 Report: HFG Outline and Recommendations. Report under NCHRP Project 17-18(8). Transportation Research Board, Washington, DC. Lerner, N., Huey, R. W., McGee, H. W., and Sullivan, A. (1995) Older driver perception-reaction time for intersection sight distance and object detection. Volume I, Final Report. Rep. No. FHWA-RD-93-168, Federal Highway Administration, Washington, D.C. Lunenfeld, H. and Alexander, G. J., A User’s Guide to Positive Guidance FHWA Report FHWA-SA-90- 017, Federal Highway Administration, Washington, DC 1990. Manual on Uniform Traffic Control Devices for Streets and Highways (2003). Federal Highway Administration, U.S. Department of Transportation, Washington, DC. McGee, H. W., Moore, W., Knapp, B. G., and Sanders, J. H. (1978) Decision sight distance for highway design and traffic control requirements. Rep. No. FHWA-RD-78-78, U.S. Department of Transportation, Washington, D.C. McLean, J. R. and Hoffman, E. R. (1972) The effects of lane width on driver steering and performance. Australian Road Research Board Proceedings, Vol. 6, Part 3, pp. 418 - 440. McNees, R. W. (1982) In situ study determining lane-maneuvering distance for three- and four-lane freeways for various traffic-volume conditions. Transportation Research Record, 869, 37-43. OECD (1990) Behavioral adaptations to changes in the road transport system. Organization for Economic Co-operation and Development, Road Transport Research, Scientific Expert Group, Paris, France. Ontario Ministry of Transportation, Ontario Traffic Manual, “Sign Design Principles”, 2000. Olson, P. L. (2002). "Driver perception and response time." Human Factors in Traffic Safety, R. E. Dewar and P. L. Olson, eds., Lawyers & Judges Publishing Company, Tucson, Arizona. Olson, P. L., Cleveland, D. E., Fancher, P. S., and Schneider, L. W. (1984) Parameters affecting stopping sight distance. Rep. No. UMTRI-84-15, University of Michigan Transportation Research Institute, NCHRP Project 1508. C-68

Olson, P. L. and Sivak, M. (1983) Improved low-beam photometrics. Rep. No. UMTRI-83-9, University of Michigan Transportation Research Institute, Ann Arbor, MI. Parker, M. R. Jr. (1997) Effects of raising and lowering speed limits on selected roadway sections. Rep. No. FHWA-RD-92-084, Federal Highway Administration, Washington, D.C. Parker, M.R. and Zeeger, C.V. Traffic Conflicts for Safety and Operations. FHWA-IP-88-026 (Engineer’s Guide) and FHWA-IP-88-027 (Observer’s Guide) Federal Highway Administration, Washington, DC Pline, J. (Editor) (1999) Traffic Engineering Handbook. Institute of Transportation Engineers, Washington, DC. Pline, J. (Editor) (2001) Traffic Control Devices Handbook. Institute of Transportation Engineers, Washington, DC. Polus, A., Fitzpatrick, K., and Fambro, D. B. (2000). Predicting operating speeds on tangent sections of two-lane rural highways. Transportation Research Record, 1737, 50-57. Rajalin, S., Hassel, S.-O., and Summala, H. (1997) Close following drivers on two-lane highways. Accident Analysis & Prevention, 29(6), 723-729. Ranney, T., Masalonis, A. J., and Simmons, L. A. (1996) Immediate and long-term effects of glare from following vehicles on target detection in driving simulator. Transportation Research Record, 1550, 16- 22. Roper, V. J. and Howard, E. A. (1938) Seeing with motorcar headlamps. Illumination Engineering, 33, 412-438. Shinar, D. (1985) The effects of expectancy, clothing reflectance, and detection criterion on nighttime pedestrian visibility. Human Factors, 27(3), 327-333. Shinar, D., McDowell, E., and Rockwell, T. H. (1977) Eye movements in curve negotiation. Human Factors, 19(1), 63-71. Smiley, A. and Rochford, S.L. Behavioral adaptation and anti-lock brake systems. Final Report prepared for Transport Canada, Ottawa, October, 1991. Also presented with B. Grant at IEA 1991, Paris, France, July, 1991. Staplin, L., Lococo, K., Byington, S., and Harkey, D. (2001). Highway Design Handbook for Older Drivers and Pedestrians. Report number FHWA-RD-01-103. U.S. Department of Transportation, Washington, DC. Staplin, L. K., Lococo, K., and Sim, J. (1993) Traffic maneuver problems of older drivers. Publication No. FHWA-RD-92-092, Federal Highway Administration, Washington, DC. Taylor, J.I., and Thompson, H.T., Identification Of Hazardous Locations: A Users Manual, FHWA-RD- 77- 82, Federal Highway Administration, Washington, DC. Weaver, G. D. and Glennon, J. C. (1972) Design and striping for safe passing operations. Highway Research Record, 390, 36-39. C-69

Wortman, R. H. and Matthais, J. S. (1983) Evaluation of driver behavior at signalized intersections. Transportation Research Record, 904. Yagar, S. and Van Aerde, M. (1983) Geometric and environmental effects. Transportation Research, 17A(4), 315-325. C-70

The example driving situation consists of a 55-mph, two-lane rural roadway which approaches a 35-mph curve followed by a Stop-controlled intersection. The intersection approach is to a main highway, i.e., requiring application of destination guide signing. Driver requirements in this situation are as follows: 1. Reduce speed from 55 to 35 m.p.h. to negotiate curve 2. Process traffic control device information related to intersection, e.g., Destination name sign 3. Stop for intersection 1. Step 1- Collect Field Data and Prepare Site Diagram The labeled site diagram is shown below. Example Site Diagram 2. Step 2- Conduct Preliminary Engineering Analyses This example requires a sight distance analysis to two separate potential hazards. The first is a 35 mph curve which requires slowing from 55 mph; and the second is an intersection which is heavily signed with a Stop sign and two guide signs, containing multiple route shields, symbols, and destination names. The approach roadways to each hazard point are separately treated as follows: (1) curve approach, and (2) signed intersection approach. 2.1 Curve Approach Segment Steps 2A thru 2D – Examine Site with respect to AASHTO Design and DSD Criteria For the purpose of this example, it is assumed that geometrics conform to AASHTO and that DSD criteria (e.g., visually cluttered environmental conditions) do not apply. Step 2E – Examine Traffic Control Devices for Compliance with the MUTCD Chapter 2C of the MUTCD specifies requirements for warning signs. The curve warning sign in the example is a “W1-2, Horizontal Alignment Sign” with a 35-mph advisory speed plate. Section 2C-05 of the MUTCD specifies an “advance placement guideline” for warning signs. Given the requirement to slow from 55 to 35 mph, the recommended distance in Table 2C-4 is 350 feet. Attachment A: Example Application: Sight Distance Diagnostic Procedure AT-1

2.1.1 Signed Intersection Approach Segment Steps 2A thru 2D – Examine Site with respect to AASHTO Design and DSD Criteria For the purpose of this example, it is assumed that geometrics conform to AASHTO and that DSD criteria (e.g., visually cluttered environmental conditions) do not apply. Step 2E – Examine Traffic Control Devices for Compliance with the MUTCD This segment is a stop-signed intersection approach containing signs to multiple routes and destinations. Chapter 2D of the MUTCD provides requirements for guide signs on conventional roads. Signs in the example consist of a “directional assembly” with destination name signs and route shields. Required advance distances and spacing of these signs is given in Figure 2D-2. Typically, when a series of guide signs is sequentially placed along the approach to an intersection there is a 100-to-200 foot separation between the first two signs. The minimum spacing between signs is 100 feet, i.e., intended to enable drivers to read the entire message on either sign. Section 2D.06 requires 6-inch letter heights for a 35- mph roadway. Specifications for Stop sign size and placement are contained in Chapter 2A of the MUTCD. As shown in Figure 2A-2, the Stop sign should be set back a minimum of 12 feet from the intersection. The recommended letter height is 8 inches. 3. Step 4 – Establish Roadway Segments This example requires a sight distance analysis to two separate potential hazards. The first is slowing from 55 mph to the 35 mph posted curve advisory speed; and the second is a stop signed approach to an intersection containing signs to multiple routes and destinations. As above, the approach roadways are separately discussed. 3.1 Curve Approach Segment The roadway segment requiring the driver to slow from 55 mph to a 35-mph curve is labeled in accordance with Steps 4A and 4b and is shown below. The two sight distance driver response scenarios are the following: Case 1. Direct line of sight to hazard, i.e., 55-mph speed zone to 35-mph curve Σ∆ΗΑΖ--------Æ Α Case 2. Intervening traffic control device, i.e., 35-mph advisory speed sign warning of hazard Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆--->TCD-----ÆΑ AT-2

This roadway is diagrammed below. 3.2 Signed Intersection Approach Segment On this roadway section, 35-mph motorists are confronted with a stop-signed intersection and two guide signs containing destination names and route shields. Due to the fact that sight distance to the intersection is limited by a curve on the approach, a sight distance analysis is critical. The component section diagram is labeled in accordance with Steps 4A and 4b and shown below. The sight distance driver response scenarios are the following: Case 1. Direct line of sight to hazard, i.e., 35-mph speed zone to intersection Σ∆ΗΑΖ--------Æ Α Case 2. Three intervening traffic control devices, i.e., A route shield assembly: Σ∆ΤΧ∆1-----ÆΛ∆ΤΧ∆1--->TCD1-----ÆΑ A destination name sign: Σ∆ΤΧ∆2-----ÆΛ∆ΤΧ∆2--->TCD2-----ÆΑ A Stop sign: Σ∆ΤΧ∆3-----ÆΛ∆ΤΧ∆3--->TCD3-----ÆΑ This roadway segment is diagrammed below. Intersection Approach Segment Diagram AT-3

4. Step 5 – Analyze Component Driving Task Requirements 4.1 Curve Approach Segment The roadway section, requiring the driver to slow from 55 mph to a 35-mph curve, considers sight distance to the curve and legibility distance requirements posed by the advisory speed sign. Step 5A – Determine the relevant design sight distance application The applicable design sight distance is Slowing Sight Distance, i.e., the required distance ahead for a driver to observe a curve (e.g., potential hazard) ahead and adjust its speed accordingly. In the event that certain visual noise conditions or other factors are present which would render the curve as difficult-to- perceive, then consideration must be given to applicable Decision Sight Distance criteria [Section 5.2.3]. Where a traffic control device is present, driver information processing time is required to observe and comprehend the sign as well as slow to a safe curve negotiation speed. In the current example, i.e., a rural uncluttered environment, the DSD criterion is not applied. Step 5B – Determine the driving task requirements Considering the two possibilities in this case, i.e., Case 1 in which the driver observes the curve ahead without seeing the sign, and Case 2 whereby the driver observes and comprehends the sign, the requirements are as follows: Case 1. Direct line of sight to hazard, i.e., 55-mph speed zone to 35-mph curve Σ∆ΗΑΖ--------Æ Α The sight distance requirement in this case is simply that the driver observes the curve ahead and slows to a safe speed. Case 2. Intervening traffic control device, i.e., 35-mph advisory speed sign warning of hazard Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆--->TCD-----ÆΑ The sight distance requirement in this case is that the driver observes the sign, comprehends the sign message, and slows to a safe speed. Step 5C – Quantify the applicable PRT and MT requirements for each driving task Case 1. Direct line of sight to hazard, i.e., 55-mph speed zone to 35-mph curve Σ∆ΗΑΖ--------Æ Α Recalling that DSD does not apply, the design PRT value of 2.5 s is applied; thus the PRT component of sight distance is 202 feet, i.e., 2.5 s times 80.85 fps. The MT requirement (4.0 s) is derived from the need to slow from 55 mph to 35 mph at a comfortable deceleration rate, i.e., .23g1, requiring 261 feet. Thus the total PRT and MT sight distance requirement is 463 feet. It is noteworthy, however, that the 2001 AASHTO Green Book acknowledges that its deceleration data may be outdated and that more rapid (albeit uncomfortable) decelerations are common. A typical such deceleration is .35g [Knipling, 1993], resulting in an MT of 2.6s. It is 1 Derived from Exhibit 2-25, AASHTO Policy on Geometric Design of Streets and Highways, 2001. For safety purposes, wet weather deceleration is considered. AT-4

also known that most reasonably alert drivers are able to initiate braking within a PRT of 1.6 s [Section 5.2.1]. Applying these performance parameters to slowing from 55 to 35 mph, the total required PRT distance is 129 feet plus 172 feet MT distance, or 301 feet. It is unlikely that the need to slow to 35 mph would be visually evident from an advance distance of either 301or 463 feet. Therefore, the critical sight distance consideration is based on the application of the speed advisory sign. Case 2. Intervening traffic control device, i.e., 35-mph advisory speed sign warning of hazard, Σ∆ΤΧ∆----- ÆΛ∆ΤΧ∆--->TCD-----ÆΑ In this case the driver needs to detect the sign, read the sign, and decelerate to the safe curve speed. A critical requirement for sight in advance of a highway i.e., allowing time to comprehend the sign’s message, is known as Legibility Distance. There is a considerable body of knowledge regarding sign legibility distance requirements [Smiley, 2000] For simple warning signs, the MUTCD specifies a 250-foot legibility distance for symbol signs (see Table 2C-04) applied in “condition B”, e.g., slowing for a curve. Moreover, the MUTCD considers that a deceleration value of 11.2 fpss be applied for determining warning sign placement. Consider the driving task requirements as follows: 2.0 s are needed to detect and comprehend (e.g., minimum 1.0 s detection time plus 1.0 symbol comprehension) the simple warning sign message prior to the initiation of slowing, the deceleration requirement would be .32g or approximately the equivalent slowing rate of skidding on wet pavement. In this example the required PRT and MT distances would be 161 and 189 feet respectively, for a total of 350 feet. The MUTCD-recommended warning sign advance placement of 137½ feet (interpolation from Table 2C- 04) in addition to the indicated 250-foot legibility distance provides 387½ feet of information lead distance. Therefore the recommended MUTCD warning sign placement in advance of the curve is adequate. For signs with complex messages, i.e., sets of destination names or symbols in combination with symbols, message comprehension may require significantly more legibility distance. The next example illustrates such a situation. 4.2 Signed Intersection Approach Segment On this roadway section, 35-mph motorists are confronted with a stop-signed intersection and two guide signs containing destination names and route shields. Since sight distance to the intersection is limited by a curve on the approach, a sight distance analysis is critical. Step 5A – Determine the relevant design sight distance application As the driver approaches a Stop-signed intersection, there must sufficient available Stopping Sight Distance [Section 5.2.1] to enable stopping at the stop line. (While negotiating intersection involves the application of Intersection Sight Distance, the current example is limited to approaching the intersection.) AT-5

Step 5B – Determine the driving task requirements Considering the two possibilities in this case, i.e., Case 1 in which the driver proceeds to the intersection ahead while ignoring the signs, and Case 2 whereby the driver observes and comprehends the intermediate signs, the requirements are as follows: Case 1. Direct line of sight to hazard, i.e., 35-mph speed zone to intersection Σ∆ΗΑΖ--------Æ Α The sight distance requirement (to accommodate travel time) in this case is simply that the driver observes the intersection ahead and safely slows to a stop. Case 2. Three intervening traffic control devices, i.e., A route shield assembly: Σ∆ΤΧ∆1-----ÆΛ∆ΤΧ∆1--->TCD1-----ÆΑ1 A destination name sign: Σ∆ΤΧ∆2-----ÆΛ∆ΤΧ∆2--->TCD2-----ÆΑ2 A Stop sign: Σ∆ΤΧ∆3-----ÆΛ∆ΤΧ∆3--->TCD3-----ÆΑ3 where TCD1 is a route shield assembly bearing two route designations, TCD2 is a destination guide sign with two destination names and directional arrows, and TCD3 is a Stop sign. The sight distance requirement in this case is that the driver detects and comprehends the signs, and slows to a safe stop at the stop line. Step 5C – Quantify the applicable PRT and MT requirements for each driving task Case 1. Direct line of sight to hazard, i.e., speed reduction from 35-mph speed to stop at the stop line. Σ∆ΗΑΖ--------Æ Α The design Stopping Sight Distance does not accommodate information-processing requirements of the intervening guide signs. The AASHTO design SSD value [AASHTO Green Book, 2001] for a 35-mph approach is the range, 225 to 250 feet, which accounts for both the PRT and MT tasks. It should be noted that the above sight distance would barely accommodate the physical placement of the two guide sign assemblies that are shown in the figure. Moreover, the information-processing load imposed by the signs requires significant attention in terms of sight distance requirements. Therefore the Case 2 condition is treated below. The general model, Σ∆ΤΧ∆-----ÆΛ∆ΤΧ∆--->TCD-----ÆΑ entails the following considerations. First, there must be sufficient sight distance so that the sign is detected prior to time required to comprehend the sign’s message, thus application of the Σ∆ΤΧ∆ AT-6

term. This advance distance is not specified in the MUTCD. Nevertheless, 2.5 seconds is desirable for this sign detection task, although less time may be adequate as motorists who are looking for signs are generally aware of the expected position in their field of view. The more essential approach sight distance to a traffic control device is that required to comprehend its message. The symbol in the above model, Λ∆ΤΧ∆ refers to “legibility distance”, i.e., the approach distance at which a TCD legend is read or its symbol message is comprehended. The legibility distance of a legend sign is determined multiplying a “Legibility Index” (i.e., the distance at which a given unit of letter height is readable) by the letter height. The applicable Legibility Index values are shown in the table below. For example, the legibility distance typically associated with 6-inch letter height is 40 times 6 or 240 feet. Legibility Index: Legibility distance based on letter height Metric US Customary 4.8 meters/centimeter 40 feet/inch The legibility distance of symbol signs has been researched2 in a laboratory study [Dewar and Swanson, 1997] and found to significantly exceed that of legend signs (despite the high degree of variability in the study data). For example, the mean legibility distance for the right curve arrow symbol was determined to be 283 meters (with a standard deviation of 68 meters). Consider that a 55-mph approach allowing a 2.5-second advance sight distance and 1.0-second reading time would consume only 86 meters, pure symbol signs are not expected to result in an information processing problem. The required PRT for this example roadway segment is comprised of three components, i.e., detection of the signs, comprehending the sign messages, and detecting the intersection. Each is separately discussed. Sign Detection Upon a driver’s detection of the first sign, the second and third would require minimal detection time. The recommended detection time for the first sign is 2.5 seconds; however the second two signs are likely to be detected much more rapidly. “Alerted” PRT responses are known to occur in as little as 1.0 to 1.5 seconds. Moreover, signs can be quickly detected as drivers know where to look for signs and typically scan toward expected sign locations. Therefore, a conservative sign detection PRT for the example roadway segment is (2.5 + 1.5 + 1.5) or 5.5 seconds. Sign Comprehension Sign comprehension consists of the sign reading task plus the process of making the resultant decision, e.g., right or left turn in response to the sign’s information. The PRT requirement3 is based on sign-response reading and decision time, for which general rules are noted in the table on the next page. The first guide sign assembly contains two numbers and two symbols, requiring 3.0 seconds of reading time; the second contains two designation name and two symbols, also requiring 3.0 seconds; and the third is a simple and familiar one-word regulatory sign, requiring one second. 2 See page 97 of the Traffic Control Devices Handbook [Pline, 2001] 3 Smiley, Ontario Ministry of Transportation, Ontario Traffic Manual, “Sign Design Principles”, 2000 AT-7

Thus the total sign reading time is 7.0 seconds. This estimate is highly conservative, as drivers would likely scan the guide signs seeking only a particular name or route number; however, it is necessary to provide sufficient information-processing sight as some drivers may need the entire set of information. An additional 3.0 seconds is considered for decision time responses to the three signs. Thus the total comprehension time for the three signs is ten seconds. Sign Comprehension PRT Requirements Reading Time requirements are: ½ second for each word or number, or 1 second per symbol, with 1 second as a minimum for total reading time. In the event of the sign’s containing redundant information, the reading time computation should be limited to critical words. The suggested formula for estimating sign reading time is, reading time (seconds) = 1(number of symbols) + 0.5(number of words and numbers). For messages exceeding 4 words, the sign requires multiple glances and the driver must look back to the road and at the sign again. Therefore, for every additional 4 words and numbers, or every 2 symbols, an additional ¾-second should be added to the reading time. When the driver is sufficiently close to see a sign at an angle, the sign is not visible for the last ½ second. Therefore, ½ second should be added to the required reading time. An exception applies to signs requiring a maneuver before the sign is reached, as no further reading is required. Decision Time, i.e., to make a choice and imitate a maneuver if required. Considering the driver’s alerted state having read the sign, decision time can range from one second for commonplace maneuvers (e.g. stop, reduce speed) to 2.5 seconds or more when confronted with a complex highway geometric situation. Intersection Detection Distance As noted above under the Case 1 (Σ∆ΗΑΖ---Æ Α) discussion, the Stopping Sight Distance Requirement considers a 2.5-second PRT. A summary of the above-noted PRT requirements, if separately considered, is shown in the table below. Driving Task PRT Requirement (seconds) Perceive initial guide sign 2.5 Perceive next 3 signs, @ 1.5 s 4.5 Comprehend initial guide sign 4.0 Comprehend second guide sign 4.0 Comprehend Stop Sign 2.0 Perceive intersection 2.5 Total 19.5 AT-8

The above sum of PRT requirements would apply to a serial task process. However, a realistic assessment of PRT requirements considers that many the above tasks are concurrent. For example, the Stop sign comprehension and decision-making tasks would not logically entail a separate process of perceiving the intersection, thus conceivably reducing the total PRT by 2.5 seconds. In addition, following a driver’s 2.5-second detection of the initial sign, the subsequent two signs would likely be detected with a minimum detection time (e.g., 1.0 seconds rather than 1.5 seconds), thus conceivably reducing the total PRT by another 1.0 second. Therefore, subtracting 3.5 seconds from the serial total of 19.5 seconds, the estimated PRT requirement becomes 16.0 seconds. The MT requirement, i.e., to slow from 35 mph to a stop at the specified AASHTO g-force, calculates to 4.7 seconds over a distance of 120 feet. The extent to which the deceleration process would occur concurrently with the various sign-response tasks is uncertain. However, it is logical (and best serves liability concerns) to allow time for comprehension of all signs prior to the initiation of the slowing response. Therefore the overall sight distance requirement is 16.0 seconds of sign information processing at 35 mph (51.45 fps) or 823 feet, plus the 120-foot deceleration distance, for a total of 943 feet. A final consideration is the necessity that drivers have sufficient time to comprehend a sign’s message during the interval when the message is discernable. Therefore, an essential sight distance diagnostic step is to compare the available sign legibility distance (i.e., available reading distance) with distance traveled during reading PRT (i.e., required reading distance and decision time). The table below contrasts the distance traveled during PRT with the legibility distance. While the guide signs in this example accommodate both reading time and associated decision time, the decision component of PRT can obviously be accomplished after the driver passes the sign. Legibility Distance, ft PRT Distance, ft Sign #1 6-inch letters, 2 Numbers + 2 Symbols 240 231 Sign #2 6-inch letters, 2 Numbers + 2 Symbols 240 231 Sign #3 8-inch letters, 1 Word Word, 320 Symbol, 51 AT-9

References Dewar, R.E., Kline, D.W., Schieber, F., Swanson, H.A., “Symbol Signing Design for Older Drivers”, Federal Highway Administration, Report # FHWA-RD-94-069, Washington, D. C., 1994. Knipling, R. R. et al., Assessment of IVHS Countermeasures for Collision Avoidance: Rear-end Crashes, Report NRD-50, National Highway safety Administration, Washington, DC 1993 Lerner, N.D., Steinberg, G.V., Huey, R.W., and Hanscom, F.R. Understanding Driver Maneuver Errors, Final Report, Contract DTFH61-96-C-00015, Federal Highway Administration, Washington, D.C., July 1999 Pline, J. (Editor) (2001) Traffic Control Devices Handbook. Institute of Transportation Engineers, Washington, DC. Smiley, A. Ontario Ministry of Transportation, Ontario Traffic Manual, “Sign Design Principles”, 2000 AT-10