U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
202-366-4000
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
REPORT |
This report is an archived publication and may contain dated technical, contact, and link information |
|
Publication Number: FHWA-HRT-17-048 Date: May 2018 |
Publication Number: FHWA-HRT-17-048 Date: May 2018 |
This chapter provides a literature and policy review, including agency policy manuals and memorandums; standard plans and details; and published, peer-reviewed guidance from technical support organizations.
A variety of sources were researched and reviewed, including standards and guidance documents, published research, conference proceedings, and various other literature on signing or marking for complex interchanges. FHWA has sponsored recent research related to complex interchanges, including Simulator Study of Signs for a Complex Interchange and Complex Interchange Spreadsheet Tool, Driver Expectations When Navigating Complex Interchanges, and Collecting and Analyzing Stakeholder Feedback for Signing at Complex Interchanges.(1–3) The research had the following objectives:
This report is intended to complement the previous three efforts, rather than duplicate the literature review and information-gathering efforts already performed. Therefore, the report focuses on identifying the critical elements of these previous efforts and other key pieces of literature in addition to summarizing existing guidance and state of practice for exit signing and markings at complex interchanges. Combined, these elements will influence how the remaining project tasks are conducted. The key findings are summarized in the subsequent sections of this chapter, which is organized as follows:
The following subsections focus on driver expectations and perceptual factors.
Drivers have expectations for many aspects of driving, including speed, traffic, the roadway’s geometry, and the information supplied to them—including when, how, and where that information will be provided. Roadway conditions that contribute to driver expectations include roadway alignment, width, shoulders, surface texture, and signs and markings.(4) Violations of driver expectations can increase driver confusion, frustration, and workload, which may lead to navigational errors or potentially unsafe driving behavior (e.g., speed variability or erratic maneuvers). Driver expectancy, therefore, has a major impact on highway safety and operations.
Understanding the drivers’ expectations provides information on the types of scenarios that may contribute to the overall complexity of an interchange and serves as a guide when designing treatment options for evaluation.
Russell (1998) describes the technique of “commentary driving,” in which verbal comments are made while driving to indicate initial expectations of drivers and when those expectations are violated. When used appropriately, such techniques can be used to flag potential problem sites that may require additional evaluation.(4) Russell also identified the following factors that affect the information needs of drivers:(4)
Richard and Lichty (2013) conducted a thorough literature review of previous work on driver navigation problems and driver expectations at interchanges and intersections.(2) Various measures are identified that can be used to record driver behavior and expectations, including driver errors (e.g., number of missed exits, unnecessary lane changes (ULCs), and erratic maneuvers), operational measures (e.g., lane change distance, response time, and vehicle speed), and subjective measures (e.g., sign expectations and certainty of choice selection). The researchers employed these performance measures to make the following inferences about driver expectations:(2)
Although such performance measures can be used to deduce driver expectations, it is difficult to draw direct conclusions about expectations without specifically asking about them as they are often intertwined with performance factors (i.e., sight distance) and driver motivations, preferences, and familiarity, among other factors.(2)
In addition, little research explicitly identifies driver expectations for specific interchange elements. Nonetheless, overall themes of design principles or guidance can be applied to interchange design and, thus, will be useful in subsequent tasks of the current project. Richard and Lichty (2013) identified 10 principles based on a thorough review of existing research; these design principles and supporting research are shown in table 4.(2)
In an effort to isolate driver expectations in specific complex interchange scenarios, Richard and Lichty (2013) developed video scenarios based on geometries and interchange elements identified in the literature review and complexity factors identified by Fitzpatrick et al. (2013).(1,2) These videos were used as part of a focus group to identify the sources of expectation-related problems that drivers encounter and potential solutions/countermeasures that drivers suggest. The project team identified the following key driver expectations for the navigation of complex interchanges:(2)
As stated previously, drivers form expectations about where and how information is provided to them. This indicates a need for consistency, and if there is consistency in signing within and between States, then signing will more reliably meet driver expectations. In addition, drivers will form expectations about the upcoming geometry of the roadway and the actions they should take based on the design and placement of signs and markings. The location and layout of a sign, as well as how information is grouped within a sign, will all influence how drivers interpret the sign.(1,2) These factors must, therefore, be considered during design and implementation to ensure that the signs will be effective and intuitive/usable to drivers. This is especially important for complex situations that may be inherently confusing to drivers. The following examples are perceptual factors that may affect how drivers interpret signs.
Lateral Sign Placement
One study compared a yellow left-exit panel at the bottom of the guide sign and a yellow left-exit-number plaque on the top of the guide sign to a plain/green exit-number plaque atop the guide sign. Although the green left-exit plaque was less noticeable to drivers than other left-exit notations, sign placement on the sign bridge was a stronger cue than the left-exit notation; most participants remained in the rightmost lane when viewing the left-exit panel, indicating that the left-exit panel may be confused with an exit-only panel.(2) Similarly, a simulator study showed minimal difference in a yellow left-exit plaque at the top left of the sign and a yellow left-exit panel at the bottom of the sign when both signs were placed above the left lane.(1) These findings indicate that the lateral position of the sign as well as the placement of information within a sign can influence driver lane selection and expectations about the geometry of the roadway.
This has implications for other signing concepts as well (i.e., sign spreading). Some research has shown that spreading a lot of information across multiple signs on a sign bridge can lead to incorrect lane changes or ULCs as drivers may position themselves underneath the sign that contains their intended destination.(2) Therefore, the lateral location of pull-through signs on a sign bridge is important.
Separation and Organization of Information
Separation cues and lateral placement of destinations on a sign can influence how drivers associate destinations with specific lanes. For example, Richard and Lichty (2013) examined different types of destination separators on guide signs for a two-lane exit that divides after the exit to determine whether the sign indicated an immediate split or a downstream split (i.e., they could use either lane to exit, and then, they would need to change lanes after the exit).(2) In each condition, the exit panel extended the full length of the guide sign so that both destinations were above the exit panel. The researchers then examined multiline separation (destinations stacked vertically on top of one another), vertical separator lines (destinations on the same horizontal row, separated by a vertical line), and hyphen separators (destinations on the same horizontal row, separated by a hyphen). Multiline and hyphen separators were found to be better than vertical separators at communicating that both lanes would allow drivers to reach both destinations; the vertical separator lines caused drivers to think they had to change lanes immediately to reach their destination.(2) Similarly, a simulator study examined three signing conditions for a Y-split: a split-sign configuration in which all three signs used vertical separator lines, a multiline separation configuration in which all three signs used a vertically stacked format, and a condition in which the two advance signs used the multiline separation, and the sign at the gore used the vertical separator lines. The results indicated that the lateral location of the destination on the sign influenced drivers’ lane-changing decisions.(1)
It is important to note that some sign elements (e.g., vertical separator lines) may have different effects when used in different scenarios and in combination with different sign elements. For example, a study by Katz et al. (2014) evaluating different sign elements for combined lane use and destination signing indicated that the presence of vertical separator lines did not have a significant effect on driver comprehension of guide signs.(9)
Visual Presentation of Lane Information
The visual perspective on the alignment of lane information can also influence driver behavior and lane selection.(2) Some research has shown that conventional guide signs may produce fewer lane placement errors than certain types of diagrammatic signs, or that modified diagrammatic signs providing separate arrows for each lane (i.e., arrow-per-lane (APL) signs) are more effective in communicating lane assignment than the diagrammatic sign and other conventional signing options.(10–12) Furthermore, some research has examined driver interpretation of such signs, and lane movements may also vary depending on the particular situation. For example, modified diagrammatic signs may be better than diagrammatic signs in some interchange scenarios, but not in others, or that visual cues (e.g., the number of lanes) may influence the distinctiveness and effectiveness of a diagrammatic sign.(2,13)
As indicated in previous sections, driver expectations and perceptual factors that influence guide sign interpretation can affect driver behavior and comprehension of the roadway geometry. Such considerations should be taken into account in the design of signing and markings to make them more intuitive and useful to drivers; failing to do so can lead to unnecessary or incorrect lane changes, erratic maneuvers, and driver confusion. These factors should not only be considered in the design process, but should also be further explored as they could help identify minor changes in current designs that could be used to improve signing at complex interchanges.
Doctor, Merritt, and Moler (2009) describe a complex interchange as “a facility that typically contains many lanes, usually four or more in each direction, and carries high traffic volumes through a maze of tightly spaced ramps and connectors.”(14) Complex interchanges do not typically have conventional layout patterns; instead, each interchange is unique, and thus, the current guidance and practices are not always sufficient for complex conditions.(14) Because each interchange is unique, the challenges for complex interchanges are often the result of the unique interaction between multiple components, rather than one particular scenario. These components may include roadway geometry variables, signing and markings, traffic volume, driver-expectancy violations, and driver workload. Therefore, it is difficult to explicitly define a complex interchange.
A variety of resources address the geometric and signing conditions that contribute to interchange complexity. For example, the Guidelines for Ramp and Interchange Spacing (Ray et al. 2011) discusses complex geometric situations and topics, such as ramp and interchange spacing, collector–distributor (C/D) roadways, ramp braids, and weave sections, and provides guidance in evaluating various safety and operational considerations of potential solutions.(15)
Lichty, Bacon, and Richard (2014) gathered feedback on complex interchanges through telephone interviews with stakeholders representing 17 State transportation departments and through a web activity with stakeholders from 32 regions. Stakeholders, including roadway engineers and other stakeholders who have responsibilities related to interchange planning, design, or maintenance, identified specific elements that cause problems for drivers and, thus, contribute to the complexity of an interchange, which are shown in table 5.(3)
HOVs = high-occupancy vehicles.
Stakeholder feedback revealed that multiple routes converging or diverging within a short distance are a common indication of a complex interchange.(3) The stakeholders were also asked to discuss examples of when they had to address a problem at a complex interchange and identify HFs challenges that they encounter at such interchanges. The following topics were mentioned most often:
Other topics included left/right exit plaques, route continuity, diverging diamond interchanges, lane balance, and arrows on signs.
Closely spaced interchanges, routes, or access points are a common problem because drivers have to make more decisions in a shorter amount of time and engineers are more constrained in the amount of space they have to provide information.(3) Such circumstances lead to other challenges, such as short weaving sections, C/D roadways, exit lane splits, information overload, multiple routes that run concurrently, lack of space for advance signing, density of access points, and the number of decisions the driver is required to make at the interchange.
Fitzpatrick et al. (2013) developed a complexity rating tool to evaluate the complexity of an interchange. The researchers compiled an initial list of noteworthy variables for inclusion in the tool and then held an expert panel discussion to finalize the list of variables that contributed to interchange complexity. The resulting tool focused on three interchange-wide characteristics, a selection of cross-section characteristics at the terminus of the speed-change lane of each ramp, and ramp-specific characteristics that are dependent on whether the ramp is an entrance or exit ramp.(1) The tool contains threshold values for scoring each variable and weights to assign relative importance and measure of complexity for a given factor (i.e., factors with higher weights are considered to have a greater impact on the complexity of an interchange than factors with lower weights); the 32 factors and their weights are shown in table 6.(1)
Lane continuity violations and weaving sections (less than 0.5 mi in length) were given the largest weights as they were considered to be the biggest contributors to driver workload and perceived complexity.(1) Other factors with higher weights include number of entrance ramps per mile, ramp densities, number of general-purpose lanes, loop ramps or curved approaches to ramps, left-side ramps, and ramps with multiple destinations. Similarly, stakeholder feedback revealed that multiple routes converging or diverging within a short distance are a common indication of a complex interchange.(3)
Additional research and crash data support that factors such as interchange spacing and ramp characteristics play a significant role in interchange complexity. For example, research in California and Washington State shows that inserting an additional interchange between existing interchanges could increase freeway fatal injury crash frequencies by as much as 88 percent.(17) This is not surprising, as decreased spacing between interchanges will likely increase driver workload. Some other crash-related findings include the following:
Design guidance for interchange type selection, layout, and geometric design can be found in several key resources, including AASHTO’s A Policy on Geometric Design of Highways and Streets, commonly referred to as the Green Book.(16) Many States base their State design manuals on information contained in the AASHTO Green Book. In the United Kingdom, the resource that contains similar types of information is the Design Manual for Roads and Bridges, a document that also forms the basis for Middle Eastern and some South Asian design manuals.(21)
The design of TCDs for interchanges is typically undertaken in a separate process using separate design documentation. The Manual on Uniform Traffic Control Devices (MUTCD) contains illustrations and text identifying the requirements and recommendations for signing at interchanges.(22) In general, MUTCD language does not consider entire systems as a whole, although the illustrations indicate a systematic approach to typical scenarios. MUTCD users must select discrete TCDs based on individual expertise with the process of designing for particular conditions.
In the process of exploring how design guidance is selected for interchange projects, the project team sought feedback from practitioners to discuss their experiences in locating, screening, using, and interpreting resources.
A leading reference for geometric design guidance is the AASHTO Green Book.(16) Support for design decisions and methodology for selecting cross-section and geometric features are provided in other documents, including the following:
In addition, the FHWA’s National Highway Institute offers various training classes to help increase awareness and application of many of these additional resources. With some transportation agencies facing staffing challenges, practitioners may lack the technical background necessary to effectively design freeway interchange components, particularly traffic signing. Designers who do not have a strong background in the application of traffic engineering principles and an understanding of HFs may be unaware of these resources. Even with knowledge of these resources, recommended practices may be misapplied in complex situations, leading to poor choices in the selection, layout, and fabrication of freeway guide signing.
For the geometric design of interchanges, the AASHTO Green Book addresses interchange configuration and ramp geometry in several sections, including information in section 10.1.(16) Section 10.9.3 of the Green Book provides information on four-leg interchange designs but does not specifically address how those interchange configurations might be modified to fit local conditions in ways that introduce complexity.
Section 10.9.5 of the Green Book addresses issues related to driver expectation, such as route continuity, lane balance, and the selection of auxiliary lane termination designs.
Section 10.9.6 of the Green Book addresses issues related to interchange complexity, including ramp terminal spacing and ramp terminal design. In particular, figure 10-73 and the associated text describe the problems associated with the “tapered design” for multilane entrance ramps. The Green Book cautions against using the tapered design, as it causes the “inside merge,” or “forced merge,” wherein two vehicles compete for the same space with no clear assignment of right-of-way and no escape option (e.g., a shoulder).
In addressing the use of multilane exit ramps, the Green Book discusses the lane changing required for the parallel design, while discussion concerning issues related to signing of option lanes and the forced merge tapered design is notably absent.
While not specifically addressing signing, the Green Book does include several photos that illustrate important design features, including signing.(16) In some instances, the photos in the Green Book may not be consistent with practices from the MUTCD.(22)
The MUTCD lists freeway signing treatments that potentially reduce driver workload and provide pertinent information on movements. The following items, which are from section 2E.07 of the MUTCD, are intended for application as operational needs warrant but are all considered applicable to complex interchange mitigation methods:(22)
The project team conducted a cursory examination of the intersection design practices of roughly 10 States/provinces, representing locations across the United States and in Canada. The examination of the interchanges considered such design elements as layout and ramp geometry and the application of traffic signing, pavement markings, and other elements to the system. These elements include upstream lane additions and eliminations, longitudinal origin and method of addition or elimination of lanes, lane type or function, and lane reduction in close proximity to exiting traffic.
Beginning in the 1960s, interchange designs were typically characterized by ample accommodation of ramps, wider median strips, and potential accommodations for additional lanes in future years. However, often, current traffic volumes have exceeded the projections from that era, and ramps have capacity constraints. Reconstruction of these ramps and the addition of lanes can create geometric constraints, particularly for lane and shoulder width, sight distance, and lane arrangements upstream of decision points.
In the 1980s and 1990s, interchange reconstruction projects were often constrained by funding sources. Environmental considerations, including community opposition to freeway construction or expansion, often resulted in the construction of an improvement that was less than optimal and constrained for the construction of future capacity, typically because of bridge abutments, retaining walls, and noise barriers.
The project team considered an examination of these overarching trends an important aspect of understanding how design processes are influenced by prevailing trends in the cultural, economic, and policy realms. Those trends can cause transportation officials to select designs that may not satisfy all desired goals of the project.
In researching prevailing practice, the project team examined four aspects of interchange design and operations: interchange layout, interchange modification strategies, interchange signing practices, and interchange pavement marking practices.
Interchange layout, in reconstruction projects, is particularly constrained by adjacent land uses and development. Increasingly, designers are turning to more costly methods of providing sufficient space for ramp connections, including grade separations, large retaining walls, and fill sections.
As interchange reconstruction projects occur, optimal ramp spacing, configuration, and signing are often challenging to provide. Removal of existing ramps may be fraught with political difficulties and efforts to reduce cost, and impacts may result in decisions to not use design features that have the potential to reduce interchange complexity such as braided ramp sections and additional lanes. The proper selection of interchange type and differentiation between service interchanges (those serving arterial roadways with ramp junctions that are not free flowing) and system interchanges (those serving intersecting freeways and expressways and characterized by free-flow ramps) is also key in preserving system performance.
When confronted with poor traffic operations performance and safety deficiencies, agencies may seek to modify existing interchanges. In some instances, these modifications may make an interchange previously not considered complex into one that exhibits characteristics of a complex interchange. Some combinations of interchange characteristics may yield more undesirable results regarding safety performance or operations than if the two characteristics were separately implemented. Determining the safety performance of interrelated elements can often be difficult; as such, determinations are rarely addressed by design guidance for specific instances, although some scenarios are addressed. One example would be the application of information on ramp terminal spacing. The development of an interchange complexity checklist may be one tool that could assist the practitioner in avoiding combinations of features that exhibit poor safety performance while at the same time illustrating practices that demand consideration.
In the survey of practice, the project team identified 12 interchange modification strategies applied to existing infrastructure where the core configuration of the interchange remained unchanged:
Evaluation of current traffic signing practices in the United States and Canada and countries with available published documentation was conducted to help identify areas where significant variations in practice were observed. The following six areas of significant concern were identified:
The following are six other practices of concern for interchange complexity that were also examined, but not in detail, for this report:
The review of agency practices indicated that agency proficiency with freeway operations appeared to be highly correlated with the use of pavement marking treatments that improve motorist comprehension of lane use and freeway geometric changes.
In particular, the use of “drop line” (wide dotted lane line) markings, dotted extension lines, and gore markings is critical to providing for reduced workload in the “guidance” portion of the driving task. FHWA HFs research posits that there are three tasks related to the operation of a vehicle. Roadway users navigate within the network, using guide signs and other information, and choose a path of travel based on pavement markings, regulatory and guide signs, and other roadway and roadside features. The physical operation of the vehicle, the “control” task, requires appropriate inputs associated with the guidance task.
Typically, a distinction is made between the dotted lane line markings and the dotted extensions. In Washington State, Minnesota, and several other States, the dotted lane line markings are used only adjacent to full-width non-continuing lanes. In some States, their use is restricted to exit-only lanes, so as to not create confusion between exit-only lanes and lanes subject to a downstream lane reduction. The Minnesota Department of Transportation (MnDOT) and Washington State Department of Transportation (WSDOT) call for a 3-ft line with a 12-ft space, permissible per the guidance statements in MUTCD sections 3A.06 and 3B.04. Other States, such as North Carolina, do not distinguish between the dotted lane line and dotted extensions in pattern or width, and some States have eliminated the narrower, closely spaced pattern of the dotted extension (typically a 2-ft line with a 6-ft space) in favor of using 3-ft lines with 9-ft spaces for all dotted markings. Ongoing TCD pooled fund study efforts may help provide the research results that justify additional design guidance and changes to the MUTCD.
Reviews of pavement marking practices indicated that some agencies are extremely proficient in providing uniform markings to indicate the present and downstream status of individual lanes; the delineation associated with exit and entrance ramps; and the marking of gore areas, particularly those in areas with horizontal alignment changes. However, the project team also noted several inconsistent pavement-marking applications, particularly as they are associated with complex interchange features:
©Esri.
Figure 2. Photo. Satellite imagery of northbound Interstate 5 (I-5) C/D roadway north of I-90 with multiple successive lane-reduction arrows.(28)
Source: FHWA.
Figure 3. Photo. Washington State Route (SR) 520 westbound near 92nd Avenue northeast.
In several States, the project team noted that practices differed between locations, between installations in different time periods, and between regions or districts within a State. While some variations are to be expected, variations in projects in the same administrative region and along the same freeway corridor point to issues with consistent practice within an agency.
Based on a cursory review of agency practices and using notes prepared on previous projects, particularly projects involving the preparation of contract plans, the project team identified the following as potentially contributory to inconsistent application of design principles in interchange design:
The findings of this literature and standards review will help guide the efforts of the current project as important complex interchange characteristics are identified, types of challenges that need to be addressed within these interchanges are analyzed, and strategies to address the constraints associated with complex interchanges are identified, while also seeking to improve uniformity and decrease driver confusion.
Previous research and feedback from stakeholders has identified specific scenarios that make interchanges and common challenges associated with them (e.g., close spacing of interchanges results in lack of space for advance signing, density of access points, and short weaving sections) complex, as well as some practices that have been implemented to try and address these challenges.
Previous complex interchange research efforts have also identified driver expectations and subsequent design principles and have begun to work toward identifying solutions for problems that drivers encounter. This information can be used to identify gaps in research or areas that would benefit from additional research and applied to existing complex interchanges. Understanding driver expectations will not only provide insight into what types of scenarios may contribute to the overall complexity of an interchange, but will also help identify modifications that may unintentionally increase the complexity of an interchange.
Driver expectations and design principles will also be used as a guide when designing and implementing signing and marking strategies for evaluation to ensure that signs are intuitive and usable to drivers. In addition, this information will be useful when creating the resulting guidance, visualizations, and decision criteria that practitioners can use. Although guidance that can be adapted to a variety of scenarios will ultimately be developed, it should be consistently based on key design principles and driver expectations. Multiple suggestions and decisionmaking tools that practitioners can use to reduce driver confusion in an attempt to reduce the challenges presented in a complex scenario should be included.
As potential strategies are identified, one should also look into possible modifications that can be made to make them more cost effective, while still providing the same benefits. For example, the current arrow standards for the APL-guide signs require larger signs and structures that may not be financially feasible in some situations. In this case, a reduced arrow size could be evaluated to determine if the signs would still be effective if implemented at a smaller size. Previous stakeholder feedback has already identified some constraints that practitioners face, and stakeholder interviews and working group discussions will help identify additional modifications that could be made to ensure that the signing and marking strategies will be useful to practitioners.