Roundabout Practices (2016)

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.

4 States has increased in tandem with the development and pub- lication of design guides and research (4). Types of Roundabouts Although the first two roundabouts constructed in the United States were multilane, the Kittelson & Associates website indicates that the majority of roundabouts constructed since have been single-lane roundabouts—as of 2013, 71%. How- ever, this is a decrease from 2005, when approximately 78% of known roundabouts in the United States were single-lane. The first known mini-roundabouts in the United States, 17 of them, were constructed in 2000; However, FHWA has only identi- fied 12 mini-roundabouts as of December 2013 (5). Figure 6 displays the cumulative percent of roundabouts constructed in the United States by type. (Unlike Figure 4, Figure 6 only includes roundabouts where both the roundabout type and year constructed are known.) Although all roundabouts are generally circular—Figures 7, 8, and 9 show mini-, single-lane, and multilane roundabouts, respectively—roundabouts exist that are not pure circles, with oblong, elliptical, raindrop, and dumbbell shapes (6–8). Examples include the oblong-shaped Rochester roundabout shown in Figure 10, located in Rochester Borough, Pennsyl- vania (10). The I-135/1st Street roundabout in Newton, Kansas, shown in Figure 11, is an example of an elliptical roundabout (11). The I-70/Avon Road roundabouts shown in Figure 12 displays two examples of raindrop-shaped roundabout designs located at the interchange ramp terminals (12). A series of dumbbell-shaped roundabouts are located on Keystone Park- way in Carmel, Indiana; the Keystone Parkway/E. 116th Street intersection dumbbell roundabout, as shown in Figure 13 (13). ROUNDABOUT POLICIES A 2010 study of existing statewide policies was reviewed to assess the status of roundabout policies in the United States (14). The 2010 review was conducted by examining infor- mation available online, including state agency roundabout websites, or other state DOT web pages. The review split the roundabout policy types into six categories. For the purposes of this synthesis, the two strongest roundabout categories (“Strong” and “Justify”) were combined into a new category termed “Preferred.” The five categories are shown in Table 1. ROUNDABOUTS IN THE UNITED STATES Although the first traffic circle in the United States was built at New York City’s Columbus Circle in 1905, with one-way circulation, the roundabout was officially invented in 1966, when the off-side priority rule (meaning entering drivers yield to circulating drivers) was adopted in Great Britain. Accord- ing to NCHRP Synthesis 264, roundabouts were constructed in Australia and France in the 1970s, and were officially moved into practice in France in 1984. In the meantime, however, in this country, traffic circles fell out of favor in the 1950s, as a result of operational and safety problems associ- ated with rotary intersections, including high speeds, weaving issues, and the tendency to lock up at higher volumes. The first two modern roundabouts in the United States were con- structed in 1990 in Summerlin, Nevada (1). This chapter discusses the growth of roundabouts in the United States since 1990, along with current statewide round- about policies; statewide statutes, codes, and laws pertaining to roundabouts; public outreach efforts; current research on roundabout performance; and a snapshot of the most effec- tive practices related to the design of roundabouts. Growth of Roundabouts In the first 10 years following the introduction of roundabouts in Nevada, only 30 states were known to have built at least one roundabout, as shown in Figure 2; and only four states were known to have had more than 25 roundabouts. After the open- ing of two roundabouts in Fairmont, West Virginia, in 2010 (3), every state and the District of Columbia had at least one roundabout, as shown in Figure 3. By 2013, approximately 3,200 roundabouts are estimated to have been constructed (2). In general, roundabouts have been constructed at an increas- ing rate per year since 1990, as shown in Figure 4. The data in this figure comes from an online roundabout database from Kittelson & Associates (3) and has been adjusted to reflect the likely number of roundabouts constructed in the United States by providing assumptions for the number of unknown roundabouts. These assumptions are based on the number of roundabouts missing from previous years’ records compared with the number of roundabouts that are now known to exist, and does not include “nonconforming” traffic circles (2). As shown in Figure 5, the number of roundabouts in the United chapter two LITERATURE REVIEW

5 FIGURE 2 Location of roundabouts in the United States as of 2000. Source: Rodegerdts (2014) (2).

6 FIGURE 3 Location of known roundabouts in the United States as of 2013. Source: Rodegerdts (2014) (2). FIGURE 4 Estimated cumulative roundabouts in the United States. Source: Rodegerdts (2014) (2).

7 FIGURE 5 Timeline of roundabouts implementation and guidance development in the United States (4). 0% 20% 40% 60% 80% 100% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 Single-Lane Multilane Mini-Roundabout FIGURE 6 Cumulative percent of known roundabouts by type in the United States as of 2013. Source: Rodegerdts (2014) (2). FIGURE 7 Mini-roundabout, Vierling Drive/County Road 79 intersection in Shakopee, Minnesota. Source: City of Shakopee (7). FIGURE 8 Single-lane roundabout, 133rd Street/Lamar Avenue intersection in Overland Park, Kansas. Source: City of Overland Park, Kansas (8).

8 Based on the categories displayed in Table 1, each state, along with the District of Columbia, was assigned to a group. Information from the 2010 review and assignment was updated for this synthesis. It can be noted that the categorization of roundabout policy type was somewhat subjective, especially in the division between “Allow” and “Encourage”—but was somewhat more straightforward in the difference between “Encourage” and “Evaluate,” which was often as simple as the difference between “should” and “shall,” respectively. An example of the policy text associated with each policy type category is given in Table 2, which lists example roundabout policy types from five states. The number of states with each policy type was tabulated, and is displayed in Table 3. As shown, the most prevalent policy types were “Allow” and “Encourage,” with 13 and 18 states, respectively. Currently, 13 states formally require the analysis of a roundabout alternative as denoted by the “Evaluate” and “Preferred” policy type categories. The remaining eight states have no formal or informal roundabout policy. The policy type categories were mapped in order to denote regional roundabout policy type trends (Figure 14). The 2010 study on roundabout policies correlated the number of roundabouts in the state to the strength of the state- FIGURE 9 Multilane roundabout, SW Stafford Road/SW Borland Road intersection in Tualatin, Oregon. Source: Kittelson & Associates, Inc. (9). FIGURE 10 Rochester roundabout, Brighton Avenue/Adams Street (SR-68)/Rhode Island Avenue (SR-18) intersection in Rochester Borough, Pennsylvania. Source: Whitman Requardt & Associates, LLP (10). FIGURE 11 I-135/1st Street roundabout in Newton, Kansas. Source: Burns & McDonnell (11). FIGURE 12 I-70/Avon Road roundabouts in Avon, Colorado. Source: Google (12). FIGURE 13 Keystone Parkway/E 116th Street intersection in Carmel, Indiana. Source: Microsoft Corporation (13).

9 Category Description None No policy or mention of roundabouts from the state DOT. Consequently, the state neither encourages nor discourages roundabouts. Allow The state allows the consideration of roundabouts. Encourage The state encourages the consideration of roundabouts. Evaluate The state requires the evaluation of a roundabout alternative. Preferred The state requires the evaluation of a roundabout, and justification when a roundabout is not the preferred alternative. TABLE 1 TYPES OF ROUNDABOUT POLICIES Policy Type Sample Policy Text (Source) None N/A Allow A modern roundabout is an alternative form of intersection control to traffic signals and [multiway] stop control intersections. Therefore, roundabouts may be considered only when these intersection control types are warranted. (Kentucky) Encourage Those locations that meet or nearly meet [signal] warrants should be given consideration for roundabout installation. Intersections that are, or proposed to be, all-way stop controlled may also be good candidate locations for a roundabout. (Connecticut) Evaluate Roundabouts are the preferred safety and operational alternative for a wide range of intersections of public roads. A roundabout shall be considered as an alternative in the following instances: (1) Any intersection in a project that is being designed as new or is being reconstructed. (2) All existing intersections that have been identified as needing major safety or operational improvements. (3) All signal requests at intersections (provide justification in the Traffic Engineering Study if a roundabout is not selected). (Georgia) Preferred When the analysis shows that a roundabout is a feasible alternative, it should be considered the Department’s preferred alternative as a result of the proven substantial safety benefits and other operational benefits. (New York) N/A = Not Available. TABLE 2 EXAMPLE ROUNDABOUT POLICY TYPES Policy Type No. of States % None 7 14 Allow 13 25 Encourage 18 35 Evaluate 8 16 Preferred 5 10 Total 51 100 TABLE 3 SUMMARY OF STATEWIDE ROUNDABOUT POLICIES FIGURE 14 Roundabout policy type by state updated from the 2010 study (14). wide roundabout policy (14). The study offered the following observations: • An established policy typically indicates that a person or persons in leadership capacity are supportive of the policy. • An established policy provides roundabout proponents with a position of influence within the agency by having regulatory backing. • An agency with an established roundabout policy could indicate an organization open to change and innova- tion, which promotes an environment conducive to implementation. • The enactment of a policy could indicate the breaking down of informal internal barriers, which otherwise could hinder the potential implementation. • With more states enacting policies, new and revised poli- cies have precedents for success. • The implementation of a roundabout policy allows for the development of uniform and simplified procedures, standards, and regulations, thereby further increasing the chance for successful implementation of roundabouts. • The establishment of roundabout policy validates round- abouts as an alternative. • A policy helps to ensure the continuation of the round- about program, especially when one of the early round- about champions leaves the agency. • The formation of a policy helps to institutionalize roundabouts in the state DOT’s practice. Several states have adopted intersection control evalua- tion (ICE) policies to provide a more proactive method for

10 evaluating intersection alternatives using performance-based design. Consequently, roundabouts are being considered earlier in the project development process, alongside other intersection treatments such as all-way stop-controlled inter- sections, signalized intersections, alternative intersection forms, and grade-separated forms. The ICE policies are helping to identify short- and long-term benefits that previously were not considered, including the evaluation of life-cycle costs (4). STATE STATUTES, CODES, AND LAWS The Manual on Uniform Traffic Control Devices (MUTCD) (15) requires a yield sign on all approaches to a roundabout. Section 2B.09 YIELD Sign Applications reads as follows: A YIELD (R1-2) sign shall be used to assign right-of-way at the entrance to a roundabout. YIELD signs at roundabouts shall be used to control the approach roadways and shall not be used to control the circulatory roadway. If all roundabout approaches shall have a YIELD sign and all roundabout approaches operate under yield control, then all entering traffic must yield to circulating traffic regardless of whether or not a state has a specific statute, code, or law. Nevada, like many other states, includes information in its Driver’s Handbook (16), as shown in Figure 15. PUBLIC OUTREACH EFFORTS Although the safety and operational benefits of roundabouts are well documented, some states have been slow to build roundabouts. “A major impediment [to the construction of roundabouts] is the negative perception held by some drivers and elected officials” (17), which has been termed “roundabout anxiety” (18) or driver “angst” (19). In addition, confusion about the differences between roundabouts and rotaries, along with negative experience with rotaries, continues to linger (20). As has been demonstrated on countless occasions, the public in most jurisdictions will have a negative opinion of round- abouts before the installation of the first one (21). According to a telephone survey of Bellingham, Washington, area residents prior to the construction of a roundabout, only 36% of those surveyed were in favor (22); one year after construction, the percentage in favor of the roundabout increased to 70% (Fig- ure 16). The Montana Department of Transportation (MDT) developed a synthesis report on education and outreach related to roundabouts (23), confirming public reluctance throughout the United States, and similarly finding that people remain skeptical until they have driven a roundabout. Public outreach for roundabouts comes in many differ- ent formats. Some examples, though not a comprehensive list, are included in the Roundabout Outreach and Education Toolbox on the FHWA Office of Safety’s website (24), which includes a range of options for public outreach, including websites, brochures, flyers, videos, and even coloring books. (The coloring book, developed by the city of Bend, Oregon, educates children on the basics of roundabouts, with a goal of demystifying the newer intersection type for a younger audi- ence.) The Montana synthesis also provides an overview of education and outreach material used throughout the United States, with recommendations for when and where different outreach strategies may be appropriate. Some states are providing information on how to navigate a roundabout: Figure 17 shows an example from the Indiana Drivers Manual (25). States are also using websites to com- municate to the public. The South Dakota DOT (SDDOT) site (26) lists information on: 1. What is a roundabout 2. How to drive a roundabout 3. Roundabout benefits 4. Pedestrians and bicyclists. A screenshot of the SDDOT roundabout website is shown in Figure 18, and a screenshot of the MDT roundabout web- site is shown in Figure 19. ROUNDABOUT PERFORMANCE This section documents methods used by state DOTs to deter- mine the expected operational and safety performance of roundabouts. Operational Performance NCHRP Report 572: Roundabouts in the United States (28) provided the first roundabout capacity model anchored to empirically observed performance in the United States at a national level. The results of NCHRP Report 572 were incorporated into the Highway Capacity Manual 2010 (29). The HCM 2010 procedure derives approach capacity from a regression-based analysis and also incorporates a calibra- tion procedure using gap-acceptance parameters (30). Sev- eral jurisdictions have developed and documented calibration efforts, including the California DOT (Caltrans, 31), city of Bend, Oregon (32), and the Kansas DOT (KDOT, 33). The Wisconsin DOT (WisDOT) also recently completed a study evaluating the operational characteristics of roundabouts (34), and the city of Carmel, Indiana, developed a new model with higher capacities reflective of the local driver population (35). Results from KDOT, WisDOT, and Georgia DOT (GDOT) (36) were within data ranges reported by NCHRP Report 572. However, the cities of Bend and Carmel have documented results indicating higher capacities on roundabouts in their communities, both of which have large concentrations of roundabouts. Table 4 shows a sample of capacity model cali- bration parameters.

11 As of the writing of this synthesis, FHWA is leading a study to update the capacity equations. The results of this research are expected to be published in 2016. Preliminary results pre- sented at the TRB 4th International Roundabout Conference in 2014 (37) indicate roundabouts are operating at greater capacities than estimated by the HCM 2010 model based on the Appendixes to NCHRP Report 572 (38). Initial results indi- cate geometric effects such as inscribed circle diameter, entry lane width, entry angle, and splitter island width are not strong enough to include in the capacity model. Furthermore, the exponential form, consistent with the model form in NCHRP Report 572, fits the same or better than other linear forms. FIGURE 15 Roundabout guidance in the Nevada Driver’s Handbook (16).

12 FIGURE 16 Percent of drivers who support roundabouts—Telephone surveys of Bellingham, Washington, area residents (22). FIGURE 17 Roundabout explanation in the Indiana Drivers Manual (25).

13 Roundabouts in Corridors NCHRP Report 772: Evaluating the Performance of Round- abouts in Corridors (30), published in 2014, studied nine roundabout corridors in the United States. As a result of the study, a methodology for estimating travel speed was devel- oped for inclusion in the HCM. The study identified “a need for corridor-specific evaluation to determine which form of intersection control is preferred on a given corridor.” Lastly, a corridor comparison document (CCD) was developed to pro- vide a framework for comparing the performance of a corridor with roundabouts to a corridor with signalized intersections or other intersection treatments to arrive at a design solution. The CCD includes information on different corridor users, typical performance measures, and examples illustrating the use of the CCD. Roundabout Analysis A 2012 report, Roundabout Software Evaluation (39), eval- uated software tools available for analyzing roundabout capacity. The report found that “a software package that has the capability of performing capacity analysis using U.S. based models is desirable.” The report did not recommend a particular software package or packages, but instead clarified that “many other considerations, unique to each agency, are required in order to arrive at a recommendation for a soft- ware package.” The report provided an evaluation matrix to aid in decision making. ROUNDABOUT DESIGN Although the first roundabout built in the U.S. opened in 1990, the first state DOT design guidance documents were not developed for several years, with Maryland State High- way Administration (SHA) and Florida DOT publishing documents in 1995 and 1996, respectively. The first federal design guide, NCHRP 672: Roundabouts: An Informational Guide (40), followed in 2000. To date, there are still state DOTs that have not incorporated roundabout design into their design manuals and guidance. NCHRP Report 672: Roundabouts: An Informational Guide, 2nd Edition (6), was published in 2010, and officially adopted by FHWA as the second edition of the federal guide. FIGURE 18 South Dakota Department of Transportation Roundabouts website (26).

14 FIGURE 19 Montana Department of Transportation Roundabouts website (27 ).

15 Although some states make reference to NCHRP Report 672 for design guidance, other states go further than just referring to the federal guidance and include some material in their own manual or online, such as what is found in Maryland’s Round- about Design Guidelines (41): The Maryland State Highway Administration (SHA) has adopted the NCHRP Report 672: Roundabouts: An Informational Guide, Second Edition as our Roundabout Design Guideline. The infor- mation contained in this document is considered a Supplement to the NCHRP Report 672 and is intended to document SHA’s approach to the design of roundabouts. Other states have developed standalone design guidance. For instance, the Kansas Roundabout Guide, 2nd Edition (42) states: This guide is intended to provide practitioners and the general public with information and guidance related to roundabouts in the State of Kansas and serve as a companion to NCHRP Report 672: Roundabouts: An Informational Guide, 2nd Edition. Design Vehicle Accommodations The type and size of vehicle that will use the roundabout affects the design of the intersection, as with other intersection forms. For example, the Georgia DOT Roundabout Design Checklist (43) states: It should be recognized that unlike conventional intersection forms (e.g., signalization, stop control, etc.) the configuration and lay- out of a roundabout can be dramatically affected by the results of capacity, fastest path, and truck turning template studies and thus often requires higher level of engineering during the concept phase. Although accommodating larger vehicles is important, there are other factors to consider. Concerning the percentage of trucks at a roundabout, Washington State DOT (WSDOT) guidance suggests, “Designers should also recognize that among that percentage, WB-67s may only represent a small sample of the entire truck volume on any given day.” WSDOT’s guidance allows trucks to encroach on adjacent lanes with the intent to reduce the overall size of the intersection. Figure 20 shows an example of this design technique for both round- abouts and signalized intersections. Illumination Illumination at roundabouts is provided for safety and opera- tions. Very few states have their own unique guidance on illu- mination at roundabouts. For instance, the Kansas Roundabout Guide draws from resources including NCHRP Report 672, which relied on the Design Guide for Roundabout Light- ing written by the Illuminating Engineering Society (IES). AASHTO also provides illumination guidance in the Road- way Lighting Design Guide. Calibration Parameter NCHRP Report 572 (28)* (Nationwide 2007) Bend Roundabout Guide (32) (Bend, Oregon 2010) Caltrans Roundabout Guide (31) (California 2007) Developing Capacity Models for Local Roundabouts (35) (Carmel, Indiana 2011) Wisconsin Roundabout Evaluation (34) (Wisconsin 2011) Georgia Calibration (36) (Georgia 2013) Single-Lane Roundabouts Critical Headway 5.1 s 4.1 4.5–5.3 s 3.4–3.8 s 4.8–5.5 s 4.75 s Follow-up Headway 3.2 s 2.7 2.3–2.8 s 2.1–2.4 s 2.6–3.8 s 3.27 s Multilane Roundabouts Critical Headway Left lane 4.5 s — 4.4–5.1 s — 4.1–4.8 s — Right lane 4.2 s — 4.0–4.8 s — 3.4–4.4 s — Follow- up Headway Left lane 3.4 s — 1.8–2.7 s — 2.5–3.1 s — Right lane 3.1 s — 2.1–2.3 s — 2.2–3.0 s — *Results from Kansas are similar to NCHRP Report 572 (33). — = Not Reported. TABLE 4 SUMMARY OF CRITICAL HEADWAY AND FOLLOW-UP HEADWAY FACTORS FIGURE 20 WB-65 Right-Turn Movement Over-Tracking Comparison (44).