Evaluating the Performance of Corridors with Roundabouts (2014)

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

Evaluating the Performance of Corridors with Roundabouts Chapter 1–Background Page 1-1 CHAPTER 1. BACKGROUND This report summarizes the findings of NCHRP Project 03 100, Evaluating the Performance of Corridors with Roundabouts. The intended audience for this report is researchers, practitioners, and policy makers who establish federal, state, and local guidelines for roundabouts. This introductory chapter presents the problem statement and research objective, scope of study, research approach, and a summary of the literature review conducted for this project. 1.1. PROBLEM STATEMENT Roundabouts are increasingly recognized as an intersection control strategy that can fulfill multiple performance goals related to traffic operation and safety and that meet societal goals related to sustainability, livability, complete streets, context sensitive design, economic development, and others. Some transportation agencies have recently constructed or approved the use of a series of roundabouts on an arterial rather than the traditional solution of coordinated signalized intersections. While anecdotal reports suggest that functionally interdependent roundabouts on a corridor are successful in meeting performance goals, liƒle research has been conducted to objectively determine the efficacy of this alternative as compared to a series of coordinated signalized intersections. The performance of traffic signal systems on arterials is well researched and documented, and methods to predict their performance are well established. Performance measures for isolated roundabouts exist, and safety research has consistently shown that signalized intersections have higher injury crash rates when compared to roundabouts. In contrast, qualitative and quantitative information on the performance of a set of functionally interdependent roundabouts on arterials is lacking. 1.2. RESEARCH OBJECTIVE The objective of this research is to provide traffic engineers, transportation planners, and other practitioners with performance measurement and evaluation methods to comprehensively evaluate the performance of functionally interdependent roundabouts on arterials, thus enabling a comparison with signalized intersections, in order to arrive at a design solution. For the purposes of this research, a series of roundabouts shall include at least three roundabouts that function interdependently on an arterial. The research plan developed to achieve this objective focused on the delivery of two key products: 1. Performance measurement tools and techniques based on quantitative, empirical data that can assist in the evaluation of a roundabout corridor.

Evaluating the Performance of Corridors with Roundabouts Page 1-2 Chapter 1–Background 2. A set of guidelines for corridor comparisons that incorporates both quantitative and qualitative components. The results of this study can be grouped into two major categories: An assessment of the performance of existing corridors that employ a series of roundabouts. This assessment comprises a general evaluation of their development and success, obtained through field observations and interviews with corridor operators, and a detailed data collection effort for operational data to aid in performance prediction. Tools to enable alternatives analysis for corridors employing a variety of intersection control treatments, whether they be roundabouts, traffic signals, or stop control. These tools include an overall framework for comparison (called a Corridor Comparison Document) and predictive tools for operational performance intended to supplement existing predictive tools in the Highway Capacity Manual. As a result of these two categories of research products, this report is a hybrid of (1) content that is intended for inclusion in other major resource documents (e.g., Highway Capacity Manual and Roundabouts: An Informational Guide) and (2) stand alone content in the form of a Corridor Comparison Document that can be used directly. To achieve these objectives, the research team undertook the following broad tasks: Conducted focused outreach efforts with operators of roundabout corridors to understand the actual characteristics and lessons learned. Identified the quantitative elements related to the operational models, data collection, and recommended analysis methodology. Identified the qualitative elements that could augment the quantitative elements and predictive operations models to support corridor treatment evaluations, comparisons, and recommendations. Examples include access management considerations, safety performance, access to non motorized transportation users, constructability, and how well the arterial treatment fits within the broader city design and cultural context. Collected traffic operations field data at nine roundabout corridors using proven and emerging tools and techniques to develop a field data– driven methodology for evaluating roundabout corridors, and compared them to signalized intersection treatments. Developed a predictive procedure for travel time on a roundabout corridor. The procedure incorporates models developed from the field data collected as part of this project, and is presented in a manner consistent with the auto procedure of the Highway Capacity Manual (HCM) 2010 Urban Streets Chapter. The procedure is recommended for inclusion into the next edition of the HCM. Created a practitioner focused guidance framework (called a Corridor Comparison Document) that provides a holistic approach to considering, evaluating, and supporting corridor treatment decisions. The framework

Evaluating the Performance of Corridors with Roundabouts Chapter 1–Background Page 1-3 includes groups of performance measures and prioritized “tiers” of evaluation considerations consistent with multiple corridor contexts. Four example applications illustrate the use of the guidance framework. 1.3. BACKGROUND AND LITERATURE REVIEW Comparatively speaking, the transportation profession’s understanding of signalized intersection corridor operation is more developed than its understanding of roundabout corridors. For signalized intersections, considerable research and experience has gone into evaluating and timing intersections in both isolated and coordinated models. Analytical and simulation modeling has been common for considering the interaction between adjacent signalized intersections, and significant documentation exists for optimizing flow along signalized corridors. Practitioners have a solid base of experience and well developed “gut feels” for how the familiar signalized corridor should operate. In contrast, roundabout evaluations have largely focused on isolated intersections, using only simulation tools with any regularity in practice to evaluate roundabouts in corridors. Many practitioners have never seen a roundabout corridor in person, much less have a “gut feel” for how it would operate. The intent of this project is to close that gap for practitioners by improving the basis for good decisions. A variety of analytical methods and tools are available to quantify projected roundabout intersection operations, yet prior to this project very li‚le documentation was available that quantifies the operational a‚ributes of roundabout corridors. Data collection for this project focused on elements affecting traffic operations, so that a predictive model for operations could be developed. Other project activities, such as corridor owner interviews, were more comprehensive and included elements such as planning, pedestrian and bicycle user experience, public involvement, construction, and maintenance. 1.3.1. SAFETY AND ACCESS MANAGEMENT CONSIDERATIONS Roundabouts have well documented safety benefits compared to other types of traffic control, and these safety benefits are the predominant a‚ractiveness compared to signalized intersections (Gross et al. 2012, NCHRP Report 572, NCHRP Report 672, Persaud et al. 2001). Safety relationships of isolated roundabouts are likely to transfer to roundabout corridors; there are no specific characteristics of roundabouts in series diminishing the safety performance of the roundabout junction itself. On a corridor level, roundabouts create more access management opportunities compared to signalized intersections. One key differentiating consideration between corridor types may be safety at midblock access points. Opportunities to use roundabout U turning qualities could potentially eliminate left turns to or from driveways along the corridor. Reducing turns at driveways would reduce vehicle conflicts at these locations and positively influence overall corridor safety performance. In addition to reduced conflicts, depending on the spacing of the roundabouts, segment operating speeds could be reduced compared to signalized corridors and, therefore, could reduce crash severity. Roundabout

Evaluating the Performance of Corridors with Roundabouts Page 1-4 Chapter 1–Background corridors could possibly reduce crash frequency because slower operating speeds decrease stopping sight distance requirements and, therefore, increase the opportunity to avoid crashes. Access management principles are well established in the literature (e.g., TRB Access Management Manual). 1.3.2. OPERATIONAL CONSIDERATIONS The transportation profession has an extensive body of knowledge for traffic operations at isolated roundabouts and urban corridors with signalized intersections. Lile research has been documented on traffic operations on roundabout corridors, particularly in the United States. Roundabout corridors have unique operational characteristics compared to their signalized intersection counterparts. Fundamentally, the notion of moving platoons of vehicles to maximize the performance efficiency of signalized intersections is not applicable to roundabouts, where gap acceptance principles allow more dispersed flows to mingle within the intersections. Travel time is a natural performance measure for roundabout and signalized corridors. Roundabouts have increased geometric delay compared to signalized intersections by virtue of their shape; therefore, defining travel time performance measures is of paramount interest. 1.3.2.1. Roundabouts in Isolation Prior to development of models based on observed performance of roundabouts in the United States, operational analysis of individual roundabouts in the United States has been mostly conducted using methodologies and software developed internationally. There have been some contributions from countries such as France and Germany, but methods from the United Kingdom and Australia have dominated US practice. There are conceptual differences between UK and Australian schools of thought discussed below. The UK’s Transport Research Laboratory (TRL) developed their capacity analysis techniques using empirical regression methods (Kimber 1980). Within this methodology, roundabout approach capacity is highly dependent upon geometric features of entries, such as the width, radius, and angle. The methodology was developed based upon extensive field observations of near capacity roundabouts in the UK. The RODEL and Arcady software packages implement the results of the TRL research findings. The most recent version of Arcady at the time of this publication includes a “linked roundabout” feature for analyzing adjacent roundabouts. Arcady adjusts flow entering the downstream roundabout based upon operations at the upstream roundabout. SIDRA, an Australian software package developed by Akçelik and Associates Pty Ltd., analyzes roundabouts as well as stop controlled and signalized intersections. Within SIDRA’s methodology, roundabout capacity is primarily a function of gap acceptance (i.e., entering vehicles accepting gaps in the flow of circulating traffic). At one time, practitioners in the United States generally applied a capacity reduction factor (“environmental factor” in SIDRA’s terminology) of 1.2 to account for observed reductions in capacity compared to

Evaluating the Performance of Corridors with Roundabouts Chapter 1–Background Page 1-5 Australian roundabouts. However, the most recent version of SIDRA at the time of this publication incorporates the methodology of the HCM 2010 directly. The HCM 2010 (TRB 2010) contains a roundabout analysis procedure based upon data collected in the US as part of NCHRP Report 572. The methodology is limited to one or two lane roundabouts with no more than four legs. This procedure took a hybrid approach, where approach capacity was empirically derived from regression based analysis, while also incorporating behavioral gap acceptance parameters that can be user calibrated. Regardless of the way capacity is estimated, all analytical methods principally compare approach capacity to approach volume. They use equations to then predict performance measures such as approach delay and vehicle queuing. Mauro (2010) compiled a selection of analysis techniques for roundabout capacity and performance. Those techniques contain calculations for additional service measures, such as queue length and waiting time, which are varied based on the level of saturation at the intersection. He also discusses time spent in the intersection, which contributes to calculating a level of service for the roundabout using the methodology in the 2000 edition of the HCM. He ultimately uses his method for determining capacity to estimate a roundabout’s reliability (i.e., the probability that the intersection does not fail and that demand does not exceed the capacity of any single entry). While the HCM methodology has changed for the 2010 edition, queue length and waiting time remain valid measures for consideration in any roundabout analysis procedure. Few of the international or domestic procedures explicitly account for the impacts of adjacent intersections (including roundabout intersections), nor do they provide a means of analyzing multiple roundabouts at once to gauge cumulative performance. Users analyzing a roundabout corridor as a series of isolated roundabouts may not account for platooning and queue spillback effects. Additionally, there is no means of assessing corridor wide metrics such as travel time. Some practitioners in the United States have used microsimulation software (such as VISSIM and Paramics) to analyze individual and multiple roundabouts. This does represent a means of analyzing roundabout corridors, but like all applications of microsimulation it requires more time and specialized skills on the part of the analyst compared to other analysis tools. 1.3.2.2. Roundabout Corridors A study of a roundabout corridor in Golden, Colorado (Ariniello, 2004), reviewed crash rates, operating speeds, travel times, and sales tax revenue along the corridor. A portion of South Golden Road between Ulysses Street and Johnson Street was considered for study, where four roundabouts were installed in a corridor of approximately a half mile in length. The five lane corridor served several residential areas and many businesses, including several fast food restaurants, a large grocery store, and a small shopping center. The composition of the traffic mix was not specified, but a high number of driveways were identified in the report, suggesting that turning traffic was substantial and included large delivery and service vehicles in addition to passenger vehicles

Evaluating the Performance of Corridors with Roundabouts Page 1-6 Chapter 1–Background carrying customers and residents. The site description specifically mentioned the existence of horse trailers entering and exiting a veterinary office in the corridor. Ariniello concluded that installing the roundabouts resulted in slower speeds between major intersections in the corridor, but there were also lower travel times compared to when the corridor was signalized (reduced from 78 to 68 seconds through the corridor). The analysis also revealed less delay at business access points. Before installing the roundabouts, the average measured delay was 28 seconds with a maximum of 118 seconds. After the installation, the average delay was reduced to 13 seconds with a maximum of 40 seconds. Between 1996 and 2004, traffic volumes increased from 11,500 to 15,500 vehicles per day, while the number of annual crashes dropped from 123 to 19. Calculated crash rates declined by 88 percent, from 5.9 to 0.4 crashes per million vehicle miles; injury crashes were reduced from 31 in the three years prior to installation to one in the 4.5 years after—a 93 percent decline in injury crashes. Sales tax revenue along the corridor increased 60 percent and 75,000 square feet of retail/office space was built after installation. Isebrands et al. (2008) reviewed corridors in Brown County, Wisconsin, and Edina, Minnesota. They found total crashes at one of the Wisconsin roundabouts were reduced by one per year and injury crashes were nearly eliminated. Another roundabout in the Wisconsin corridor did not have enough data after installation to make a definitive conclusion on crashes. Access management treatments and a series of three roundabouts were used along the Minnesota corridor to address traffic operation and safety performance. Although the roundabouts were open for only a relatively short time when the study was conducted, the city indicated to Isebrands et al. that vehicle operations improved from levels of service (LOS) between B and F prior to opening to LOS ranging from A to D after opening. They also found no reduction or change in access to local businesses. 1.3.2.3. Signalized Corridors Numerous studies on the operations of signalized corridors have been conducted and documented. Signalized corridor analysis is a mature area of study, and the fundamentals of signalized corridor operation can be related to those of roundabout corridors for metrics such as travel time and delay. Signalized corridor studies provide relevant information to be˜er understand and compare the performance of roundabout corridors for similar measures of effectiveness. The HCM 2010 introduces a method for evaluating the quality of service on an urban street using measures for four travel modes—automobiles, transit, pedestrians, and bicycles—based on user perceptions of quality of service. Exhibit 8 3 of the HCM 2010 lists components of traveler perception models used to generate service measures contributing to quality of service. The portion of that exhibit pertaining to urban street segments and intersections is depicted as Exhibit 1 1. The automobile traveler perception model for urban street segments is not used to determine LOS, but it is provided in the HCM 2010 as a performance measure to facilitate multimodal analyses. Other automobile related components (e.g., through delay), as well as components from other

Evaluating the Performance of Corridors with Roundabouts Chapter 1–Background Page 1-7 modes (e.g., vehicle volume and speed), do contribute to the calculation of LOS for intersections and segments. System Element Mode Model Components Urban Street Segment Automobile Stops per mile, left-turn lane presence Pedestrian Pedestrian density, sidewalk width, perceived separation between pedestrians and motor vehicles, motor vehicle volume and speed Bicycle Perceived separation between bicycles and motor vehicles, pavement quality, automobile and heavy vehicle volume and speed Transit Service frequency, perceived speed, pedestrian LOS Signalized Intersection Pedestrian Street crossing delay, pedestrian exposure to turning vehicle conflicts, crossing distance Bicycle Perceived separation between bicycles and motor vehicles, crossing distance For the automobile mode, Dowling et al. (2008) found that stops per mile was the key quality of service measure for signalized arterials based on extensive driver surveys. Stops are a significant consideration to drivers and are key inputs in evaluating energy consumption and exhaust emissions. The research also emphasized the importance of incorporating all road users. While the research did not explicitly incorporate roundabouts, the parameters for describing pedestrian and bicycle quality of service (e.g., sidewalk width, buffer separation to vehicular traffic, presence of on street parking, or expected delay at crossing points) may rate a roundabout corridor favorably over an equivalent capacity signalized arterial. Bonneson et al. also contributed to methodologies and service measures considered by the HCM 2010. In the first of two reports from NCHRP Project 03 79 (Bonneson et al. 2008a), researchers summarized findings of then current practices in real time performance measurement of urban streets. They specifically described three measurement concepts: area wide measurement, segment based measurement, and signal based measurement. Area wide measurement techniques typically use probe vehicles and some type of wireless technology. This technique is used to sample a large number of vehicles on the urban street system at a few dispersed locations; the sample is then used to estimate aggregate performance measures that describe facility performance for the previous hour or more. Segment based measurement techniques are used to measure performance on a specified street segment by monitoring traffic flow along the segment. Segment based techniques typically use one or more vehicle detectors, such as inductance loops or cameras, to monitor traffic flow on the segment. This technique estimates the performance of the monitored segment with a reasonable accuracy and with a frequency suitable for responsive signal control applications. Exhibit 1-1: Components of Traveler-Perception Models Used to Generate Service Measures (TRB 2010)

Evaluating the Performance of Corridors with Roundabouts Page 1-8 Chapter 1–Background Signal based measurement techniques measure performance on a specified street segment by monitoring traffic flow along the segment and the signal timing status of the signalized intersection that bounds the segment. Techniques following this approach typically use detectors to monitor traffic flow and receive information about the status of the phase serving the through traffic movement. According to the researchers’ findings, these techniques estimate the segment performance with a high degree of accuracy and with a frequency suitable for responsive or adaptive signal control applications. They noted travel time and travel speed were not directly measured by any of the techniques. Rather, they were estimated by combining the delay and running time measurements. This approach to travel speed estimation was intended to overcome challenges they identified in previous research that were associated with the direct measurement (or prediction) of travel speed on urban street segments. Using the findings and recommendations from efforts documented in the first report, the NCHRP Project 03 79 researchers proceeded to evaluate a selection of alternative performance prediction procedures (Bonneson et al. 2008b). The focus of their evaluation was on procedures that predicted measures (i.e., running time, delay, and stop rate) to describe the operational performance of automobile traffic flow on urban streets. One procedure was used to estimate running time and the other was used to estimate signal control delay. They found several factors affecting those two service measures, as shown in Exhibit 1 2. Ultimately, the researchers developed several procedures that were included in the HCM 2010 urban street performance evaluation methodology, with the intention of improving the accuracy of the estimated running time and control delay. Many of these procedures are also applicable to roundabout corridors, although they have not yet been applied to or calibrated for roundabout corridor evaluation. Those procedures were: delay due to turning vehicles, running time (including free flow speed), arrival flow profile, actuated phase duration, stop rate at a signalized intersection, and capacity constraints. The stop rate prediction procedure was developed to extend the range of performance measures predicted by the HCM 2010 methodology. The accuracy of the proposed procedures was evaluated by comparing the predicted performance measures with those obtained from a traffic simulation model. The findings from their analysis indicated the predicted delay from the proposed procedures was within one or two seconds of that obtained from the simulation model. A similarly good fit was found when comparing the predicted stop rate with that obtained from the simulation model. The researchers’ analysis also indicated the proposed procedures yielded a reasonably good estimate of the

Evaluating the Performance of Corridors with Roundabouts Chapter 1–Background Page 1-9 simulated travel speed. Although the urban street procedure was developed for signalized corridors, many of the factors, including running time, delay due to turning vehicles, capacity constraints, and LOS, are relevant to roundabout corridors and could be measured and applied when analyzing a roundabout corridor. Service Measure Factor Running Time Influence of segment length on free-flow speed Delay due to vehicles turning right from a through lane Delay due to vehicles turning left from a through lane Factors influencing free-flow speed (e.g., access point density, lane width, lateral clearance) Delay due to proximity of other vehicles (i.e., effect of traffic density on speed) Delay due to on-street parking maneuvers Signal-Control Delay Basic signal coordination (i.e., platoon dispersion) Green interval timing (i.e., average phase duration) Semi-actuated signal coordination (i.e., signal offset relationship) Upstream signal metering and queue spillback 1.3.2.4. Unique Features of Auto Travel on Roundabout Corridors The concept of geometric delay is an important one in comparing the total delay of roundabouts to that of signalized intersections since it is a significant difference between the two corridor types. All vehicles are expected to slow to an appropriate speed for negotiating a roundabout; therefore, they experience a delay based on the geometry of the intersection. According to Akçelik (2011), geometric delay is determined as a function of approach and exit cruise speeds as well as negotiation speeds, which depend on the geometric characteristics of the roundabout. Akçelik added that steps could be taken to approximate the value of geometric delay and add it to the control delay computed by the HCM procedure. 1.3.2.5. Mixed Signal/Roundabout Corridors Research has been conducted on corridors containing signals and roundabouts. This section summarizes the research most relevant to this project. Bared and Edara (2005) simulated the traffic impacts of roundabouts. They investigated two scenarios: 1. Urban single lane and dual lane roundabouts were modeled in VISSIM and compared with the results of RODEL and SIDRA. Their comparison with data collected from various sites in the United States showed VISSIM results were closer to field data than the RODEL and SIDRA results. Exhibit 1-2: Factors Affecting Service Measures in Estimating Travel Time on Urban Streets (Bonneson et al. 2008b)

Evaluating the Performance of Corridors with Roundabouts Page 1-10 Chapter 1–Background 2. The impact of signalized intersection proximity to roundabouts was studied using a model developed by the researchers. More specifically, they studied the impact of a coordinated signalized arterial when a roundabout is inserted within an arterial corridor. Results of average delay measures were comparable to the signalization alternative when the roundabout was operating below capacity. However, at heavy volumes, when the roundabout was operating at capacity, the performance of signalization in the model was slightly beer. The researchers did not report on comparing the model’s results with field data or how well they were correlated. Isebrands et al. (2008) examined two signalized corridor location case studies that contained roundabouts: one in Ames, Iowa, and one in Woodbury, Minnesota. For the Ames corridor, researchers coded the details into VISSIM, using existing vehicle volumes and intersection timing plans. They evaluated three alternatives: (1) optimized signal timing with the existing signalized corridor, (2) a two lane roundabout at one selected intersection, and (3) optimized signal timing with left turn lanes at the same intersection. Once the system was calibrated to replicate existing conditions, they aempted to optimize signal timings and coordinate the system for each alternative, but they were unable to achieve an optimal coordination plan due to geometry and other constraints. However, the best possible progression was sought with offsets and signal timings. The resulting timing plan with the existing geometry alternative had much higher travel time, stopped delay, and average delay than the other two alternatives, as shown in Exhibit 1 3. The signal with left turn lanes had slightly more stopped delay for both the northbound and southbound directions of travel than the roundabout alternative. However, the two alternatives had similar amounts of average delay for both directions. The signal with the left turn alternative had slightly less average delay for the northbound direction of travel, while the roundabout had slightly less delay for the southbound direction of travel. The corridor in Woodbury had three major intersections: signals for the two northernmost and a roundabout at the southern intersection. Two alternatives were evaluated for the southern intersection: a four way stop and a two lane roundabout. Both alternatives were modeled in VISSIM, and results are shown in Exhibit 1 4 for average delay, stopped delay, and travel time for passenger vehicles. Vehicles turning onto and off of the system mid corridor were not included in the analysis. Data in Exhibit 1 4 indicate lile difference in total travel time for both the northbound and southbound corridors between the two alternatives. Average delay was 10 and 17 seconds longer with the four way stop alternative for both northbound and southbound directions of travel, respectively, than for the roundabout alternative. Stopped delay was slightly longer with the four way stop alternative for both northbound and southbound directions than for the roundabout alternative.

Evaluating the Performance of Corridors with Roundabouts Chapter 1–Background Page 1-11 Exhibit 1-3: Comparison of Alternatives for the Ames, Iowa Corridor (Isebrands et al. 2008)

Evaluating the Performance of Corridors with Roundabouts Page 1-12 Chapter 1–Background A subsequent review of the data from Ames and Woodbury (Hallmark et al. 2010) using VISSIM led researchers to conclude that, based on results from the two case studies, roundabouts had minimal impact on corridor travel time. At the Ames site, signals with left-turn lanes and roundabout alternatives had similar results, considering both directions of travel together, suggesting a roundabout in this scenario did not provide a significant advantage in terms of traffic operations through the corridor as compared to the alternative where the Exhibit 1-4: Comparison of Alternatives for the Woodbury, Minnesota Corridor (Isebrands et al. 2008)

Evaluating the Performance of Corridors with Roundabouts Chapter 1–Background Page 1-13 left turn lanes were added. In Woodbury, average stopped delay was 10 and 17 seconds longer for the four way stop alternative for both directions, respectively, compared to the roundabout alternative. 1.3.2.6. International Experience in Roundabout Corridor Evaluations There is lile documented international experience of roundabout corridor evaluations similar to those considered for this project. An informal survey at a roundabout workshop held during the International Symposium on Highway Capacity and Quality of Service in Stockholm, Sweden (June 2011) revealed few cited examples of roundabout corridors in other countries, and limited experience or guidance for analysis practices. The workshop featured aendees from Germany, Sweden, Finland, Denmark, Australia, Poland, Portugal, Spain, Italy, and the United Kingdom, among others. Of all the aendees, the only work toward a roundabout corridor evaluation method was being conducted in Australia. 1.3.3. OTHER ROUNDABOUT CORRIDOR CONSIDERATIONS Emissions, non motorized transportation modes, constructability, and corridor context are additional considerations potentially providing differentiating characteristics between corridor types. The literature review explored past research in these areas and found they are generally less documented in comparison to operations and safety. 1.3.3.1. Emissions Studies examining effects on emissions generally determined isolated roundabouts performed at least as well as traffic signals for key pollutants. In addition to providing traffic operations analysis, the SIDRA software package provides emissions and fuel consumption data for roundabouts and other types of intersections. SIDRA uses a “four mode elemental model” to calculate emissions, considering time vehicles are cruising, decelerating, idling, and accelerating (Akçelik & Associates Pty Ltd 2011). Myers et al. (2005) used SIDRA to compare the performance of roundabout and existing control devices at 13 study intersections in Northern Virginia. The study showed a slight decrease in fuel consumption and emissions of four gases (carbon monoxide, hydrocarbons, nitric oxide, and carbon dioxide) during peak periods at the intersections where single lane roundabouts would be appropriate. At intersections where multi lane roundabouts would be appropriate, SIDRA predicted fuel savings of 14%, 9%, and 15% during the a.m., midday, and p.m. peak hours, respectively. The savings were in comparison to the existing control devices at the intersection. The majority of roundabout air quality research in the United States is relatively simplistic and similar to the Northern Virginia study in the sense that it merely reports the outputs of traffic analysis software such as SIDRA. As such, though the actual software may change from study to study, and the actual study sites may vary, the Northern Virginia study is representative of the types of studies that have been conducted and documented in the United States.

Evaluating the Performance of Corridors with Roundabouts Page 1-14 Chapter 1–Background Coelho et al. (2009) developed a traffic and emission decision support (TEDS) tool for urban highway corridors. They analyzed a highway corridor in Portugal containing a roundabout, a traffic signal, and a speed control traffic signal; the first two treatments are typical of those found in use in the United States, while a speed control traffic signal is not. A speed control signal is installed with speed detection devices as part of a system used to reduce speeds. In the system, individual vehicle speeds are detected upstream of the signal and if the detected speed remains below a programmed speed threshold, the signal rests in green. When the signal detects a vehicle traveling over the speed threshold, it displays a fixed clearance time, followed by a red time (of fixed or variable length) and a minimum green time, to the approaching driver. The Coelho et al. analysis suggested the roundabout intersection produced emissions similar to those of the traffic signal but more than those of the speed control signal in three of four emission types, as shown in Exhibit 1 5. The primary conclusion of the report was that the greatest percentage of vehicle emissions in the highway segment occurred at the traffic interruptions (signals and roundabout), due to the final acceleration back to cruise speed and to stop and go cycles where there were queues. The traffic interruptions were only 24 percent of the total segment distance, but together produced more than 50 percent of total emissions of the segment for all pollutants and, in the worst situation, 75 percent of overall carbon monoxide (CO) emissions. Zone CO NO HC CO2 Percentage of Total Distance Basic Highway Segments 25 41 38 49 76 Traffic Signal 32 21 25 19 8 Speed-Control Traffic Signal 26 17 15 11 8 Roundabout 17 21 22 21 8 Note: CO – Carbon monoxide HC – Hydrocarbon NO – Nitric oxide CO2 – Carbon dioxide 1.3.3.2. Non-motorized Transportation Modes NCHRP Report 616 Multimodal Levels of Service Analysis for Urban Streets documents user perspectives of a facility’s quality of service for pedestrian and bicycle modes, in addition to auto and transit modes (Dowling et al. 2008). This research was incorporated into the HCM 2010. Exhibit 1 1, presented earlier in this chapter, lists components influencing pedestrian and bicycle LOS on urban arterials with signals. It is likely the same components would influence pedestrian quality of service and, thus, pedestrian LOS on roundabout corridors. In some cases, design and operating differences between signalized corridors and roundabout corridors may generally increase or decrease pedestrian or bicycle LOS or components of LOS. For example, increasing travel speed is Exhibit 1-5: Percentage of Emissions Produced by Zone Type (Coelho et al. 2009)

Evaluating the Performance of Corridors with Roundabouts Chapter 1–Background Page 1-15 associated with decreasing pedestrian LOS. On a roundabout corridor, speeds in the vicinity of roundabouts are limited by the geometry of the roundabouts. If roundabouts are close enough together to limit speed on an entire corridor, pedestrian LOS may increase compared to an equivalent signalized corridor. At an unsignalized roundabout, pedestrian delay is generally reduced compared to a signalized intersection because a crossing can legally be made whenever a gap is present instead of waiting for a Walk indication. However, the lack of a signalized crossing may be perceived as detrimental to the quality of service for pedestrians. 1.3.3.3. Constructability In new arterial corridors there is relatively lile difference in constructing roundabout treatments compared to signalized intersections. However, in retrofit conditions a roundabout’s footprint and approach geometry treatments increase right of way needs, construction staging, and traffic maintenance during construction requirements. 1.3.3.4. Context-Sensitive Design Finally, roundabouts offer unique supporting design qualities to particular corridor contexts. Fundamentally, roundabouts offer distinct physical and visual separations between roadways approaching and continuing through an urban environment. A roundabout arterial corridor could strongly support redevelopment objectives for a community. Roundabouts offer gateway and speed reduction changes and could be especially effective on highways that become a “main street.” Landscaped medians and landscaped central islands may be particularly conducive to desired land use contexts. As an example, the La Jolla Boulevard corridor retrofit in the Bird Rock neighborhood of San Diego, California, supported a road diet and dramatically changed the corridor context from the former traditional arterial treatment. 1.3.4. LITERATURE REVIEW SUMMARY The literature review shows progress in beer understanding the operational characteristics of roundabouts in the United States, particularly isolated roundabouts. Studies of signalized corridors and corridors with mixed control also provide insight on potential service measures and analysis methods. Key findings from the literature review are as follows: The methodology for determining the safety performance of roundabouts compared to other forms of control is well established by the Highway Safety Manual (HSM). In addition, the combination of the HSM and the TRB Access Management Manual provides considerable insight on the impact of various access management techniques that can be used in a corridor of roundabouts or signals. The operational methods for evaluating corridors of roundabouts are lacking key methods. Evaluating other aspects, such as emissions, non motorized modes, constructability, and context sensitive design, contributes to a holistic

Evaluating the Performance of Corridors with Roundabouts Page 1-16 Chapter 1–Background corridor evaluation. These aspects may not be able to be readily quantified in this research project, certainly not to any level of statistical significance, but anecdotal evidence from a variety of corridors can still prove useful to practitioners. As a result, the data collection plan placed emphasis on capturing a variety of corridor contexts to gain insight on the considerations that led to the development of each corridor.