Crossing Solutions at Roundabouts and Channelized Turn Lanes for Pedestrians with Vision Disabilities: A Guidebook (2017)

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.

78 C H A P T E R 7 7.1 Assessment of Pedestrians Crossing at Roundabouts and Channelized Turn Lanes This chapter provides a method for the assessment of a pedestrian crossing at a roundabout or intersection with CTLs. The method is divided into thirteen principal steps geared at quantifying the performance of a given site. The method is based on input variables available to the analyst, including site geometry, traffic volumes, and other factors. These inputs are used to estimate operational characteristics, including vehicle speed, driver yielding, gap availability, and utiliza- tion rates of crossable yields and gaps. These operational characteristics feed into three performance checks that are integrated into the overall design processes for roundabouts and CTLs discussed in Chapter 2. These new per- formance checks for pedestrian accessibility are: (1) crossing sight distance, (2) pedestrian delay, and (3) level of risk. Many of the models and interim steps used to predict these performance measures are sensitive to the effects of crossing treatments and can be used to predict performance for new and existing sites. This chapter provides the overall methodology used for crossing assessment, while details for the various models are given in NCHRP Web-Only Document 222. 7.1.1 Crossing Performance Checks The crossing assessment method is geared at estimating three key performance checks, which jointly attempt to describe the accessibility of a site. These performance measures are (1) the crossing sight distance, (2) the estimated level of crossing delay, and (3) the expected level of risk for blind travelers. These measures are combined with other performance checks on wayfinding presented in Chapter 6 to allow for an overall accessibility evaluation of a site. The first performance check, crossing sight distance, is a design parameter used to provide clear lines of sight between the driver and the pedestrian to provide appropriate reaction and braking time. A driver with adequate time to see the pedestrian can make adequate decisions about yielding. More generally, the driver has sufficient time to react should the pedestrian step into the roadway. For sighted pedestrians, adequate sight distance is directly linked to their abil- ity to make gap acceptance decisions. But for blind pedestrians, having a clear line of sight is criti- cally linked to the amount and quality of audible information that is available to make crossing decisions. Crossing sight distance is determined from the design of the roundabout and CTL, and is a function of the approaching vehicle speed, the crossing width, and the walking speed of pedestrians. In general, faster vehicle speed, longer crossings, and slower walking speed result in an increase in the crossing sight distance requirements. Crossing Assessment

Crossing Assessment 79 The second performance check, pedestrian delay, is one commonly used by transportation analysts to evaluate the level and quality of service of pedestrian facilities for sighted pedestrians. In the context of this method, the delay is focused on the expected experience of a pedestrian who is blind. Crossing delay is a direct function of the availability of crossing opportunities in the form of crossable gaps and yields. With more crossing opportunities, delay is expected to decrease. Differences in delay between sighted and blind pedestrians may be associated with dif- ferences in the rate of utilization of the crossing opportunities. The utilization rates are in turn related to attributes of the vehicle stream, the auditory environment, and ultimately the indi- vidual making the decision. It is noted here that in many of the crossing trials performed in this and prior research, the experienced delay did not seem to be as important to blind participants as the level of risk. Accordingly, the relative weight of delay is conceptually less important than the weight of the risk score. Nonetheless, extraordinarily high delays are considered an impediment to accessibility, which is why the measure is included in this methodology. Extraordinary delays may also lead to acceptance of risky crossing opportunities. The last performance check, level of risk, is arguably the most important performance mea- sure for any crossing, as it estimates the likelihood of a poor crossing decision given attributes of the site. For the field studies that form the empirical basis of this research, risk was estimated through intervention events (participants being physically stopped from stepping into the road- way by a certified orientation and mobility specialist), through expert ratings of crossing risk, and through measurements of time-to-contact—a measure of time between a pedestrian deci- sion and the next vehicle arrival. All three metrics are surrogate safety measures, as no actual crash data are available for this analysis. However, all three metrics are documented in the lit- erature as valid measurements of pedestrian risk and have been previously used in accessibility assessment studies. Together, the three performance checks (as well as the various operational characteristics used as inputs in their calculation) are intended to provide a multifaceted look at the expected cross- ing performance of the studied crosswalk. As with any performance measure, their usefulness is limited by their ability to be measured objectively and predicted from available data. 7.1.2 Setting Performance Targets The three performance checks are intended to enable a quantitative assessment of the acces- sibility of a crosswalk at a roundabout or a CTL at an intersection. Through the quantitative nature of the performance checks, it is generally possible to (1) conduct a relative comparison of two sites or (2) conduct a before-and-after assessment of the same site. Regardless of the type of assessment, the performance targets should yield evidence as to which site or treatment results in better relative accessibility performance. It is much more challenging to use these checks to conduct an absolute assessment of acces- sibility. In other words, once a crossing assessment has been completed, and once estimates for risk, delay, and confidence score performance measures have been obtained, can a given site be classified as being “accessible”? The question of whether a performance level is acceptable is ultimately a policy decision by the appropriate agency. As an example, for general pedestrian delay, the Highway Capacity Manual 2010 (TRB, 2010) provides a letter-grade assessment of the Levels of Service (LOS) of a pedestrian crossing based on the estimated average pedestrian delay. Pedestrian delay at two-way stop- controlled intersections less than 5 seconds per pedestrian is considered LOS A, while a delay greater than 45 seconds is considered LOS F. For signalized intersections, LOS thresholds are based on a user-perception score, which incorporates delay as one of several factors. However, even with the letter-grade LOS being determined by the Highway Capacity Manual 2010 methodologies,

80 Crossing Solutions at Roundabouts and Channelized Turn Lanes for Pedestrians with Vision Disabilities: A Guidebook the decision of what LOS is acceptable is an agency decision. In other words, the performance target for pedestrian LOS is an agency policy decision. In the context of this research, the performance target for accessibility also lies with the appro- priate implementing agency or agencies. The performance checks and prediction tools presented in this document are intended to support these policy decisions through quantitative metrics, but as a research publication, this document does not set the standard. Minimum standards for accessibility, as a civil rights issue in the United States, are set by the U.S. Access Board and adopted by other agencies. 7.1.3 Limitations of the Methodology It is important to emphasize the limitations of the crossing assessment method and the per- formance checks presented in this chapter. The predictive models and performance estimation methods are based on a limited number of study sites that are believed to be representative, but nonetheless describe only a small subset of all roundabouts and CTLs that exist around the country. Further, all field studies showed high variability of performance across participants. The field-measured performance is thus only a snapshot of the true complexity of pedestrian decision-making, especially for pedestrians who are blind. The methods put forth in this chapter are intended to provide an approximation of the expected performance to aid engineers and planners in evaluating design alternatives and assist in the selection of crossing treatments to enhance the accessibility of a given or proposed site. The limitations of the methodology are primarily due to two factors: (1) variability in the geometry, signing, marking, and other features of roundabouts and CTLs chosen for the study and (2) high variability of performance across participants. Variability in the geometry of studied sites may affect the range of observed vehicle speed, con- flicting traffic volume, and local and regional differences of driver behavior. These site attributes may, in turn, affect yielding rate and gap availability, which are key inputs in the performance estimation. Variability in participant behavior and skill level may, in turn, affect yield and gap utilization rate, which are also critically linked to the performance measures. The analyst is encouraged to check for these limitations by comparing local data to the field measurements presented in this research, and details published in NCHRP Web-Only Document 222. For example, results from a region with general high driver yield compliance and frequent pedes- trian activity are likely not transferable to areas with low compliance and low expectancy of a driver encountering a pedestrian and vice versa. 7.1.4 Value of Direct Field Measurements The procedures and models presented in this chapter present a way to estimate the expected accessibility of a new intersection based on available geometric and traffic operational input variables. However, in some instances, an analyst may be interested in evaluating the accessibility of an existing site and in identifying treatments that may enhance the accessibility performance of such sites. For existing sites, direct field measurements of accessibility may represent a viable and preferred alternative to predicting performance. The clear benefit of direct field measurements is that any bias and error from applying national models to a local site are avoided. In that sense, driver behavioral difference, driving culture, and local context are uniquely tied to the site in question; this can be a big advantage. Given local context, participants may be accustomed to crossing at single-lane roundabouts due to frequent use of this intersection form in the local area. Similarly, certain treatments may be very effective

Crossing Assessment 81 in an area, where such treatments are used routinely at other intersection forms. In short, locally observed accessibility performance data is likely to be more accurate and representative of the “true” accessibility of a site in question. On the other hand, a field accessibility assessment is resource intensive, requires trained staff, and may involve the use of human subjects, which requires approval from an Institutional Review Board. As such, a full-scale accessibility audit may be out of scope for many sites in question. NCHRP Web-Only Document 222 provides detailed field protocols for conducting this accessibility audit using the methods that also form the basis of this report. As an alternative to a full accessibility audit, an agency may select a subset of studies, as permit- ted by the available resources, to calibrate for local context. For example, if a crossing indicator study with blind participants is not possible due to resource constraints or Institutional Review Board approval requirements, one or more of the input variables may be measured directly in the field. A field study of driver yielding behavior is generally very feasible and requires minimal resources. Similarly, a local study of gap availability is generally feasible. In some cases, a local gap study may even be desirable if conditions at adjacent intersections (such as an upstream signal) are expected to affect the gap availability distribution. As general guidance, direct measurements of driver and pedestrian behavior under local oper- ating conditions are expected to provide a better accessibility assessment than national-scale predictive models, provided that the local studies are executed by trained and qualified staff and follow the study protocols put forth in the final project report (or comparable). 7.2 Methodology The crossing assessment methodology consists of thirteen principal steps that are evaluated sequentially. The methodology obtains key input and performance targets from the overall site design process described in Chapter 2. A key characteristic of the method is that it is iterative. Should a performance check fail to meet a specified performance target, it may require changes to the design and recalculation of the performance checks as described in Chapter 2. The meth- odology flow chart is shown in Figure 7-1 and discussed in detail in the following sections. To use the crossing assessment methodology, initial site-related data need to be gathered. The data are entered into various models developed as part of crossing assessment and eventually the model results are used for final crossing assessment performance measures. A summary of the required input data and their application in each of the crossing models is shown in Table 7-1. 7.3 Methodological Steps In this section, each of the steps shown in Figure 7-1 is described in more detail. For steps with significant computations, only the key equations are shown here, with additional information on model derivation provided in the NCHRP Web-Only Document 222. The methodology is applied to each approach of the roundabout, and separately to entry and exit legs, as well as to CTLs. Before embarking on the steps, the analyst needs to obtain geometry inputs and performance targets. In the overall design process described in Chapter 2, the analyst defines the candidate design and crossing configuration of the roundabout or CTL to be evaluated for accessibility. The initial design should contain sufficient detail to specify the number of lanes, design radii, cross- walk location, and other geometric details. The initial design may be obtained from an engineer- ing design project at approximately the 10% to 25% completion level. At this stage, the design is expected to provide sufficient geometric and operational details, while still allowing flexibility

82 Crossing Solutions at Roundabouts and Channelized Turn Lanes for Pedestrians with Vision Disabilities: A Guidebook Figure 7-1. Methodology flow chart. Figure 7-1 showing a thirteen-step methodology for assessing crossing performance. Step Zero obtains design data from the overall process described in Chapter 2, as well as performance targets set by the agency. The twelve principal steps of the methodology are as follows: (0) obtain geometry inputs, (1) gather site data and other inputs, (2) predict vehicle speed at crosswalk, (3) calculate crossing sight distance, (4) check sight distance provi- sion, (5) predict crossing opportunities in the form of gaps and yields, (6) estimate utilization of gaps and yields, (7) evaluate audible environment and noise effects, (8) estimate pedestrian delay, (9) check pedestrian delay, (10) estimate crossing risk, (11) check crossing risk, (12) check visibility of traffic control devices, and (13) com- plete crosswalk assessment. The analysis sequence is linear, with potential for iteration after each of the three performance checks in steps 4, 9, and 11.

Crossing Assessment 83 for design adjustments and treatment provision as needed. Key design elements needed for the crossing assessment include the number of lanes, lane widths, crosswalk location, treatment details, and design radii for the intersection itself. The initial design should also be sensitive to other performance elements that are specified in various guidelines or standards (e.g., design vehicle). The initial design may or may not include specialized treatments intended to enhance the accessibility of the site. Before starting with the principal procedure, the analyst reviews and notes performance targets for the three accessibility performance checks based on agency guidelines or standards. Pedestrian accessibility performance objectives based on federal guidelines and previously conducted studies can serve as target values, but the specification of target standards is the responsibility of the agency conducting the assessment. This report intends to provide the quantitative assessment methodology to estimate the performance measures needed in those standards. 7.3.1 Step 1. Gather Site Data and Other Inputs The analyst gathers engineering inputs or selects default conditions specified by the method- ology. These inputs include traffic conditions and roadway factors, as well as geometric details of the roundabout or CTL in question. The overall design of the roundabout or CTL in question was transferred to the crossing assessment in step 0. In this step, design details necessary for the crossing assessment are extracted, along with other traffic operational factors. See Table 7-1 for a listing of required input data. Table 7-1. Required inputs for the crossing assessment method. Step Equation/Table Required User Input Step 2. Predict speed at crosswalk Equation 7-1 Table 7-2 and Table 7-3 Fastest path radius Treatment effect Step 3. Calculate crossing sight distance Equation 7-2 Equation 7-3 Vehicle speed at crosswalk (from Step 2) Approach geometry Pedestrian walking speed Step 4. Check Sight Distance Provisions Expert judgment CAD drawing Crossing sight distance (from Step 3) Step 5. Calculate crossing opportunity (gaps and yields) Equation 7-4 Equation 7-5 Equation 7-6 Equation 7-7 Approach geometry and treatment, Gap acceptance parameters Pedestrian walking speed Traffic volume on approach Vehicle speed at crosswalk (from Step 2) Step 6. Estimate utilization of gaps and yields Table 7-4 Table 7-5 Approach geometry Step 7. Evaluate audible environment and noise effect Expert judgment Appendix A Local observation Surrounding lane uses Steps 8 and 9. Estimate pedestrian delay Equation 7-9 through Equation 7-11 Gap and yield opportunities (from Step 5) Gap and yield utilization (from Step 6) Steps 10 and 11. Estimate crossing risk Equation 7-12 Vehicle speed at crosswalk (from Step 2) Noise (from Step 7) Sight distance (from Step 4)

84 Crossing Solutions at Roundabouts and Channelized Turn Lanes for Pedestrians with Vision Disabilities: A Guidebook 7.3.2 Step 2. Predict Vehicle Speed at Crosswalk Vehicular speed has been identified as a key measure affecting pedestrian accessibility. This step predicts the free-flow speeds under low volume conditions that can be expected in the vicinity of the crosswalk. The analyst may obtain speed estimates through field measurements at comparable sites, or use the speed prediction equations presented below. Speed prediction is required for computing other aspects of accessibility such as calculating the required crossing sight distance and driver yielding rate at the crosswalk, and predicting the rate of intervention and risk events. The model for predicting the speed of the crosswalk is the theoretical fastest path speed method described in NCHRP Report 672 for roundabouts. It was found to also apply to vehicle free-flow speeds through CTLs. The vehicle speed model in Equation 7-1 estimates the free-flow speed at the crosswalk, FFS, as a function of the fastest path radius, R, for a curve with positive superelevation e = +0.02 (drainage toward the outside, which is most common). Equation 7-1. Fastest path radius calculation for vehicle speed. FFS 3.4415 R , for e 0.020.3861= = + The equation predicts the 85th percentile free-flow speed expected at the crosswalk as a func- tion of fastest path radius (in feet) that is believed to control the speed at the crosswalk. For roundabout entries, this speed is generally calculated using the entry path radius, R1. For round- about exits, a composite equivalent radius may be used to estimate the speed under consider- ation of both the radii in the circle and on the exit itself. At a roundabout entry, this speed is principally a function of the R1 radius shown in Figure 7-2. For exiting vehicles, the analyst can estimate an equivalent composite radius from the terms R2, R4, and R5 depending on whether the conflicting movement is a right-turning vehicle from the immediate upstream entry, or a through, or left-turning vehicle from another entry. Since vehicles Figure 7-2. Roundabout vehicle path radii (Source: NCHRP Report 672).

Crossing Assessment 85 have an opportunity to accelerate leaving the roundabout, their actual speeds at the crosswalk are expected to be higher than those predicted by the respective controlling radii. As such, the speed is estimated at the fastest path radii, adjusted by the acceleration of vehicles as described in NCHRP Report 672. For CTLs, the equivalent of the R1 radius is used to estimate the speed. The equivalent radius computations are summarized in Table 7-2. The free-flow speed at the crosswalk can also be impacted by certain treatments that are installed specifically with the goal of reducing vehicle speeds. Several sites were evaluated in prior research with various forms of raised crosswalks or speed tables installed to slow traffic, and some sites even had speed humps in advance of the crosswalk with a similar goal. The Traffic Control Devices Handbook provides some guidance for estimating the speed-reducing effects of traffic-calming measure as shown in Table 7-3 (Seyfried, 2013). Specific design attributes of the traffic-calming measure (e.g., height of the speed hump or speed table, transition slope) are not reflected in the Traffic Control Devices Handbook guidance. Further, the Traffic Control Devices Handbook data refer to standard intersections, and do not consider the speed-reducing impacts of roundabout or CTL geometry. As such, it is advisable to use the average reduction or percentage reduction in speed as an approximation of the effect, rather than the absolute measured speed. 7.3.3 Step 3. Calculate Crossing Sight Distance The crossing sight distance corresponds to the distance required by pedestrians to recognize the presence of conflicting vehicular traffic and determine crossing opportunities at intersections and roundabouts. The distance is established through sight triangles that allow a pedestrian to evaluate potential conflicts with approaching vehicles. Similarly, the resulting sight triangles also assure that the driver has a clear view of a pedestrian waiting to cross or approaching the crosswalk. For pedestrians who are blind, the crossing sight distance applies in that any visual Table 7-2. Equivalent composite radius for speed estimation. Approach Vehicle Movement Equivalent Composite Radius RBT Entry Right, through, and left R1 RBT Exit Right R5 with acceleration constraint RBT Exit Through R2 with acceleration constraint RBT Exit Left R4 with acceleration constraint CTL Right R1 equivalent at CTL Table 7-3. Speed impacts due to traffic-calming measures (adapted from the Traffic Control Devices Handbook). Traffic-Calming Measure Sample Size 85th Percentile Speed after Calming in mi/h (Std. Dev.) Average Change in Speed after Calming in mi/h (Std. Dev.) Average Percentage Change (Std. Dev.) 12 ft hump 179 27.4 (4.0) -7.6 (3.5) -22% (9%) 14 ft hump 15 25.6 (2.1) -7.7 (2.1) -23% (6%) 22 ft table 58 30.1 (2.7) -6.6 (3.2) -18% (8%) Longer tables 10 31.6 (2.8) -3.2 (2.4) -9% (7%)

86 Crossing Solutions at Roundabouts and Channelized Turn Lanes for Pedestrians with Vision Disabilities: A Guidebook obstructions are also expected to impact the audible environment at the crosswalk and the ability to hear approaching vehicles without sound obstructions or deflections. The methodology developed to determine crossing sight distance adequacy at a roundabout or CTL has been adapted from the sight distance performance checks for vehicles at roundabouts from NCHRP Report 672, calculations and definitions from the Green Book, and the pedestrian mode methodology in Chapter 19 of the Highway Capacity Manual 2010. The four pedestrian movements at a roundabout—crossing from the curb to the splitter island at entry, crossing from the splitter island to the curb at entry, crossing from the curb to the splitter island at exit, crossing from the splitter island to the curb at exit—are all different for several reasons, including: • Traffic is approaching from the left when crossing from the curb, but from the right when crossing from the splitter island; • Traffic is moving only in front of the pedestrian when crossing from the curb (quiet behind the pedestrian), while it is moving both in front of and behind the pedestrian when crossing from the splitter island; and • Entering traffic is decelerating in approach of the yield line, while exiting traffic is accelerating as drivers exit the roundabout. Since traffic patterns at each conflicting approach are judged independently, there are sight distances and sight triangles associated with each location and its conflicting approaches. The entry crossing locations have one potential conflict with vehicles entering the roundabout. The exit crossing locations are subject to two conflicting movements with different trajectories: traffic from the immediate upstream entry approach (right turns), and traffic circulating from other upstream approaches (through and left turn movements). The sight distance (d) is calculated as a function of the conflicting vehicle speed (V) and the pedestrian critical headway (tc) Equation 7-2. Crossing sight distance calculation. ( ) ( )( )= 1.467 ,d V tn n n c where, dn = distance along approach leg n upstream of the crosswalk for crossing, ft; Vn = free-flow speed of conflicting vehicle movement on approach n, mph; and tn,c = critical headway required by a pedestrian crossing approach n. The critical headway describes the minimum amount of time necessary for a pedestrian to cross the roadway. The critical headway calculation is directly derived from the pedestrian analysis method covered in the two-way stop-controlled intersection methodology of the Highway Capacity Manual 2010. Equation 7-3. Estimating pedestrian critical headway. ,t L S tn c n p s)(= + where, Ln = crosswalk length for a specific traffic stream, ft; Sp = average pedestrian walking speed, ft/s (default = 3.5 ft/s); ts = pedestrian start-up time and end clearance time, s (default = 2 s). The vehicle speed parameter is the same as was estimated in Step 2. At a roundabout entry, this speed is principally a function of the R1 radius shown in Figure 7-2. For exiting vehicle, the analyst

Crossing Assessment 87 uses the controlling radius for the particular movement from radii R2, R4, and R5 depending on whether the conflicting movement is a right-turning, through, or left-turning vehicle. For all exit-leg movements, the actual speed is adjusted to account for the vehicle’s ability to accelerate before reaching the crosswalk as shown in Table 7-2. Once the minimum distance, d, is determined for all possible conflicting movements, the designer should plot the distance along the centerline of the direction of travel. Figure 7-3 shows the necessary sight distance, d, for each crossing location at the entry and exit of a roundabout. The length of each d may be longer or shorter than shown relative to the roundabout geometry, depending on the speed and critical headway times used in the calculation. After plotting the distance from the pedestrian location, the sight triangle is determined as shown in Figure 7-4. Any sight obstruction should be eliminated from the sight triangles for bet- ter pedestrian visibility. The figure focuses on showing examples for just two of the crosswalks. But just like the rest of the crossing assessment method, the evaluation needs to be performed for each crosswalk, entry and exit, and both for crossings originating from the island and originating from the curb. 7.3.4 Step 4. Check Sight Distance Provisions In this step, the calculated required crossing sight distance is checked against the design of the roundabout or CTL to see if sufficient sight distance is provided. The required length of sight distance is measured along the center of the approaching roadway in advance of the crosswalk. Figure 7-5 illustrates this for a roundabout for both entry and exit legs. The figure includes a two- lane entry (south entry, shown in blue), a two-lane exit (north exit, shown in red), and a three- lane entry and exit (east entry and exit, shown in green). Sight distances are shown based on the field-measured vehicle speed at the crosswalk, which was approximately 13 mi/h to 15 mi/h because of the raised crosswalks installed on the tested approaches. Without this treatment, the sight distance requirements would have been significantly longer. The figure further shows Figure 7-3. Minimum sight distance along the actual vehicle path for roundabouts. Figure 7-3 shows a schematic of a roundabout with calculated sight distances drawn for entry and exit legs, and for both crossings from the curb and crossings from the splitter island.

Figure 7-4. Pedestrian sight triangles for each crossing location. Figure 7-4 shows a schematic of a roundabout with estimated sight triangles drawn based on the calculated sight distances. Sight triangles are drawn for entry and exit legs, and for both crossings from the curb and crossings from the splitter island. Figure 7-5. Sight distance for two-lane and three-lane roundabout approaches. Figure 7-5 shows an example application of the sight distance calculations for a two-lane entry (south entry, shown in blue), a two-lane exit (north exit, shown in red), and a three-lane entry and exit (east entry and exit, shown in green). Sight distances are shown as arrows based on the field-measured vehicle speed at the crosswalk, which was approximately 13 mi/h to 15 mi/h because of the raised crosswalks installed on the tested approaches. The figure further shows the resulting sight “triangles” drawn relative from the respective waiting positions (on both curb and island) for a pedestrian to the end of the measured sight distance.

Crossing Assessment 89 the resulting sight “triangles” drawn relative from the respective waiting positions (on both curb and island) for a pedestrian to the end of the measured sight distance. It is evident from this example that the three-lane crossings result in a longer required sight distance (336 ft for traffic exiting from circle, 236 ft for traffic exiting from south to east right turn, and 213 ft for traffic entering the circle) relative to the two-lane crossings (235 ft, 164 ft, and 153 ft for the corresponding distances). This is intuitive, as the required crossing time for pedestrians (exposure time in the street) is longer for a three-lane crossing, thereby increasing the sight distance requirements. The sight triangles between the pedestrian crosswalk landing and the end of the measured sight distance should be clear of obstacles and obstruction, including tall bushes, signal control- ler cabinets, walls, or buildings. If the crossing sight distance is not provided, pedestrians will not be able to see (and presumably hear) far enough to be able to accept a sufficiently large gap in traffic. Similarly, drivers may not be able to see a pedestrian waiting to cross or beginning to cross, which is expected to impact their propensity to yield as well as their ability to react in time to avoid a potential collision. Increased vehicle speeds, longer crossing distances, and slower pedestrian walking speeds all contribute to longer sight distance requirements. If the sight distance check fails, the designer has the choice of modifying the design to reduce the sight distance requirements (e.g., through tighter radii, fewer lanes, or a raised crosswalk to reduce speeds) or may decide to move the crosswalk (e.g., further from the circulating lane for an exit crossing). Figure 7-5 illustrates the effect of crossing distances for a roundabout with two-lane and three- lane crossings. Figure 7-6 shows two CTL approaches to a signalized intersection. The east approach has a required crossing sight distance of 203 ft for a single-lane crossing. For the north approach, the presence of a raised crosswalk reduces vehicle speeds and thereby the sight distance to 129 ft. 7.3.5 Step 5. Predict Crossing Opportunities (Gaps and Yields) This step predicts the availability of crossing opportunities in the form of crossable gaps between moving vehicles, as well as vehicle yields. The availability of crossable gaps can be estimated from traffic flow relationships by taking into account platooning or bunching effects that may result from signals upstream of the crosswalk in question. A predictive equation for gap opportunities is presented below. Pedestrian crossable gap opportunities P(CG-Opp) are predicted as shown in Equation 7-4 as a function of critical headway for crossable gap (tc) and average headway (tavg). The equation shows the equation that can be used to estimate the probability of encountering a gap greater than the critical gap. Equation 7-4. Estimating Pedestrian Crossable Gap Opportunities from Traffic Flow Theory (May, 1990). ≥( ) ( )− = = −tP CG Opp P headway ec t t c avg where, tc = critical headway for crossable gap (s) tavg = average headway, defined as (3,600 s/h)/(vehicular volume in vehicles/hour) In the absence of pedestrian platoons, the critical headway for pedestrians can be calculated by Equation 7-5 following the pedestrian delay methodology at two-way stop-controlled intersec- tions in the Highway Capacity Manual 2010.

90 Crossing Solutions at Roundabouts and Channelized Turn Lanes for Pedestrians with Vision Disabilities: A Guidebook Equation 7-5. Pedestrian critical headway from the Highway Capacity Manual, Chapter 19. t L S tc p s= + where, L = crosswalk length (ft) Sp = average pedestrian walking speed (ft/s), and ts = pedestrian start-up and clearance time (s), default = 2 sec. In addition to crossable gaps, driver yielding events also present crossing opportunities. A predictive equation for estimating the likelihood of driver yielding is given below, as a function of geometry and other prevailing traffic conditions. Yield opportunities are predicted, as shown in Equation 7-6, as a function of fastest path radius at the crosswalk (Rcrosswalk) and the presence of an RRFB at the approach. The fastest path radius (in feet) is a continuous variable and RRFB is a binary variable that is 1 if a round- about approach is equipped with RRFB and 0 if no RRFB is present. Figure 7-6. Sight distance for CTLs with and without raised crosswalks. Figure 7-6 shows an example application of the sight distance calculations at a CTL. Sight distances are shown as arrows and the resulting sight “triangles” drawn relative from the respective waiting positions (on both curb and island) for a pedestrian to the end of the measured sight distance.

Crossing Assessment 91 Equation 7-6. Estimating probability of yielding. 0.065 * 11.9 * 82.6( ) ( ) ( ) ( ) ( )= − + +P Yield Rcrosswalk RRFB The model predicts a base yield probability of 82.6%, which is reduced by 6.5% for each 100 ft increase in the fastest path radius. The presence of an RRFB increases the yield probability by 11.9% after controlling for radius. The model has been calibrated from data at two-lane round- abouts. It is expected that yield rates at single-lane roundabouts are higher than the estimate from Equation 7-6, while yield rates at three-lane roundabouts are lower. The probability of yield crossing opportunity P(Y-Opp) is different than the probability of driver yielding, P(Yield). The term P(Y-Opp) is calculated on the basis of all encountered vehicles, and it is a better representation of the yield encountered rate that a pedestrian is likely to experience. A reasonable approach for estimating P(Y-Opp) from P(Yield) is to subtract the probability of crossable gaps from the total number of vehicle events (see Equation 7-7): Equation 7-7. Estimating yield opportunities from yield probabilities. ( ) ( )( )( )= −- 100% -P Y Opp P Yield P CG Opp This approach assures that the sum of P(Y-Opp) and P(CG-Opp) is less than or equal to 1 as is required by definition. 7.3.6 Step 6. Estimate Utilization of Gaps and Yields In this step the analyst estimates the rate of utilization of gap and yield crossing opportunities. The utilization rate of gaps is calculated as the ratio of the number of crossings a blind pedes- trian is expected to take in a gap over the total estimated number of gap crossing opportunities available. Yield utilization is similarly calculated as the ratio of the number of yields utilized or accepted over the total number of yields available. The gap utilization rate of pedestrians who are blind is generally more conservative than that of sighted pedestrians, with the biggest differences being additional latency time after a vehicle passes the crosswalk until a decision to cross is made. Sighted pedestrians will often visually identify a gap in traffic approaching the crosswalk and initiate crossing as soon as the gap opens in front of them. Research has generally shown that a blind pedestrian requires additional time for the noise of the vehicle to subside before choosing to cross in a gap. The additional decision latency time results in blind travelers rejecting gaps that a sighted person may have utilized. Gap opportunity utilization is estimated from the average gap opportunity utilizations observed at study locations in NCHRP Project 03-78B, and are shown in Table 7-4. There is pres- ently insufficient data in the literature to derive more sophisticated gap utilization models, but Approach Average Gap Utilization Sample Size Std. Error 1 Lane Entry 66.5% 6 2.55% 1 Lane Exit 60.8% 6 2.92% 2 Lane Entry 82.3% 12 2.21% 2 Lane Exit 65.7% 11 3.00% CTL 57.9% 12 2.05% Table 7-4. Estimated average gap utilization for blind pedestrians.

92 Crossing Solutions at Roundabouts and Channelized Turn Lanes for Pedestrians with Vision Disabilities: A Guidebook analysts are encouraged to use local data or estimates should those be available. It is also noted that the relatively high gap utilization of 82.3% at two-lane entries compared to other locations may be biased by the specific sites studied in the research. Similar to the concept of gap utilization, not all yield events may result in a utilized crossing. Pedestrians who are blind may not utilize a yield crossing opportunity because of high ambi- ent noise, quiet vehicles, uncertainty of driver intent, or other reasons that result in not having confidence in their judgment. A non-utilized yield is not necessarily an event “missed” by the pedestrian, as the decision to reject the yield may be made consciously. Yield opportunity utilization is estimated from the average yield opportunity utilizations observed at study locations and is shown in Table 7-5. There is presently insufficient data in the literature to derive more sophisticated yield utilization models, but analysts are encouraged to use local data or estimates should those be available. 7.3.7 Step 7. Evaluate Audible Environment and Noise Effects Research has linked the accessibility of a site for a pedestrian who is blind to the availability of adequate audible cues. This is intuitive, as a blind traveler relies on hearing to navigate and make crossing decisions. An adequate audible environment is therefore critical to assure that a blind traveler can independently and safely navigate a crossing. In this step, the analyst should identify and flag any concerns about the audible environment. The outcome is a yes/no check on whether audibility is likely to be compromised at the site. To date, no quantitative method exists to accomplish this, but some guidance is provided below, as well as in Appendix A. The availability of audible cues is related to the presence of noise sources in the vicinity of the site, as well as obstacles that may interfere with the ability to clearly hear approaching vehicles. Such obstacles may include signs, poles, or landscaping that may impact audibility in a mat- ter similar to their impact on sight distance. The principal question is whether the person can adequately hear the approaching vehicle (referred to as the signal in human factors research) to the background noise. Having an adequate signal-to-noise ratio is critical to assure that the conflicting vehicle can be heard and distinguished from other noise sources. In evaluating the audible environment, the first and foremost audibility consideration is the location of the crosswalk relative to sources of noise. In the case of a CTL, most of the traffic noise is generated at the main intersection. It is generally expected that smaller radius CTLs result in smaller channelization islands, which, in turn, place the pedestrian closer to that noise source. In a similar fashion, crossing from the channelization island to the curb is expected to have higher levels of interfering noise (from behind the pedestrian) than crossings from the curb to the island. For roundabouts, the separation between the crosswalk and the circulatory roadway affects the level of noise at the crosswalk. Noise levels are further expected to be different between entry legs (quiet traffic slowing down in approach of the roundabout) and exit legs (louder traffic Approach Average Gap Utilization Sample Size Std. Error 1 Lane Entry 67.0% 6 2.79% 1 Lane Exit 68.5% 6 3.30% 2 Lane Entry 72.7% 17 22.09% 2 Lane Exit 70.5% 16 1.22% CTL 35.7% 12 1.24% Table 7-5. Estimated average yield utilization for blind pedestrians.

Crossing Assessment 93 accelerating away from the roundabout). Similar to CTLs, the splitter island is expected to have the highest levels of noise, with traffic traversing in front of and behind the waiting pedes- trian. Landscaping has the potential to minimize the noise behind the waiting pedestrian when installed on the splitter island, but may limit lines of sight from the driver to the pedestrian. Other noise sources may exist in the vicinity of the site that have high impact on the blind person’s ability to hear conflicting traffic and distinguish it from background noise. Common examples of this include nearby freeways (especially at interchanges), work zones or construc- tion activity, or general industrial activity. Noise levels are also often amplified in locations with a high percentage of trucks and other heavy vehicles. 7.3.8 Step 8. Estimate Pedestrian Delay The second accessibility performance check is pedestrian delay. NCHRP Report 674 showed a link between pedestrian delay and the probability of crossing at a crosswalk. The probability of crossing at a crosswalk, P(Cross), is described in Equation 7-8 as a function of the probability of yielding, P(Y), the probability of yield utilization, P(GO | Y), the probability of encountering a crossable gap, P(G), and the probability of utilizing that crossable gab, P(GO | G). Equation 7-8. Estimating the probability of crossing. ( ) ( ) ( ) ( )( ) = +_ * _ _ * _P Cross P Y Opp P GO Y Opp P CG Opp P GO CG Opp The components of P(Cross) were all estimated in previous steps. This research developed models to predict pedestrian delay at roundabouts and intersections with CTLs as a function of P(Cross). These models allow analysts to estimate pedestrian delay for new sites if the input vari- ables are known. Since the models are sensitive to the utilization measures, the delay estimation can distinguish between blind and sighted pedestrians, who may be presented with the same gap and yield opportunities, but have different rates of utilizing these opportunities. Separate models were developed for single-lane CTL approaches, single-lane roundabout approaches, and two-lane roundabout approaches. Pedestrian delay for single-lane CTL approaches is predicted as shown in Equation 7-9 as a function of P(Cross). Equation 7-9. Calculating pedestrian delay for single-lane CTL approaches. ( )( )= −10.75 9.95 *d LN P Crossp Pedestrian delay for single-lane roundabouts is predicted, as shown in Equation 7-10, as a function of P(Cross). Equation 7-10. Calculating pedestrian delay for single-lane RBT approaches. ( )( )= −9.37 9.78 *d LN P Crossp Pedestrian delay for two-lane approaches (two-lane roundabouts) is predicted, as shown in Equation 7-11, as a function of P(Cross). Equation 7-11. Calculating pedestrian delay for two-lane RBT approaches. ( )( )= −6.14 8.53 *d LN P Crossp The delay term, dp, in Equation 7-9 through Equation 7-11, is measured in seconds per pedes- trian. The equations are applied separately to each portion of the crossing, which in the case of a roundabout means the total delay or the sum of delay for the entry and exit legs.

94 Crossing Solutions at Roundabouts and Channelized Turn Lanes for Pedestrians with Vision Disabilities: A Guidebook The quantity increases with a decreasing probability of crossing, P(Cross), which in turn decreases with reduced availability and utilization of gaps and yields. As such, a low-volume site (i.e., with lots of gaps) or a high-yielding site is expected to result in low delay, provided that the utilization of crossing opportunities is adequate. As traffic volumes increase (reducing the availability of gaps) and as vehicle speeds increase (reducing the number of yields), the delay per pedestrian is expected to increase. As an alternative to this pedestrian delay methodology, the analyst may choose to refer to the method in the Highway Capacity Manual 2010, or conduct a simulation study. However, it is emphasized here that the Highway Capacity Manual 2010 method does not account for opportunity utilization of less than 100%. For simulation, a method for considering varying gap and yield availability and utilization distributions is described in Schroeder, Rouphail, and Hughes (2008). 7.3.9 Step 9. Check Pedestrian Delay The calculated pedestrian delay has to be compared to the agency performance target to deter- mine whether it is acceptable. The Highway Capacity Manual 2010 defines pedestrian LOS for unsignalized intersections on the basis of the average delay per pedestrian, although these per- formance thresholds are not calibrated for blind travelers. Table 7-6 shows the Highway Capacity Manual 2010 thresholds for delay. The LOS in Table 7-6 is defined on a per approach basis. In the case of a roundabout, this means that the entry and exit leg delays should be added together before applying the thresh- olds. For a CTL, the total crossing delay should be considered, which adds whatever delay the pedestrian experiences crossing one or more of the intersecting streets to the calculated CTL delay. The analyst may use the Highway Capacity Manual 2010 methodology for signalized intersections to estimate the pedestrian delay of the full crossing. In Table 7-6, it is further shown that the likelihood of risk-taking increases significantly with longer wait times. While this refers primarily to sighted pedestrians (no studies with blind travel- ers have been conducted to date), high delay times are nonetheless cause for concern and should be avoided. The agency may thus choose to adopt stricter performance thresholds than those shown in the table. 7.3.10 Step 10. Estimate Crossing Risk The third, and arguably most critical, accessibility performance check is the expected level of pedestrian risk. The level of risk is determined in field studies from certified orientation and mobility specialist intervention events, observer ratings, time-to-contact measurements, and Table 7-6. Pedestrian LOS thresholds for unsignalized intersections from the Highway Capacity Manual. LOS Control Delay (s/ped) Comments A 0-5 Usually no conflicting traffic B 5-10 Occasionally some delay due to conflicting traffic C 10-20 Delay noticeable to pedestrians, but not inconveniencing D 20-30 Delay noticeable and irritating, increased likelihood of risk-taking E 30-45 Delay approaches tolerance level, risk-taking behavior likely F >45 Delay exceeds tolerance level, high likelihood of pedestrian risk-taking

Crossing Assessment 95 video observations. These risk assessment factors are correlated to the characteristics of the studied crosswalk to arrive at a risk prediction model. The model predicts the likelihood of a risky decision as a function of different variables. The intervention model predicts the likelihood of the crossing decisions a blind pedestrian might make, which would have resulted in a certified orientation and mobility specialist inter- vention. The intervention model, P(INT) is predicted as shown in Equation 7-12 as a function of noise (NOISE), average crosswalk speed (XSPD_AVE), and sight distance (SIGHT_D). Vari- ables NOISE and SIGHT_D are binary variables and equal to 1 if the noise level is high and the required crossing sight distance is not provided, respectively. Noise level and sight distance were estimated in Steps 4 and 7, respectively. XSPD_AVE is a continuous variable and is defined for speeds higher than 10 mph. Equation 7-12. Estimating the probability of interventions. ( ) ( ) ( )( ) = + + −0.0629 * 0.0020 * _ 0.0230 * _ 0.0177P INT NOISE XSPD AVE SIGHT D 7.3.11 Step 11. Check Crossing Risk The calculated crossing risk has to be compared to the agency performance target to deter- mine whether it is acceptable. There is presently no standardized guidance for what level of risk or what rate of interventions is acceptable. Clearly, an intervention rate of zero would be desirable to reduce the risk as much as possible. In the language of the ADA legislation, however, a crossing should provide equivalent access to persons with and without a disability. To date, no comprehensive study exists comparing the rate of interventions between blind and sighted pedestrians, therefore guidance is limited. Based on research conducted for FHWA at two-lane roundabouts (Schroeder et al., 2015), researchers concluded that an intervention rate of 3% or less is similar to the rate of interven- tions at single-lane roundabouts, and may be considered accessible in many cases. Rates of inter- vention above 5% were considered as likely to present a significant barrier for blind travelers crossing at these locations, and rates of intervention above 10% were considered as representing a challenging and risky crossing environment. It is emphasized here that these thresholds are not based on any formal guidance available, nor should they be used as the basis for policy and categorization of roundabouts. These thresholds are merely introduced to help distinguish and categorize sites for the purpose of analysis and discussion. An agency should set its own thresholds for the purpose of evaluating sites and decid- ing on the need for further treatments. 7.3.12 Step 12. Visibility of Traffic Control Devices The accessibility framework and method presented in this chapter may result in the provision of treatments intended to enhance accessibility of pedestrians who are blind at roundabouts and CTLs. These treatments encompass a range of geometric and design changes in the roundabouts, as well as the installation of traffic control devices in the form of traffic signals, beacons, signs, and markings. Traffic control devices on roads open to public travel have important functions in providing guidance and information to road users. The visibility of such physical aids is espe- cially important for motorists, bicyclists, and pedestrians navigating complex roundabouts and intersections with CTLs. The basic question in this context of visibility is whether traffic control devices can be seen by drivers as they approach the crosswalk and whether pedestrians can see or hear the device.

96 Crossing Solutions at Roundabouts and Channelized Turn Lanes for Pedestrians with Vision Disabilities: A Guidebook An underlying consideration of whether traffic control devices are understood by drivers and pedestrians also plays into the question of visibility. The key difference between visibility and sight distance (discussed in the another step of the crossing assessment method) is that visibility considers weather drivers and pedestrians can see (and properly interpret) traffic control devices, while crossing sight distance is strictly tied to physical obstructions and the line of sight between drivers and the pedestrian. The principles underlying the visibility performance checks presented in this section are com- piled from MUTCD, the Traffic Control Devices Handbook, NCHRP Report 672, and other sources. 7.3.12.1 Visibility Considerations for Signs and Markings Traffic signs and pavement markings are designed and placed in a way that they are legible to the road user for whom it is intended. Proper visibility of these traffic control devices assures that they are understandable in time to provide information for a proper decision. This decision can be for the purpose of navigation, warning, guidance, or advisory purposes. Important aspects include, but are not limited to, consistent design, daytime and nighttime visibility, proper size, and correct placement. Two key considerations exist for signage and markings, both of which test for adequate sepa- ration of traffic control devices at the crosswalk with the traffic control devices controlling the downstream merge point at the CTL or the entry at a roundabout. 1. The first consideration is whether there is sufficient separation between the crosswalk mark- ings and the markings for the yield line or stop bar downstream of the crosswalk at the roundabout entry or at the CTL merge point. The two sets of markings should be separated by at least one-vehicle length. This assures a visual separation and distinction of the two sets of markings. It also provides a one-vehicle length of storage between the yield line or stop line and the crosswalk, so that a waiting vehicle does not obstruct the crosswalk. Any subsequent vehicles can then queue upstream of the crosswalk, leaving the crossing area free (in principle). If a longer separation is needed, a separation in multiples of vehicle lengths (i.e., 20 ft, 40 ft, 60 ft) is desirable to maximize the potential for vehicles blocking the crosswalk. 2. The second consideration is whether there is appropriate separation of signs at the crosswalk from signs at the yield or stop line. In addition to checking for separation, the designer should also check for potential occlusion effects with a sign blocking one or more downstream signs. Visual obstruction may also affect the visibility of the pedestrian, but that aspect should have been identified in the crossing sight distance step. 7.3.12.2 Visibility Considerations for Signals and Beacons Six considerations exist for signal and beacon installations at roundabouts and CTLs, as follows: 1. Are signals visible to an approaching driver to provide adequate stopping sight distance per MUTCD requirements? Stopping sight distance is calculated from the approaching vehicle speed and assumed driver reaction times and deceleration rates. If stopping sight distance is not adequate, a supplemental (upstream) signal head may be needed. This visibility concern is especially important at roundabout exit-leg signals and CTLs, where the roadway curvature upstream of the signal may limit its visibility. 2. Are mounting heights correct? Overhead traffic signals need to be mounted at a sufficient height to allow large design vehicles (trucks) to pass underneath them. The general mounting height of overhead mounted signals is 15 ft. In addition, side-mounted signals need to be mounted at least 8 ft high to assure proper visibility, and to not act as a potential obstacle for pedestrian traffic. 3. Is the stop bar set back enough? MUTCD requires a separation between the vehicle stop bar and any overhead signal to assure that drivers stopped at the stop bar can comfortably see

Crossing Assessment 97 the signal display (without having to lean forward in their seat). This setback requirement may result in the need for a full or partial crosswalk relocation at roundabouts to meet this criterion at the exit leg. 4. Is the stop bar located upstream of the crosswalk? Pedestrians should cross downstream of the stop bar where vehicles wait for a red signal. For multilane crossings, where there is a high potential for multiple threat situations, an additional setback distance from the crosswalk is desirable. A stop bar downstream of the crosswalk would result in vehicles queuing onto the crosswalk, which is undesirable. This is a principle for signalized and unsignalized crosswalks and their position relative to the vehicular stop bar or yield line, respectively. 5. Is the signal or beacon control separated from other traffic control devices? Both roundabouts and CTLs have additional traffic control devices that control yielding and merging behavior at the roundabout entry and at the downstream end of the CTL. Any signals or beacons at the crosswalk need to be visibly separated to avoid driver confusion. For example, a green vehicle signal at a roundabout entry crosswalk may be misunderstood by drivers as providing a pro- tected movement into the circulating lane, unless the signal is sufficiently separated from the circulatory roadway. 6. Are audible messages provided and sufficiently separated? Any pedestrian signal or beacon installation requires the use of APS or other audible devices that convey the presence and functionality of the traffic control device to a pedestrian who is blind. These devices should be installed immediately adjacent to the crosswalk, aligned with the crossing direction, and down- stream of the approaching vehicles. Any audible devices further need to be separated from each other by at least 10 ft, or must have special speech messages, to uniquely tie the audible message to a crossing point. This is especially critical on splitter or channelization islands, which exist for both roundabout and CTLs. In some cases, larger island designs may be required to assure a separation of entry and exit devices, or of devices controlling the CTL versus the main intersection. Additional discussion on audibility considerations at both facility types is given in the next section. 7.3.13 Step 13. Complete Crosswalk Assessment When the candidate design satisfies the performance targets, the design can be finalized and the treatments can be implemented as applicable. As part of this assessment, the analyst con- ducted three explicit performance checks (Steps 4, 9, and 11), and compared estimates to the performance targets established by the agency to evaluate whether or not the candidate design meets the desired level of accessibility. The result of the crosswalk assessment is iterative by defi- nition and will prompt the analyst to accept, reject, or modify the candidate design. Depending on the outcome of the performance checks, the analyst may complete the crosswalk assessment (Step 13) or may repeat the process with a modified design after iterations in Steps 4, 9, or 11. While not explicitly called for, an assessment of vehicle impacts may be considered in this step. Chapter 2 of this guidebook presents the context of the accessibility evaluation within the broader intersection design process, which considers the expected operational and safety per- formance of each mode. By conducting the assessment of vehicle impacts in this step, the analyst may check for these impacts within the accessibility assessment.