Application of Pedestrian Crossing Treatments for Streets and Highways (2016)

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

13 CHAPTER THREE GUIDANCE AND CURRENT PRACTICES REGARDING SELECTING AND PRIORITIZING PEDESTRIAN CROSSING IMPROVEMENTS EXISTING RESOURCES FOR IDENTIFYING AND PRIORITIZING PEDESTRIAN CROSSING IMPROVEMENTS It is clear that crossings cannot be provided everywhere pedestrians may potentially cross a street. However, at certain locations, pedestrians need assistance from designs or traffic control devices or both to create sufficient gaps in traffic, to reduce speeds or conflicts with motor vehicles and exposure to potential crashes, and to reduce unnecessary delay for a level of safety and service that motorists typically expect. The guides highlighted here each offer information or process guidance that can help jurisdictions improve their practices. A number of the guides may be used in several ways, because they may provide information important for screening and identifying risks (in a systemic approach), as well as processes and safety evidence needed to identify and prioritize appropriate treatments. Other guides are also available and used by some agen- cies, but are not included here, either because they are more out of date or are not evidence based. The first resource described provides a model process and prioritization tool that any type of jurisdiction may apply to plan and prioritize any type of pedestrian or other multimodal transportation projects. NCHRP Project 07-17 developed a flexible model process that can be applied to identifying and prioritizing any types of pedestrian or bicy- cle improvements for existing roads. Prioritization Tool NCHRP Report 803: Pedestrian and Bicycle Transportation Along Existing Roads—ActiveTrans Priority Tool Guide- book (Lagerwey et al. 2015) Available at http:/onlinepubs.trb.org/onlinepubs/nchrp/ nchrp_rpt_803.pdf In 2014, researchers for NCHRP Project 07-17 created the ActiveTrans Priority Tool (APT) in an effort to provide the first national model pedestrian and bicycle prioritiza- tion tool that can be used by agencies with a wide spectrum of resources and capabilities. The tool is “a flexible, and Federal and national documents such as the MUTCD and AASHTO green books provide guidance on design and traf- fic control devices. However, these national guides alone provide insufficient information on identifying where cross- ings should be placed and how crossings are to be improved appropriate to the location type, road type, traffic volumes, pedestrian needs, and other concerns. A number of agencies have undertaken to fill some of the gaps. This chapter describes key state-of-the-practice guides and resources that have been developed to help jurisdictions understand issues related to pedestrian crossing needs, traf- fic and roadway factors that affect the safety effects of treat- ments, and how to select appropriate countermeasures. In addition, some examples of guides developed by state and local agencies are highlighted in this chapter and in case examples in chapter five. As described in chapter two, safety decisions are not made in a vacuum. Policies and local values and public interests, as well as local experiences with treatments, guide decision pri- orities, and local guides have been tailored to address such concerns. As documented by responses from jurisdictions, not all locations have the same successes with the same treat- ments, and they likewise do not want to approach pedestrian safety with the same toolbox used by other locations. This chapter describes practices and resources actually being used by jurisdictions to help make decisions. State and local jurisdictions also described some of their challenges in delivering safer pedestrian crossings. Among the challenges jurisdictions face are that safety barriers can hinder net- work access for pedestrians; consequently, there may be low crashes and low pedestrian numbers (and difficulty meet- ing traffic control device warrants), but significant latent demand and need for improved mobility and safety. A fur- ther challenge is the lack of crash-based evidence for some types of treatments, or details on where certain treatments are most effective. As discussed in the introduction, these and other issues sometimes make it difficult to determine or justify treatment priorities and to perform cost–benefit types of assessments to prioritize resources. The following two major sections of this chapter cover effective practice resources available, followed by a sum- mary of practices and guidance resources being used.

14 data-driven approach to prioritizing pedestrian and bicycle improvements along existing roads; and is designed to work for agencies at all levels of government with different pri- oritization purposes and different levels of technical capac- ity, available data, and experience” (Lagerwey et al. 2015). The APT tool consists of a guidebook and a programmed spreadsheet with preloaded prioritization factors, associated variables, and functions intended to save users time. The APT can be used for prioritizing a variety of location types, including road segments, intersections, and corridors. It also addresses facility types such as sidewalks and curb ramps. It consists of 10 steps and two phases. Phase I comprises an ini- tial scoping effort to define the purpose, determine how com- munity or agency values could be expressed through factor selection and weighting, select variables to represent the cho- sen factors, and assess data availability and technical resources. Phase II comprises the mechanics of calculating prioritization scores and developing a ranked list of improvement locations. Phase I starts at a high level and is used to tailor the tool to the specific agency’s goals and safety priorities. It includes six steps from defining the purpose to assessing data and technical resources (Figure 1). The steps proceed in a linear fashion but are also iterative. FIGURE 1 ActiveTrans Priority Tool general process. Source: Lagerwey et al. (2015). Phase II then calculates the scores for each improvement location based on the purpose, factors, weights, variables, and technical resources identified (Figure 1). The ultimate output of Phase II is a ranked list of priority pedestrian or bicycle improvement locations, along with a well-docu- mented process for how the list was developed. Through extensive multiagency and jurisdiction pilot test- ing, the APT was further refined and found to be useful in multiple planning processes, including master transportation plans; multimodal plans; Complete Streets plans; statewide needs and assessments; bicycle, pedestrian, and sidewalk master plans; Americans with Disabilities Act (ADA) tran- sition plans; districtwide or citywide Safe Routes to School plans; pedestrian and bicycle crash reduction plans and safety audits; and transit access plans (Lagerwey et al. 2015). Identifying Problems and Problem Locations A typical reactive assessment process might begin with crash analysis, community complaints, or public input during plan development that flags a problem location, whether controlled or lacking any existing facilities. From there, agencies will likely investigate the site, crash his- tory, and multimodal traffic characteristics and volumes to determine what actions, if any, to take. However, a more proactive approach could incorporate other types of systematic processes. Following are descriptions of tools available to help in various steps of the problem identifica- tion and prioritization process. In addition, some tools that have recently been developed by state and local jurisdic- tions are highlighted in chapter five. Additional guides for uncontrolled crossings have been highlighted separately following this section. Crash Analysis Tools Analyzing where crashes occur as well as the types of crashes occurring are ways to identify areas that may need improvement (Zegeer et al. 2009). The crash type provides information on the movements and actions of the motorist and pedestrian leading up to a collision that can help in selecting appropriate treatments to reduce the type of conflict that led to the crash. This information is often lacking in many state crash databases for pedestrian crashes. The spatial location of crashes is crucial to iden- tify in a systemic way land use and other characteristics of streets and the built environment that may be associ- ated with crashes or to identify high crash/crash rate loca- tions that may merit further investigation. Coding crash locations in geographic information system (GIS) or other spatially linkable format (if spatial data are not already available from the state or other transportation agency) is a key to knowing where collisions occur. Analyzing crash types can aid in identifying the types of treatments that may help remedy the problems.

15 The Pedestrian and Bicycle Crash Analysis Tool (PBCAT) software was created to help agencies type their pedestrian crashes if crash type information is lacking in their exist- ing crash databases, as it is for many (Harkey et al. 2006). PEDSAFE (further described later) is a companion tool with a matrix to help select countermeasures based on crash types or other performance objectives. Hundreds of agencies have downloaded the PBCAT software. Jurisdictions known to type and analyze their pedestrian crashes to help identify countermeasures include North Carolina; Wisconsin; and Boulder, Colorado. Once the database is developed, the Highway Safety Man- ual (HSM) (AASHTO 2010) provides guidance on a vari- ety of crash-based screening approaches, depending on the types of data available, that might be used to identify loca- tions potentially needing safety improvements. However, pedestrian crashes are relatively rare events and may not occur in concentrations at many particular locations. High crash locations also tend to move around over time. Agen- cies may want to apply systemic or proactive approaches to identify risks patterns and appropriate treatments to apply across many locations to maximize safety benefits. Systemic Screening Tools and Approaches At present there are a few tools and resources to help juris- dictions identify high risk locations before crashes occur, but an active NCHRP project (17-73) aims to develop a systemic pedestrian safety process. Two resources that are currently available include the following: • Pedestrian and Bicyclist Intersection Safety Indices (PedISI): User Guide (Carter et al. 2006, 2007), available at https://www.fhwa.dot.gov/publications/ research/safety/pedbike/06130/06130.pdf. • Pedestrian Road Safety Audit Guidelines and Prompt Lists (Nabors et al. 2007), available at: http://www. pedbikeinfo.org/cms/downloads/PedRSA.reduced.pdf. The PedISI proactive screening tool is available to help prioritize intersection crossings that may need further in- depth assessment (Carter et al. 2006). The index is a model equation and is applied at the crossing level (Figure 2). The index was developed from expert safety ratings and observed pedestrian and motorist interactions and conflicts data. The tool allows users to identify intersection crossings that may be priorities for in-depth pedestrian safety assess- ments. Observable characteristics such as number of lanes, area type, and type of traffic control, are used to produce a safety index score. Another way to diagnose problems more fully and help to narrow down potentially appropriate treatments is a road safety audit (RSA). RSAs are a formal, but qualitative, on- site safety assessment of streets, intersections, or other areas, and is conducted by an interdisciplinary, unbiased team of experts. It is important that pedestrian issues be addressed in any RSA, but pedestrian-focused RSAs can be conducted to focus on pedestrian safety issues and characteristics and suggest potentially appropriate remedies. RSAs can also be applied proactively, such as to certain types of corridors or a main thoroughfare through a small community, or even systemically, such as throughout an area. FHWA’s Pedestrian Road Safety Audit Guidelines and Prompt Lists provide more information on conducting an RSA to be used either proactively to identify locations that FIGURE 2 FHWA Pedestrian Intersection Safety Index model equation and index values. Reproduced from Table 1 in Carter et al. (2006).

16 may have safety issues, or reactively, once locations have been identified through crash data analysis, complaints, or other public or stakeholder input (Nabors et al. 2007). Pub- lic safety officials as well as engineering staff often provide input about problem areas, and can participate in RSAs. RSAs can also be conducted at any stage of the new proj- ect development process, preferably as early in the process as possible, to identify issues that could be addressed by the project, or even during the project (such as work zone design). Public input and complaints are other commonly used methods of identifying areas of concern that may need treat- ment and could be considered part of a systemic approach to address problems before they become crash statistics. A Resident’s Guide for Creating Safer Communities for Walking and Biking was created as a resource for community stakeholders to get involved in helping to make their commu- nities safer for pedestrians and bicyclists (Sandt et al. 2015). This resource also highlights 12 case examples of commu- nity-oriented programs that have helped achieve success in implementing pedestrian and bicycle safety improvements. Several of these examples highlight methods of effectively engaging all types of residents, including communities that are often underrepresented in more standard public pro- cesses. Also see Thomas et al. (2015a) for more information on public outreach methods used internationally. See case studies from Cambridge, Massachusetts, and New York City in chapter five on public outreach approaches used to assess satisfaction with traffic calming measures (Cambridge), and to engage the community about pedes- trian safety needs and culture change (New York City). The tools described in the next section can be used to help assess uncontrolled locations for application of poten- tial crosswalks and other treatments. Guides for Assessing Risks and Selecting Treatments at Uncontrolled Crossings The MUTCD does not establish clear warrants on when to place crosswalks at uncontrolled locations, but does indicate that crosswalk markings at nonintersection locations legally establish the crosswalk. However, a legal crosswalk alone may be insufficient to obtain motorist stopping or yielding to cross- ing pedestrians and achieve a high level of safety. The MUTCD further states, “In conjunction with signs and other measures, crosswalk markings help to alert road users of a designated pedestrian crossing point across roadways at locations that are not controlled by traffic control signals or STOP or YIELD signs.” The MUTCD also advises that an engineering study is needed before marking crosswalks at uncontrolled approaches and advises that the engineering study consider the number of lanes, presence of a median, distance from signalized inter- sections, pedestrian volumes and delay, average daily traffic, speed limit or 85th percentile speed, location geometry, pos- sible consolidation of crossing locations, availability of street lighting, and “other appropriate factors” (FHWA 2009, rev. 2012, pp. 383–384). The consideration of many of these ele- ments is, however, left open to interpretation. Relating to speed and geometry, stopping sight distance is among the factors that should be considered in deciding where to install (or not install) a marked crosswalk (Zegeer et al. 2005). For example, agencies should not mark a cross- walk where sight distance is poor, such as on or near a horizontal or vertical curve. The sidebar reproduces stop- ping sight distances as a function of speed from the 2009 MUTCD (FHWA 2009), and is provided in conjunction with work zone guidance. Sight triangles and potential visual obstructions, such as buildings, parking, and landscaping, should be considered in conjunction with stopping sight distance in selecting loca- tions or providing other treatments to mitigate issues. In addition to stopping sight distance for motorists, pedestrians also need adequate sight distance to select a gap and have time to cross at uncontrolled locations, known as pedestrian sight distance (Pedestrian Crossings: Uncontrolled Loca- tions 2014). As mentioned by survey respondents for this study, estab- lishing demand or need for crosswalks and other improve- ments is a challenge owing to a lack of pedestrian count data or models to estimate latent demand. Block lengths, street connectivity, and transit access are other factors that can vary locally and by jurisdiction that may affect the need to provide supplemental crossings at uncontrolled locations. Beyond establishing the need for a crosswalk, there are warrants for some treatments—in particular traffic signals and pedes- trian hybrid beacons—that may affect the ability to provide a safe crossing because many locations require more treat- ments than simply striping crosswalks. All of these factors, in addition to differing policies and priorities, help explain why jurisdictions may adopt a variety of practices with regard to providing safer crossings at uncontrolled crossing points. The next two resources described focus on additional safety risks and potential treatments to provide safer cross- ings at uncontrolled locations when community needs and engineering studies suggest locations merit crossing improvements. In addition to counting pedestrians crossing at locations along a segment, soliciting public input, consult- ing pedestrian plans, and assessing land uses and pedestrian generators, transit access, and distance to nearest controlled crossings are some of the ways that communities can iden- tify where crossing improvements may be needed (see e.g., Lagerwey et al. 2015). The existence of pedestrian crashes, particularly more severe injury crashes, may also suggest a need for crossing enhancements.

17 Stopping Sight Distance as a Function of Speed (Reproduced from FHWA 2009, Table 6E-1). Speed* Distance 20 mph 115 ft 25 mph 155 ft 30 mph 200 ft 35 mph 250 ft 40 mph 305 ft 45 mph 360 ft 50 mph 425 ft 55 mph 495 ft 60 mph 570 ft 65 mph 645 ft 70 mph 730 ft 75 mph 820 ft *Posted speed, off-peak 85th-percentile speed, or the anticipated operating speed FHWA’s Safety Effects of Marked Versus Unmarked Crosswalks at Uncontrolled Locations: Final Report and Recommended Guidelines (Zegeer et al. 2005) Available at http:/www.fhwa.dot.gov/publications/research/ safety/04100/. This study examined safety factors related to providing crosswalks at crossings with no stop or signal controls in place. The guidelines suggested in this report emerged from a study of 1,000 marked and 1,000 unmarked crosswalks at similar locations in 30 cities. The study reported that higher vehicle traffic volumes, higher pedestrian volumes, and a greater number of lanes were all associated with higher rates of pedestrian crashes. Raised medians and pedestrian cross- ing islands were associated with lower pedestrian crash rates (Zegeer et al. 2005). The study found that the presence of a marked crosswalk alone at uncontrolled locations on two-lane and low-volume, multilane roads showed no effect on pedestrian crash rates TABLE 1 MATRIX OF RECOMMENDATIONS FOR INSTALLING MARKED CROSSWALKS AND OTHER NEEDED PEDESTRIAN IMPROVEMENTS Roadway Type (number of travel lanes and median type) Vehicle ADT ≤ 9,000 Vehicle ADT > 9,000 to 12,000 Vehicle ADT > 12,000 to 15,000 Vehicle ADT > 15,000 Speed Limit* ≤ 48.3 km/h (30 mph) 56.4 km/h (35 mph) 64.4 km/h (40 mph) ≤ 48.3 km/h (30 mph) 56.4 km/h (35 mph) 64.4 km/h (40 mph) ≤ 48.3 km/h (30 mph) 56.4 km/h (35 mph) 64.4 km/h (40 mph) ≤ 48.3 km/h (30 mph) 56.4 km/h (35 mph) 64.4 km/h (40 mph) Two lanes C C P C C P C C N C P N Three lanes C C P C P P P P N P N N Multilane (four or more lanes with raised median)** C C P C P N P P N N N N Multilane (four or more lanes without raised median C P N P P N N N N N N N Note: Reproduced from Zegeer et al. (2005) Safety Effects of Marked Versus Unmarked Crosswalks at Uncontrolled Locations, Table 11, p. 54. These guidelines include intersection and midblock locations with no traffic signals or stop signs on the approach to the crossing. They do not apply to school crossings. A two- way center turn lane is not considered a median. Crosswalks should not be installed at locations that could present an increased safety risk to pedestrians, such as where there is poor sight distance, complex or confusing designs, a substantial volume of heavy trucks, or other dangers, without first providing adequate design features or traffic control devices or both. Adding crosswalks alone will not make crossings safer, nor will they necessarily result in more vehicles stopping for pedestrians. Whether or not marked crosswalks are installed, it is important to consider other pedestrian facility enhancements (e.g., raised median, traffic signal, roadway narrowing, enhanced overhead lighting, traffic-calming measures, curb extensions), as needed, to improve the safety of the crossing. These are general recommendations; good engineering judgment should be used in individual cases for deciding where to install crosswalks. * Where the speed limit exceeds 64.4 km/h (40 mph), marked crosswalks alone should not be used at unsignalized locations. ** The raised median or crossing island must be at least 1.2 m (4 ft) wide and 1.8 m (6 ft) long to serve adequately as a refuge area for pedestrians, in accordance with MUTCD and AASHTO guidelines. ADT = Average daily traffic. C = Candidate sites for marked crosswalks. Marked crosswalks must be installed carefully and selectively. Before installing new marked crosswalks, an engineering study is needed to determine whether the location is suitable for a marked crosswalk. For an engineering study, a site review may be sufficient at some locations, while a more in depth study of pedestrian volume, vehicle speed, sight distance, vehicle mix, and other factors may be needed at other sites. It is recommended that a minimum utilization of 20 pedestrian crossings per peak hour (or 15 or more elderly or child pedestrians) be confirmed at a location before placing a high priority on the installation of a marked crosswalk alone. P = Possible increase in pedestrian crash risk may occur if crosswalks are added without other pedestrian facility enhancements. These locations should be closely monitored and enhanced with other pedestrian crossing improvements, if necessary, before adding a marked crosswalk. N = Marked crosswalks alone are insufficient, since pedestrian crash risk may be increased by providing marked crosswalks alone. Consider using other treatments, such as traffic-calming treatments, traffic signals with pedestrian signals where warranted, or other substantial crossing improvement to improve crossing safety for pedestrians.

18 compared with unmarked crosswalks at similar locations. However, on multilane roads with traffic volumes above about 12,000 vehicles per day, a marked crosswalk alone was associated with higher pedestrian crash rates than similar locations with no marked crosswalks. Table 1 reproduces a matrix from the FHWA study, with guidelines on the number of lanes, volumes, and speed lim- its to guide considerations for implementing crosswalks. The guidelines further offer some suggestions for addi- tional treatments to consider for locations that fall into the P (possible increase in crashes may occur if crosswalks are used without other improvements) or N (marked crosswalks alone are insufficient) categories in the matrix. The report indicates that engineering assessment and judgment is still required before marking any crosswalk, as conditions such as sight distance or populations using the crosswalk may vary from one location to another. For example, the study also found that older pedestrians had higher crash involvement relative to their crossing expo- sure. In locations with significant older pedestrians, as well as school children or other special populations, the study cautions that treatments may need to be adjusted accord- ingly. There may be other local factors that can be diagnosed only from assessments of local data and through site evalua- tions such as a road safety audit. The study recommends that marked crosswalks alone (i.e., without traffic-calming treatments, traffic signals and pedestrian signals when warranted, or other substantial crossing improvement) are insufficient (N category) and should not be used under the following conditions: • Where the speed limit exceeds 40 mph (64.4 km/h), • On a roadway with four or more lanes without a raised median or crossing island that has (or will soon have) an average daily traffic (ADT) of 12,000 vehicles or greater, or • On a roadway with four or more lanes with a raised median or crossing island that has (or soon will have) an ADT of 15,000 or greater. The report goes on to highlight some of the types of addi- tional improvements that may help make uncontrolled cross- ings safer on higher-volume, multilane roads. The findings from the FHWA report are not intended to deter agencies from placing crosswalks at currently uncontrolled locations on multilane, higher-speed, higher-volume roads where pedestrians need to cross, but to carefully assess the location through engineering studies and, if a crossing is needed, to consider what treatments should be used to make it a safe one. See the case example description of Seattle, Washington’s Pedestrian Crosswalk Inventory and Improvement Program in chapter five. Seattle used the guidelines developed by the FHWA crosswalk study to assess all of the city’s midblock/ uncontrolled crosswalks and develop a treatment implemen- tation plan. Also, see how the District of Columbia adapted the guidance from this study for its lower speed conditions and identified appropriate treatments to provide crossing enhancements for different types of streets. TCRP 112/NCHRP 562: Improving Pedestrian Safety at Unsignalized Crossings (Fitzpatrick et al. 2006a) Available at http:/onlinepubs.trb.org/onlinepubs/nchrp/ nchrp_rpt_562.pdf This study focused in particular on traffic control devices at unsignalized crossings. The study had two main objectives: (1) to recommend selected engineering treatments to improve safety for pedestrians crossing high-volume, high-speed roadways at unsignalized intersections, in particular those served by public transportation; and (2) to recommend modifications to the Manual on Uni- form Traffic Control Devices for Streets and High- ways (MUTCD) pedestrian traffic signal warrant. (Fitzpatrick et al. 2006a) The guide does not cover school crossings. As suggested, the focus of this study was on traffic control devices, not design treatments. The study collected information on pedestrian vol- ume, street crossing width, and traffic volume and speed and examined four possible crossing treatment categories: marked crosswalk; enhanced, high-visibility or “active when present” devices; red signal/warning device or warning beacon device [primarily pedestrian hybrid beacons (PHBs) and rectangular rapid flash beacons (RRFBs)]; and conventional traffic control signals. Although significant additional research on the effects of PHBs and RRFBs has been conducted by these and other authors since this publication, there are still valuable insights that agencies may want to consider when weighing whether to use signs, signals, and beacons such as PHBs and RRFBs. The inputs into the study and summary of ongoing research needs were developed through a review of litera- ture, surveys of practitioners and pedestrians, assessments of pedestrian crossing designs, assessment of current pedes- trian signal warrants, and additional research including field studies of pedestrian and motorist behaviors such as pedes- trian walking speeds and motorist yielding. Among other findings from practitioners, the surveys found that across agencies, • A wide variety of treatments were being used, • Local agencies were more apt to experiment with inno- vative treatments and to seek modifications to current warrant requirements than were states,

19 • Practitioners were concerned that traffic signal war- rants then in place were too high to meet, • Treatment effectiveness varied by environment, and • Some practitioners interviewed at the time inter- preted the findings from FHWA’s Safety Effects of Marked Versus Unmarked Crosswalks at Uncontrolled Locations report recommendations as supporting a “mark versus do not mark” decision rather than a “mark versus more substantial treatment” decision. Interviews with pedestrians revealed that the most com- mon pedestrian concerns were • Unpredictability of motorists, in particular whether they will stop or yield at marked crosswalks; • Traffic volume (particularly turning traffic); and • Vehicle speeds. Pedestrians also typically felt safer with greater levels of vehicle control (i.e., traffic signals or red signal/beacon devices compared with warning beacons and marked cross- walks alone) (Fitzpatrick et al. 2006a). Other findings that could be considered regarding the placement (and enhancement) of crossings were the dis- tances pedestrians were willing to go out of their way to cross at a signalized intersection. The authors reported on these findings in the web-only companion document (Fitz- patrick et al. 2006b). When pedestrians at a midblock cross- ing were asked “if this crossing was not here, would you walk to the next intersection (point to intersection of inter- est)?,” about half of pedestrians indicated they would walk to the signalized intersection when the nearest was 200 ft away; and only about 25% of the respondents would walk to the nearest signalized intersection when it was located 550 ft (168 m), 950 ft (290 m), or 1,000 ft (305 m) from where they were crossing. At one site, a much higher percentage, 65% of pedestrians interviewed, indicated that they would be will- ing to walk to a signalized location 600 ft away, but at that site there were six lanes and high speed and high volume of traffic, and the only treatment at the midblock crossing was a marked crosswalk (Fitzpatrick et al. 2006b). New investigations were also conducted into the effects of three general types of treatments, including (1) high- visibility devices, which included high-visibility crosswalk markings and fluorescent yellow-green pedestrian crossing signs, in-street pedestrian crossing signs, and pedestrian crossing flags as well as other devices (in-pavement flashing crosswalk lights) and signs that attempted to draw attention to the crosswalks; (2) “active when present” warning-type flashing beacons; and (3) red signal/beacons [traffic sig- nals, pedestrian hybrid beacons (HAWK signals) and, half signals]. In addition, the study included sites with median crossing islands. Six measures of effectiveness for pedes- trian treatments were considered, including motorist com- pliance (yielding/stopping for pedestrians when required to do so), pedestrian delay, pedestrian compliance with the crosswalk, pedestrian visual search, pedestrian–motor vehi- cle conflicts, and pedestrian walking speed. Results of evaluations of the specific devices are reported in the detailed literature review and summarized in the rel- evant treatment sections in chapter four, along with many other studies. However, in general, Fitzpatrick et al. (2006a) concluded that motorist crosswalk compliance was consis- tently quite high—94% or above at all sites tested—with an average of 97% or higher for each of the signal control/bea- con-type devices. Results were more mixed for the high-vis- ibility devices and warning flashing-when-activated types of beacons (yellow-only indicators). The study also deter- mined that compliance rates for high-visibility and “flash- ing when present” types of devices were always better for any of the treatments used on two-lane roadways compared with four-lane roadways. (None of the signal treatments in the study were applied to two-lane streets.) Flags, pedestrian median islands, and high-visibility markings all had higher compliance rates on lower-speed roadways. Based on all of their findings, including the inputs from practitioners and pedestrians, and other research available at the time, the researchers developed Guidelines for Pedes- trian Crossing Treatments, as Appendix A to the report. This guidance splits primarily along speed limits (at 35 mph), with additional worksheets providing assistance for assess- ing roads with limits of greater than 35 mph, and another for streets with lower speed limits (see flowchart reproduced in Figure 3). The report developed recommendations for new guidelines for various pedestrian crossing measures. An Excel worksheet was also developed that combines Work- sheets 1 and 2 from Appendix A of the report, and allows practitioners to enter inputs to identify recommended treat- ments based on the recommendations developed in the study. (The Excel worksheet for pedestrian crossing treat- ments selection guidance from TCRP Report 112 / NCHRP Report 562 can be accessed at http://www.pedbikeinfo.org/ data/library/details.cfm?id=4949.) Countermeasure Selection Tools There are also a number of resources available to identify potentially appropriate countermeasures, once locations needing improvement have been identified. The Pedestrian and Bicycle Clearinghouse regularly updates information on pedestrian countermeasure evaluations and case exam- ples, and reviews new evidence and practices through its White Paper series and other resources. In addition, the PEDSAFE and BIKESAFE countermeasure selection tools were updated (in 2013 and 2014, respectively) with new countermeasures and case examples. PEDSAFE may also be used proactively to address more systemic goals and performance objectives.

20 Inputs for PEDSAFE Countermeasure Selection Tool 1. In what type of area is the roadway located? 2. What is the functional class of the roadway? 3. Is the problem at an intersection or midblock (roadway segment) location? 4. What is the vehicle volume at this location [expressed in terms of average daily traffic (ADT) for the primary roadway]? 5. Is vehicle speed low or high? 6. What is the number of through travel lanes (both directions)? 7. Is a traffic signal present, being considered, or not an option? 8. Is the roadway on a transit line/route (bus or rail)? 9. Is the roadway in a school zone or a school crossing? 10. Does the roadway contain a railroad crossing? 11. Is the roadway in a work zone? FIGURE 3 Flowchart for Guidelines for Pedestrian Crossing Treatments at Unsignalized Crossings. Source: Figure A-1 in Fitzpatrick et al. (2006a). PEDSAFE: Pedestrian Safety and Countermeasure Selection System (Zegeer et al. 2013) Available at http:/pedbikesafe.org/PEDSAFE/index.cfm Once a location or a systemic risk pattern or approach to treatment has been identified, the PEDSAFE countermea- sure selection system can be used to help identify potential countermeasures for any type of crossing location. PEDSAFE can be used to identify potential treatments for particular crash-type problems through the crash-type matrix or countermeasure selection tool, or in a more sys- temic approach using safety performance objectives or road- way descriptors or both in the objectives matrix or selection tool. This interactive Internet-based resource provides descriptions of 67 engineering, education, and enforcement countermeasures (Zegeer et al. 2013). The selection tool allows the user to input characteristics of the street or inter- section of interest to help narrow down the potentially appro- priate countermeasures for the context or crash-type pattern. A companion document to PEDSAFE summarizes the existing safety research: Evaluation of Pedestrian-Related Roadway Measures: A Summary of Available Research, pub- lished by the Pedestrian and Bicycle Information Center (PBIC) (Mead et al.) in April 2014. (This document was also used as a primary source for this synthesis, as previously mentioned.) PEDSAFE likely provides the most up-to-date research and expert-based guidance to practitioners for an initial identification of potentially appropriate countermeasures. In addition, PEDSAFE includes numerous case studies describing how agencies have used various treatments to help address pedestrian safety problems. Each countermea- sure description also provides a link to the safety research summary provided in the companion document. Toolbox of Countermeasures and Their Potential Effectiveness for Pedestrian Crashes (2013) Available at http:/www.pedbikeinfo.org/collateral/ PSAP%20Training/gettraining_references_pedTool boxofCountermeasures2013.pdf This toolbox documents 30 signalization, geometric, sign, marking, and operational measures that have crash- based estimates of effectiveness for pedestrian safety. The publication was last updated in February 2013 for FHWA. This publication resulted from reviews of crash-based evalu- ations of a wide range of roadway and engineering counter- measures related to enhancing pedestrian safety. The effect estimates are presented in the form of crash modification factors (CMFs), and the study references are provided. The signal-related measures that have safety benefits—that is, a CMF estimate of less than 1—include the following:

21 • Exclusive pedestrian signal phasing • Improved signal timing, including increasing pedes- trian walking period • Replacing traditional pedestrian signals with those with countdown timers • Modifying signal phasing to include a leading pedes- trian interval • Removing unwarranted signals on one-way streets • Converting permissive or permissive/protected left- turn signal phasing to protected-only phasing • Using the pedestrian hybrid (HAWK) signal • Installing traffic and pedestrian signals when warranted. Geometric treatments that were found to have a signifi- cant benefit to pedestrian crashes include the following: • Converting unsignalized intersections to a roundabout • Installing a pedestrian underpass or overpass • Installing raised medians at unsignalized crossings • Installing a raised pedestrian crossing • Installing raised refuge islands (pedestrian median islands) • Installing sidewalks or paved shoulders • Narrowing the roadway cross-section from four lanes to three lanes (two through lanes with a center turn lane). Sign, marking, and operational improvements having beneficial crash effects include: • Adding intersection lighting (applies to nighttime crashes) • Adding roadway section lighting (applies to nighttime crashes) • Improving pavement friction • Increasing enforcement • Prohibiting right turns on red • Prohibiting left turns • Restricting parking near intersections (to off street) • Providing high-visibility crosswalks • Providing high-visibility crosswalks in school zones. Estimates documented in this toolbox are included in the Tables of Countermeasures Effectiveness in chapter four. Other Pedestrian Safety Resources Safe Routes to School Guide (Pedestrian and Bicycle Infor- mation Center, et al. 2015) Available at http:/www.saferoutesinfo.org/ The Safe Routes to School Guide provides support for com- munities interested in increasing opportunities for children of all abilities to safely walk and bicycle to school. The Safe Routes to School (SRTS) concept began in Denmark in the 1970s. In 1997, the New York City borough of the Bronx, and the state of Florida both introduced pilot programs. In 2000, the U.S. Congress funded two SRTS pilot projects through the National Highway Traffic Safety Administration. Many other grassroots SRTS efforts began throughout the United States within a year of the launch of the pilot programs. The federal-aid Safe Routes to School Program was established by Congress in 2005 through comprehensive transportation legislation. Subsequent transportation legislation, including Moving Ahead for Progress in the 21st Century (MAP 21) passed in 2012 and makes SRTS activities eligible to compete for funding alongside other pedestrian and bicycle programs as part of a new program called Transportation Alternatives (Pedestrian and Bicycle Information Center, et al. 2015). The Safe Routes to School Guide (online) summarizes current recommended practices, mainly in regard to pro- cesses. However, it also provides guidance and numer- ous case studies demonstrating engineering treatments that have been implemented in school walking areas and crossings (http:/guide.saferoutesinfo.org/case_studies/ engineering.cfm). The guide outlines five principles for engineering solutions to improving the built environment for school-age pedestrians: • The physical environment often determines whether many children walk or bicycle to school. To safely walk or bicycle to school along a street or separate path, or to cross a street along the way, children need well-designed, well-built, well-maintained, and accessible facilities. • Accessibility is required. An important aspect of enabling children with disabilities to walk and bicycle to school is provision of accessible infrastructure. • The relationship of school buildings to sidewalks and street crossings can determine the level of comfort and safety a pedestrian or bicyclist experiences. All ele- ments are interconnected: the street is connected to the sidewalk and the sidewalk is connected to the building. Getting this relationship right is critical. • Focus first on implementing easy and low-cost solu- tions first to build momentum. Then identify and pri- oritize longer-term improvement needs. • Engineering treatments are matched to the type of problem (Pedestrian and Bicycle Information Center, et al. 2015). The guide also provides information on school route maps; understanding and improving the areas immediately around the school, along the school route, and at street crossings (see Crossing the Street page at http:/guide.saferoutesinfo. org/engineering/crossing_the_street.cfm); slowing down traffic; and links to other resources. The guidance is developed from existing research, although many treatments lack specific evaluation in school

22 areas. However, with the assistance of examples and case examples, the guidance is tailored to be appropriate for implementation in school walking routes, school zones, and other areas near schools. In addition, the guide notes that newer treatments, such as PHBs and RRFBs or raised inter- sections, may be tried in a community first in a school zone, where child safety is a prime concern. As such treatments gain acceptance, they may be used in other locations (Seth LaJeunesse, National Center for Safe Routes to School at the University of North Carolina Highway Safety Research Center, personal communication, Aug. 20, 2015). NCHRP Report 500, Volume 10: A guide for reducing collisions involving pedestrians (Zegeer et al. 2004) Available at http:/onlinepubs.trb.org/onlinepubs/nchrp/ nchrp_rpt_500v10.pdf AASHTO, in cooperation with FHWA, sponsored a series of reports (Report 500 series) to help state transporta- tion agencies in safety planning and implementation. The series was designed to help agencies develop plans and strat- egies to address key road safety problems to help meet their strategic safety objectives. The guide to reduce pedestrian crashes was released in 2004 (Zegeer et al. 2004). Although the report is somewhat dated in terms of evidence regarding traffic control devices and other treatments, it provides guid- ing safety principles, as well as significant information on considerations for applications of comprehensive treatments (e.g., engineering, enforcement, and education and public information) that are still relevant. In addition to planning guidance, the report outlines four major objectives for reducing collisions between motor vehi- cles and pedestrians. The four strategies include the following: 1. Reduce pedestrian exposure to vehicular traffic. 2. Improve sight distance and visibility for motor vehi- cles and pedestrians. 3. Reduce vehicle speed. 4. Improve pedestrian and motorist safety awareness and behavior. Proven, tried, or experimental but promising strategies are outlined under each of the four objectives. The strategies for objective 1—Reduce pedestrian exposure to vehicular traffic—include the following: • Providing sidewalks and curb ramps and protective barriers as needed (proven); • Providing median islands and raised medians (proven); • Adding or upgrading pedestrian signals, such as adding right-turn-on-red restrictions, pedestrian countdown signals (PCSs), and automated pedestrian detection (proven, tried, and experimental, respectively); • Providing grade-separated crossings (proven); and • Restricting or reducing traffic (proven and tried, respectively). Cautions, considerations, and keys to success are also described for each treatment. For example, continuous medians may not be appropriate for all circumstances and may sometimes result in higher speeds by reducing friction along sections. Strategies under objective 2—Improve sight distance and visibility for motor vehicles and pedestrians—include providing crosswalk enhancements (proven and tried). Crosswalk enhancements encompass a variety of engi- neering and enforcement measures, including improving conspicuity, enforcement campaigns, school route improve- ments, narrowing measures, installing and upgrading traf- fic signals, pedestrian islands, and others. The purpose of adding crosswalks at appropriate loca- tions is to guide pedestrians to the best location to cross, with enhancements such as high-visibility markings, raised crosswalks, lighting, and other physical roadway enhance- ments that reinforce the crosswalk or reduce vehicle speeds or both. This report also summarizes the FHWA crosswalk study findings, with appropriate cautions about not using only markings in the absence of other improvements on mul- tilane, high-volume, higher-speed roads. Other measures mentioned include • Crosswalk illumination/lighting (proven for nighttime crashes); • Eliminating visual obstructions/screening (tried); and • Signals to alert motorists that pedestrians are cross- ing (tried and experimental at the time), and improved reflectorization/conspicuity of pedestrians (tried). Measures identified under objective 3—Reduce vehicle speed—include road-narrowing measures through lane reductions, and lane width reductions by extending side- walks or landscaped areas through curb extensions (tried). Since publication of this report, there is evidence that reduc- tions in numbers of lanes through road diets often result in reduced traffic speeds and reduced crashes. Other measures to reduce vehicle speed include • Traffic-calming measures for sections such as ser- pentine streets and chicanes, chokers or roadway nar- rowing through a portion of a section, speed tables and humps, and low-speed shared streets (tried; e.g., Woonerfs or living streets); • Raised traffic-calming measures to lower speeds and crashes (proven); and

23 • Narrowing measures such as bulb-outs/curb exten- sions (tried). For intersections, measures include • Mini traffic circles, such as those widely used in Seattle, which had been found to significantly reduce crashes there (tried); • Speed tables/raised crosswalks and raised intersec- tions, which in some cases had improved motorist yielding to pedestrians (proven); • Roundabouts (tried); • Curb radius reductions and curb extensions, expected to reduce turning speeds at intersections (tried); and • School route improvements, such as design of the pick- up/drop-off areas, parking restrictions, school warn- ing signs and school zone speed limits, walking route maps, and adult crossing guards (tried). Educational and outreach measures are discussed in objective 4—Improve pedestrian and motorist safety aware- ness and behavior—because the approach taken by these guides is intended to be comprehensive. However, these measures are not covered in depth by this synthesis. Other sources include regularly updated information on countermeasures safety effects and proven countermea- sures. Additional sources that may aid pedestrian safety practitioners include those described below. FHWA’s Proven Safety Countermeasures Bulletins and Related Guidance Documents (FHWA 2015) Available at http:/safety.fhwa.dot.gov/provencounter measures/ The most relevant pedestrian safety measures currently included among FHWA’s proven countermeasures are road diets, medians and pedestrian crossing islands, pedestrian hybrid beacons, and roundabouts. Estimates for crash effects are described in the countermeasures descriptions in chapter four and the literature review. Crash Modification Factors (CMF) Clearinghouse Available at http:/www.cmfclearinghouse.org/ The CMF Clearinghouse contains a regularly updated searchable database of countermeasures with documented crash effects, including CMFs that were incorporated into the Highway Safety Manual (2010). The CMFs are quality rated using a star system that incorporates study design; sample size including years of data, diversity, and number of sites; size of the standard error compared with the CMF; control for potential sources of bias; and data sources (number and diversity of states). The star rating, standard errors, and descriptions of the crash types, road types, and area types applicable to the CMF provide information to help users determine the potential validity of the CMF to their situation. As new pedestrian crash (and other) CMFs are developed, they will be added to the clearinghouse. Countermeasures That Work (Goodwin et al. 2015) Countermeasures That Work describes major strategies and countermeasures and their effectiveness, costs, and implementation issues that state highway safety offices and their partners can use to help select effective behavioral countermeasures for traffic safety problems. This reference work covers nine major topic areas, including pedestrians, bicycles, and speeding and speed management (Goodwin et al. 2015). Speed Management Toolkit (see Thomas et al. 2015b) Available at http:/safety.fhwa.dot.gov/speedmgt/ref_ mats/docs/speedmanagementtoolkit_final.pdf Speed is an important consideration for pedestrian safety, with risk for adult pedestrians rising rapidly with impact speeds above approximately 25 mph (Rosén and Sander 2009). The HSM (2010) provides CMF estimates for total fatal and total injury crash effects expected for changes in average operating speed for a road. Because pedestrians are more vulnerable to injury than motor vehicle occupants when involved in a crash, reductions in speed might be expected to yield even greater safety benefits to pedestrians than for total crashes as predicted by the HSM. FHWA’s Speed Management Toolkit for speed-related countermeasures (Tables 9 and 10 of the toolkit) reproduces the estimates of total (not pedestrian) fatal and injury crash effects from the HSM (see Thomas et al. 2015b). Other speed- and crash-reducing countermeasures are also described in this toolkit, and companion documents provide examples of speed management action plans. Also see other FHWA speed management resources, including Engineering Speed Management Countermeasures: A Desktop Reference of Potential Effectiveness in Reducing Speed (2014) for tables of speed and crash-reduction effects for traffic-calming and other measures from prior studies. Methods and Practices for Setting Speed Limits (Forbes et al. 2012) Available at http:/safety.fhwa.dot.gov/speedmgt/ref_mats/ fhwasa12004/ This report discusses four methods for setting speed lim- its, including an injury minimization approach that is com- monly used in Vision Zero countries.

24 New York City, a Vision Zero city, has taken an assertive approach to managing speed, described in the case example in chapter five. Design Resources Although many of the previously mentioned resources describe design or geometric-based countermeasures and also touch on the importance of good design generally, the resources described here are the key national resources available to guide pedestrian facility design. If good design principles are followed, including pro- viding adequate separation of pedestrians from motorized traffic or designing for slower motor vehicle speeds, safety would be improved for both pedestrians and other users (Zegeer et al. 2004, p. V-2; NACTO n.d.). Good design may reduce the need for additional countermeasures to solve problems in the future. Design Resource Index (Pedestrian and Bicycle Informa- tion Center et al. 2015) Available at http:/www.pedbikeinfo.org/planning/facilities_ designresourceindex.cfm The Pedestrian and Bicycle Information Center released an index prepared for FHWA that documents specific sec- tions or page numbers where pedestrian-related design and traffic control features are treated in key national manuals (see Figure 4 for an example of some of the listings). There are also indices for on-street bicycle facilities and for shared- use paths, as well as a master list. The following nine resources (plus two other bicycle- oriented resources) were reviewed for the design index, and have information on pedestrian facilities listed in the Index: • Roadside Design Guide—AASHTO (2011) • A Policy on Geometric Design of Highways and Streets— AASHTO (2011) • Guide for the Planning, Design, and Operation of Pedestrian Facilities—AASHTO (2004) • Manual on Uniform Traffic Control Devices (MUTCD)— FHWA (2009, with 2012 revision) • Designing Walkable Urban Thoroughfares—ITE and Congress for the New Urbanism (2010) • Recommended Design Guidelines to Accommodate Pedestrians and Bicycles at Interchanges—ITE (2014) • Traffic Control Devices Handbook—ITE (2013) • Urban Street Design Guide—NACTO (2013) • Draft Guidelines: PROWAG, Shared Use Path Guidelines—U.S. Access Board, as of 2014 (Pedestrian and Bicycle Information Center, Design Resource Index n.d.) The index makes it much easier for practitioners to consult and cross-reference information from the most important guidance resources regarding particular designs or devices. PROWAG provides the most up-to-date guidance for accessible pedestrian facilities and crossings, although still in draft form. Once adopted in present or modified form, PROWAG guidance will provide required standards for accessible facilities. FIGURE 4 Image of partial listing of designs and coverage in guidebooks for pedestrian facility design in Design Resource Index, developed by the PBIC for FHWA. Source: PBIC (2015).

25 Of the other design guides mentioned earlier, the AAS- HTO guides are widely used by state and local jurisdictions. The MUTCD is a key resource for most agencies. NACTO’s Urban Street Guide (n.d.) This guide bears additional mention because a number of cities, in particular, are reporting use of this guide. See the Complete Streets section in chapter two for information on FHWA’s policy statement sanctioning use of NACTO and other (non-AASHTO) design guides. The Urban Street Design Guide articulates a Complete Streets vision, principles, and approaches to design, and pro- vides design examples for how to make different types of streets and intersections function safely for pedestrians. The guide articulates “design controls” that counter approaches from the past that were based largely on moving cars efficiently through rural highways, but have had unforeseen consequences on traf- fic speed and safety when applied in urban areas (NACTO n.d.). Guidance Focusing on Special Types of Crossings or Populations ITE, NCHRP, TCRP, FHWA, and other agencies have pro- duced many reports over the years focusing on pedestrian safety, traffic calming, and particular engineering challenges. Several design guides focus on unique types of crossings or special populations. The following three reports were developed to provide guidance regarding crossings at inter- change ramps, of rail/transit lines, and to provide guidance on installation of accessible pedestrian signals for visually impaired persons. Many others are also available, and state and local agencies often create their own design guides. • Recommended Design Guidelines to Accommodate Pedestrians and Bicycles at Interchanges: An ITE Recommended Practice (Mitman and Ridgway 2015) This guide, included in the Design Index, fills an impor- tant gap in shared knowledge on designing for pedestrians and bicyclists at interchanges. A number of the same treat- ments described in this synthesis are recommended for use by the guide. General design principles and current recom- mended practices were identified through a series of work- shops involving ITE members (Mitman and Ridgway 2015). Some of the principles include • Minimizing the number of conflict points; • Slowing ramp speeds until past crosswalks; • Placing crosswalks before the location where drivers will be looking for conflicting traffic from the left; • Highlighting conflict areas with high-visibility mark- ings, lighting, and so forth; and • Using advance stop/yield lines. A number of design examples for different scenarios are also provided. • TCRP Report 175: Guidebook on Pedestrian Crossings of Public Transit Rail Services (Fitzpatrick et al. 2015) Available at http:/www.tcrponline.org/PDFDocuments/ tcrp_rpt_175.pdf This guidebook, too recent to be included in the Design Index, assesses current practices in pedestrian rail treat- ments, and discusses issues associated with pedestrians (including special groups such as older pedestrians, child pedestrians, and people with disabilities) crossing public rail transit lines (Fitzpatrick et al. 2015). Treatments discussed in the guidebook include channel- ization, barriers, design, traffic control treatments (signs, signals, and warnings), other infrastructure including cross- ing gates and audible warning devices, and operational mea- sures such as reduced train speed and outreach programs. There is also information on railroad crossings in PED- SAFE, which links to additional resources. • NCHRP Web-Only Document 117A: Accessible Pedestrian Signals: A Guide to Best Practices (Harkey et al. 2007) Available at http:/www.apsguide.org/ This web-only NCHRP publication addresses most- effective practices for accessible pedestrian signals. The web content was developed from the report for NCHRP research project 3-62. The guide from this study is also available for download in pdf form from TRB’s online publications (Har- key et al. 2007). CURRENT PRACTICES In addition to reviewing the literature, the survey of states and local jurisdictions asked about practices and processes currently being used to prioritize pedestrian crossing improvements. These practices are summarized in the next three sections. Identifying Locations in Need of Improvement A number of different processes are used in conjunction with various types of programs and funding to identify locations in need of pedestrian crossing improvements. General Processes Used by States to Identify Locations Most states described multiple methods and processes, proj- ect types, and stakeholders that are used to identify locations

26 in need of pedestrian crossing improvements. These pro- cesses often included analysis of crash data and complaints, which are considered reactive processes—that is, they react to existing problems. A few states and local agencies are also using, or interested in using, more proactive system- atic approaches to identifying locations in need of crossing improvements. About half of states indicated use of crash data, with such data playing a major role for many. Several states have developed crash-based screening processes or use processes similar to those for all types of crashes. RSAs and scoping/plan reviews are also used to help prioritize loca- tions or opportunities by some state and local jurisdictions. In its survey response, California’s Department of Trans- portation (Caltrans) described a screening process: We first look at marked crosswalks across uncontrolled roadways where the speed limit exceeds 40 mph and the roadway has four or more lanes of travel and an ADT of 12,000 vehicles per day or greater. We also do a screen of fatal or injury collisions to look independently for pedestrian fatalities or injuries and investigate the locations that meet the parameters of the screen. Several states rely mostly or partly on local governments to initiate the process and identify specific needs, at least for part of the network. In Virginia, a twofold process is used, as described in the state’s survey response: For Cities who maintain their roadways, we rely on their inventory, crash, and volume information. They may submit for HSIP [Highway Safety Improvement Program] funds based on the risk based documentation of purpose and need. For the VDOT [Virginia DOT] maintained system (~85% of centerline miles) our network screening analysis for bike and ped crashes is done at the county route segment (length) level. A jurisdiction ranked listing of injury crash counts by severity and percentile range color coded maps . . . is generated each year. VDOT districts are then asked to define specifics for identifying and prioritizing safety improvements, using local knowledge and input, field reviews, and more detailed crash analysis when warranted. Oregon also described its process in its survey response: We don’t have a formal process. Needs are evaluated on a project-by-project basis. Public and local agencies request crossing improvements frequently. If money is available and the location meets the criteria in the Oregon Traffic Manual, a crossing is provided. About 15% of states indicated that they consult with local jurisdictions on transportation, multimodal, pedestrian, or land use plans when they exist; several states, includ- ing Arkansas, North Carolina, and Oregon, have statewide pedestrian plans that incorporate public input. Arkansas indicated the following: One of the ways we identify these locations is through our Statewide Pedestrian Master Plan. It identifies 31 recommended projects across the state. These improvements were identified through a review of existing conditions; GIS analysis of safety, connectivity, mobility, and pedestrian-oriented populations considerations; and stakeholder input (from the public, DOT staff, and other agency staff). Florida DOT (FDOT) indicated that many locations are identified by public requests, which initiate an engineering evaluation: “FDOT has also systematically identified the top 20 pedestrian crash locations in each district, with the focus of reducing crashes at those locations.” FDOT cur- rently bases prioritization on safety conditions including crash history and MUTCD warrants, but is working toward a “context-based system to supplement crash-based and war- rant based.” North Carolina DOT said the following: Candidate Pedestrian Crossing Safety locations are identified much like traditional safety location efforts— through a wide array of both information/evidence- driven corrective (historical black spot and network screening) mechanisms and approaches, as active project (new location and retrofit) efforts, general systemic (system modernization/infrastructure and connectivity improvement) efforts, community complaints/concerns and perceived risk, Pedestrian Plan-identified needs/ deficiencies, safety audits/review recommendations, development activities, and general statewide efforts aimed at improving modal accommodations and accessibility. In addition, North Carolina mentioned a wide variety of projects, programs, and agencies that participate in their efforts from the initial identification stage through program- ming and evaluation. Pennsylvania DOT uses a Bicycle & Pedestrian Checklist to assess bicycle and pedestrian needs on all construction proj- ects, but the checklist does not yet apply to resurfacing projects. Washington State DOT (WSDOT) indicated that crash evaluations and operational reviews are used for identify- ing potential needs, but that design criteria are also used for identification of needs. Planning documents and community requests are also used. Among the design guidance resources the state of Washington references are “Complete Streets, NACTO design guide, Context Sensitive Design, Practical Design, WSDOT design manual, and modal system plans, along with Guide for the Planning, Design, and Operation of Pedestrian Facilities and accessibility guidance.” General Processes Used by Local Jurisdictions to Identify Locations Similarly, a variety of types of processes were mentioned by local jurisdictions, including these by Santa Barbara, California: 1—Locations identified in annual high collision locations

27 2—Public requests for investigation 3—Locations identified as part of a plan 4—Locations identified using Safety Effects of Marked versus Unmarked Crosswalk criteria 5—Observation by Traffic Engineering or Police. Milwaukee, Wisconsin, described the following: 1. Preliminary engineering studies associated with large- scale capital improvement projects/paving 2. Land use developments 3. Complaints or requests from citizens 4. School operations, including new facilities 5. Changes in traffic patterns 6. Control needs and types identified through MUTCD warrant analysis and local policies 7. Prioritization based on federal fund or grant availabil- ity (i.e., HSIP, Transportation Alternatives Program, and Federal Aid Paving) and city capital improvement program fund allocation 8. Political influences. See the case example on San Francisco’s data-driven approach, using crash locations and collision profiles, to identify and prioritize intersections for improvement and to aid selection of low-cost countermeasures. Specific Tools/Methods Used by States to Identify Locations States were also asked about which specific procedures or tools they used to identify locations in need of pedestrian crossing safety improvements. The specific types of proce- dures reported used most often ( frequently or always used) by states include the following: • A majority of states (83%) indicated they frequently or always review for ADA compliance, using PROWAG or ADA Accessibility Guidelines guidance. • A majority (72%) also reported frequently or always assessing the need for pedestrian crossing improve- ments in conjunction with all types of projects (recon- struction, repaving, operations, and spot safety). However, in regard to measures that would help to iden- tify and treat existing problems, • Half of responding state DOTs (50%) indicated that they frequently or always conduct pedestrian crash analysis to help identify locations in need of improve- ments , and • Nearly half (47%) solicit and review public complaints (Table 2). Lower percentages of states frequently or always use the other types of procedures, including the following: • Review existing pedestrian plans—39% • Conduct road safety audits/assessments (RSAs)—36% • Collaborate with district or neighborhood assessments—28% • Solicit public input for specific project needs—25% • Perform multimodal level of service (LOS) or quality of service (QOS) assessments—5.6% • Perform pedestrian LOS or QOS assessments—2.8%. Table 2 summarizes the results for the states. The last column in Table 2 shows the weighted average of use of dif- ferent procedures (shown in descending order) with points assigned as follows: never used = 0 points; infrequently used = 1; sometimes used = 2; frequently used = 3; and always used = 4. These results show that assessment for ADA compliance comes closest, on average, to being always used (4 = always used) by all states reporting. Use of pedestrian level of ser- vice or quality of service assessments was the lowest, with a weighted average of < 1 (1 = infrequently used). (“Unable to determine” and missing answers were excluded from the weighted average calculations; if these had been included, with an assumption of “nonuse,” then the weighted averages of “use” would be considerably lower for those methods with a high missing response rate.) A variety of other types of analyses are frequently or always used by 25% of states to identify problem crossing locations. These include review of crash histories/crash reports (two states), and crash or warrants review plus site visits (two states). Other singular responses included analysis of pedes- trian count data, analysis of pedestrian generators, priority routing for school-related projects, conducting combined pedestrian/bicycle RSAs, general road safety analysis that includes pedestrians as well as every other type of crash for patterns that can be addressed, and engineering and safety/ crash investigations in response to complaints/concerns from the community and officials. Only one state (California) mentioned using a more sys- temic risk assessment approach in that it “look[s] for specific roadway and operational characteristics (multilane, inter- sections, higher volume, higher speed)” to identify locations that may need crossing safety improvements. (It should be

28 noted that only what the states indicated can be reported, and states may not have mentioned all the processes in use.) Other methods are also used by 14% of states to gather information for identifying locations. These include gather- ing comments from public meetings; looking at access in and around major train stations and transit stops (Mary- land); Safe Routes to School funding evaluation and plan overview (Wisconsin); other types of planning processes, including Transportation System Plans (Oregon), Strategic Highway Safety Plan collaboration (California), and fatal crash investigations. Specific Tools/Methods Used by Local Jurisdictions to Identify Locations There is a tendency for the local jurisdictions that responded to the survey (a select, nonrandom group) to use more pro- cedures more often to identify locations for improvements. In particular, methods relating to planning and public input are used more often by local jurisdictions than by the states. Differences in processes and emphases between states and local jurisdictions were expected, because there are many differences in the size and scope of responsibilities, particu- larly with regard to planning and public input between most local jurisdictions compared with most states. The proce- dures used frequently or always by a majority of the local jurisdictions surveyed include the following: • Assess the need for pedestrian crossing improvements in conjunction with all types of projects (reconstruc- tion, repaving, operations, and spot safety)—83% • Solicit and review public complaints—83% • Review ADA compliance—78% • Review existing pedestrian plans—67% • Collaborate with district/neighborhood evaluation—56% • Solicit public input regarding specific project needs—56% • Conduct crash analysis—56% (Table 3). The other methods asked about specifically are used fre- quently or always by fewer than half of the local jurisdictions that responded, including the following: • Road safety audits/assessments—28% • Multimodal level of service (LOS)/quality of service (QOS)—11% • Pedestrian LOS/QOS—11%. Other types of analysis, used by 22% of local jurisdic- tions, include the following: • Gaps assessment (three of 18 localities) TABLE 2 TYPES OF PROCESSES USED TO IDENTIFY LOCATIONS IN NEED OF IMPROVEMENTS—SUMMARY OF RESULTS FOR STATES Process or Method Proportion— Frequently (3) or Always (4) Proportion— Sometimes (2) Proportion— Infrequently (1) or Never (0) Proportion—Unable to Det./No Answer Weighted Average (excluding missing answers)1 Review—ADA 83.3% 8.3% 5.6% 2.8% 3.4 Review for all types projects 72.2% 19.4% 5.6% 2.8% 2.9 Crash analysis 50.0% 30.6% 19.4% NA 2.4 Solicit/review complaints 47.2% 38.9% 11.1% 2.8% 2.5 Review existing pedestrian plans 38.9% 36.1% 19.4% 5.6% 2.3 RSA 36.1% 41.7% 19.4% 2.8% 2.2 Collaborate with district/neighbor- hood evaluation 27.8% 47.2% 19.4% 5.6% 2.1 Solicit public input—specific project needs 25.0% 22.2% 47.2% 5.6% 1.8 Multimodal LOS/ QOS 5.6% 16.7% 61.1% 16.7% 1.4 Pedestrian LOS/ QOS assessment 2.8% 13.9% 72.2% 11.1% 0.8 Other type analysis 25.0% 13.9% 16.7% 44.4% NA Other methods 13.9% 8.3% 8.3% 69.4% NA Note: Thirty-seven states responded to the question. NA = not available. 1 “Unable to determine” and missing answers were excluded from the weighted average calculations; if these had been included, with an assumption of “nonuse,” then the weighted averages of “use” would be considerably lower for those methods with a high missing response rate.

29 • Safe Routes to School or school audits (two localities) • Engineering analysis (one locality) • Warrants analysis (signals, crosswalk, midblock cross- ing) (one locality mentioned). Other methods of obtaining information to identify locations, used by 17% (three) of local jurisdictions, include the following: • Staff observations/knowledge (one locality) • Community requests (one locality) • Public input for all major projects (one locality) • Tracking locations where a specific treatment (one widely used and known about in the community) is requested. Identifying Countermeasures States Countermeasure Selection Procedures With two exceptions, less than half of the states reported frequently or always using any particular resources or tools to help select pedestrian countermeasures. This may reflect, in part, the relatively few treatments or designs with crash-based safety evidence for various location types, but there are no data to support this conjecture. In addition, the responses may reflect a lack of awareness of the exist- ing tools that could be used to help select potential coun- termeasures. However, jurisdictions were not asked about their reasons for using or not using the tools, including prior awareness of the tools, only whether they used them. Rea- sons for lack of use may also vary across jurisdictions. The procedures used most often ( frequently or always used) by states (Table 4) include the following: • A majority of states (81%) report that they frequently or always assess ADA compliance during the counter- measure identification process. • The second most often used method was to solicit pub- lic input (53% frequently or always used). • Internal design resources were used frequently or always by 42% of states. • The CMF Clearinghouse was used frequently or always by 36% of states responding. • Of the states, 39% reported that they use cost–benefit assessment to help select countermeasures, whereas 44% indicated that they infrequently or never do. This finding may relate, as already mentioned, to difficul- ties in estimating crash effects for measures for which there are only behavioral data or no data. Other resources were less commonly consulted on aver- age. Again, the weighted average column reflects the aver- age level of use on a four-point scale from “never used” (0) to “always used” (4) (Table 4). TABLE 3 TYPES OF PROCESSES USED TO IDENTIFY LOCATIONS IN NEED OF IMPROVEMENTS—SUMMARY OF RESULTS FOR LOCAL JURISDICTIONS Process or Method Used Proportion— Frequently (3) or Always (4) Proportion—Sometimes (2) Proportion— Infrequently (1) or Never (0) Proportion—Unable to Det./No Answer Weighted Average (excluding missing answers) Review for all types projects 83.3% 16.7% 0 0 3.3 Solicit/review complaints 83.3% 11.1% 5.6% 0 3.3 Review—ADA 77.8% 11.1% 11.1% 0 3.2 Crash analysis 55.6% 27.8% 11.1% 5.6% 2.6 Review existing pedestrian plans 66.7% 5.6% 27.8% 0 2.6 Collaborate with district/neighbor- hood evaluation 55.6% 33.3% 11.1% 0 2.6 Solicit public input—specific project needs 55.6% 16.7% 27.8% 0 2.4 RSA 27.8% 33.3% 38.9% 0 1.9 Multimodal LOS/ QOS 11.1% 11.1% 66.7% 11.1% 1.4 Pedestrian LOS/ QOS assessment 11.1% 1 27.8% 55.6% 5.6% 1.2 Other type analysis 22.2% 5.6% 11.1% 61.1% N/A Other methods 16.7% 5.6% 0 77.8% N/A Note: Eighteen cities and counties responded to this question. 1 No agencies reported always using pedestrian LOS or QOS.

30 A few other methods were also mentioned as being used, including engineering judgment; coordination (by states) with local governments; web search of measures implemented in other states and cities; in-service evaluation of similarly treated sites; and other internal policies, procedures, and plans. Other external countermeasures resources mentioned by state agencies included the HSM (California), and a “case- dependent” approach, described by North Carolina as fol- lows: “Depends on the case, and we will attempt to find relevant sound research that pertains to the countermeasure and intended candidate application.” Other external design resources used include • Staff knowledge, including specialized staff or work team—three states; • State guidance for pedestrian crossing treatments— two states (at least one additional state has guidance under development or in the training phase); and TABLE 4 RESOURCES AND TOOLS COMMONLY USED BY STATES TO HELP SELECT COUNTERMEASURES Tool or Resource Used Proportion— Frequently (3) or Always (4) Proportion—Sometimes (2) Proportion— Infrequently (1) or Never (0) Proportion—Unable to Det./No Answer Weighted Average (excluding missing answers) ADA guidelines 80.6% 16.7% 0 2.8% 3.2 Solicit public input 52.8% 33.3% 8.3% 5.6% 2.6 Internal design resource 41.7% 16.7% 11.1% 30.6% 2.5 Cost-benefit analysis 38.9% 8.3% 44.4% 8.3% 1.8 Crash Modification Factors Clearinghouse 36.1% 36.1% 25.0% 2.8% 2.1 FHWA’s Safety Effects of Marked Versus Unmarked Crosswalks at Uncontrolled Locations 27.8% 47.2% 16.7% 8.3% 2.1 External design resources 22.2% 30.6% 13.9% 33.3% 2.2 FHWA Toolbox of Countermeasures and Their Potential Effectiveness for Pedestrian Crashes 22.2% 52.8% 22.2% 2.8% 2.0 Internal counter- measures resource 19.4% 33.3% 16.7% 30.6% 2.0 NCHRP Report 562, Improving Pedestrian Safety at Unsignalized Crossings 19.4% 41.7% 25.0% 13.9% 1.9 Consult other origi- nal research 19.4% 0 30.6% 50.0% 1.7 Use other methods 19.4% 2.8% 11.1% 66.7% N/A PEDSAFE 13.9% 25% 58.3% 2.8% 1.2 Pedestrian LOS/ QOS assessment 8.3% 11.1% 66.7% 13.9% 0.9 Other external countermeasure selection tools 5.6% 19.4% 25.0% 50.0% 1.2 Multimodal LOS/ QOS 5.6% 5.6% 69.4% 19.4% 0.8 Note: Figures showing detailed distributions of responses to how locations are identified by states and the local jurisdictions that participated are provided in Appendix B. N/A = not available.

31 • Other state manuals and technical guidance, including traffic engineering manual—two states. One state each mentioned the following resources: • State pedestrian (or bicycle) safety implementation plan • State guidelines for installation of marked crosswalks • CMFs with local data • State project development and design guidance • Operations policy directives. Regarding ADA compliance, a plurality of states (12) men- tioned using PROWAG guidelines. Several also use internal policies or design guides. Some also have specialists within the department to assist with ensuring ADA compliance. Other external design resources consulted by states to help identify appropriate treatments most often include the AASHTO (pedestrian facilities) guide, used by six states (17% of responding states). Other guidance used included the following: • NACTO guidance • ITE Traffic Control Devices Handbook • Nonspecific NCHRP reports • Nonspecific FHWA guidance • Boulder crosswalk guide and FHWA crosswalk study, both used by New Hampshire • Vermont state guidelines (VTrans 2015 Guidelines for Pedestrian Crossing Treatments). Thirteen states in all referenced using various internal high- way or road design guides, manuals, standard drawings, poli- cies, or bulletins, while one mentioned a traffic control device manual and one mentioned a pedestrian streetscape guide. Local Jurisdictions Countermeasure Selection Procedures Local jurisdictions also frequently solicit public input to help select countermeasures (68% of those responding to the survey). In addition, seeking ADA guidance is also widely used (63%), but to a somewhat lower extent than states reported (Table 5). Oakland, California, adopted a policy in 2015 for selecting pedestrian treatments at signalized intersections. The methodology was based on a review of current practices implemented by other cities (on NACTO’s distribution list), and the types of pedestrian–motor vehicle conflicts, intersections, and land use features present. Other measures often used are • Consulting the FHWA study on the safety effects of marked versus unmarked crosswalks (47%), and • Consulting internal countermeasures resources (42%). A number of local jurisdictions (53% at least sometimes) also refer to TCRP Report 112/NCHRP Report 562 on pedestrian safety at unsignalized crossings described in chapter three. Only 16% of local jurisdictions frequently perform cost– benefit assessments, with another 26% sometimes doing so. PEDSAFE was not often consulted by this set of juris- dictions. Only 16% consulted it frequently or always, with another 16% sometimes using this resource. The PBCAT tool for typing crashes, which can be used alone or in con- junction with PEDSAFE, also was not mentioned by state or local jurisdictions as a resource to help identify appropriate countermeasures, although the survey did not inquire spe- cifically about this tool. The FHWA Toolbox of Countermeasures containing CMFs for pedestrian crashes was consulted frequently by about 32% of those responding, with another 21% consult- ing it sometimes. Again, LOS/QOS tools were not frequently used by many agencies (about 10% each, with similar per- centages using these methods sometimes). Finally, a little over half of the local jurisdictions (56%) and exactly half of state jurisdictions responded that they use a formal design or treatment selection review process. When jurisdictions did have such a process, some of them appeared to be more “front end” processes such as collision analysis and warrants assessment, but were sometimes followed up with design quality assessment reviews. One agency men- tioned that the processes included review by citizen commit- tees and that sometimes issues did not turn up in the initial plans and scoping reviews, but were identified later. New York City involves a wide variety of partner agencies or divi- sions in reviewing significant treatment plans. Additional Resources or Needs Identified by Jurisdictions Perhaps some of the most helpful information came from asking the jurisdictions what type of additional guidance or resources they require. State and local jurisdictions were asked several questions about the need for additional guid- ance to identify problem locations or appropriate solutions. When asked, Does your jurisdiction need additional guidance on how to identify locations for pedestrian crossing improvements? only two (11%) of the selected local jurisdictions responding to the question indicated they needed additional guidance on identifying locations for crossing improvements. • Both of those identified needs related to numbers of users—one was a desire to be better able to quantify

32 latent demand at some locations, and the other related to deriving costs and benefits of threshold numbers (both vehicle and pedestrian volumes) for midblock crosswalks. • Another need mentioned by one of these jurisdictions was for additional guidance about channelized islands, including thresholds and benefits. In contrast, 57% of the states responding indicated that they could use additional guidance on identifying locations in need of crossing improvements. Most often (nine states), the focus was on means of prioritizing among locations need- ing treatment as opposed to initial identification of locations (Table 6). An illustrative answer from North Carolina was, TABLE 5 RESOURCES AND TOOLS COMMONLY USED BY LOCAL JURISDICTIONS TO HELP SELECT COUNTERMEASURES1 Tool or Resource Used (local) Proportion Frequently (3) or Always (4)2 Proportion Sometimes (2) Proportion Infrequently (1) or Never (0) Proportion Missing/ Unable to Answer Weighted Average (excluding missing answers) Solicit public input. 68.4% 10.5% 10.5% 10.5% 2.9 Consult ADA guidelines. 63.2% 5.3% 15.8% 15.8% 2.8 Consult internal design resource. 57.9% 5.3% 10.5% 26.3% 2.9 Consult FHWA Safety Effects of Marked Versus Unmarked Cross- walks at Uncon- trolled Locations. 47.4% 15.8% 15.8% 21.1% 2.6 Consult internal countermeasures resource. 42.1% 5.3% 15.8% 36.8% 2.4 Consult NCHRP 562, Improving Pedestrian Safety at Unsignalized Crossings. 36.8% 15.8% 26.3% 21.1% 2.1 Consult external design resources. 31.6% 10.5% 15.8% 42.1% 2.4 Consult FHWA Toolbox of Coun- termeasures and Their Potential Effectiveness for Pedestrian Crashes. 31.6% 21.1% 31.6% 15.8% 1.9 Consult other exter- nal countermeasure selection tools. 26.3% 10.5% 26.3% 36.8% 1.8 Consult Crash Modification Fac- tors Clearinghouse. 26.3% 5.3% 42.1% 26.3% 1.5 Perform cost–bene- fit analysis. 15.8% 26.3% 42.1% 15.8% 1.5 Consult PEDSAFE Countermeasure Selection Tool. 15.8% 15.8% 47.4% 21.1% 1.3 Consult other origi- nal research documents. 10.5% 15.8% 10.5% 63.2% 2.0 Perform pedestrian LOS or QOS assessment. 10.5% 10.5% 52.6% 26.3% 1.1 Perform multi- modal LOS or QOS assessment. 10.5% 15.8% 52.6% 21.1% 1.0 Use other methods. 5.3% 0.0% 5.3% 89.5% 2.0 1 Eighteen local jurisdictions responded to the question. 2 Numbers in parentheses represent the answer weight in the weighted average in the right-most column.

33 It should be noted that certainly far more candidate locations than could ever be treated can be readily identified through existing data-driven and systemic mechanisms/ approaches—but what is helpful are decision support tools that help apply limited resources to the most beneficial locations/sites that are best utilized. Additional spatial analysis and analytic-based tools would be of value. The North Carolina Department of Transportation (NCDOT) currently uses an expert team to estimate crash effects for treatments that currently lack such estimates. Other jurisdictions such as San Francisco have applied similar approaches. See the case example on San Francisco’s data- driven approach to prioritize improvements in chapter five. TABLE 6 SUMMARY OF STATE RESPONSES TO TYPE OF GUIDANCE NEEDED FOR IDENTIFYING LOCATIONS FOR PEDESTRIAN CROSSING IMPROVEMENTS Type of Guidance Needed No. of States Prioritization-related information/guidance (with specific mentions of economic and health valuations; identifying generators and receptors; incorporating community goals; using crowd-sourced data for pub- lic input; need for CMFs; use of tools such as spatial analysis, United States Road Assessment Program, warrants other states are using) 9 Proactive/systemic approach to identifying risks (versus waiting for crashes to happen), including one mention of context-based guidance 7 Coordinating with other planning (e.g., nonmotor- ized, Complete Streets) or project processes (i.e., integrating needs early in project scoping) 2 Crosswalk/rural crosswalk guidance 2 The second most common need expressed was for proactive systematic approaches to identify locations, as opposed to rely- ing only on crashes to do so (mentioned by at least seven states). Other needs mentioned related to coordinating with other planning or project processes, or guidance on implementing crosswalks (in general, or in rural locations). When asked, What type of resource that does not already exist would be most beneficial to help your jurisdiction select pedestrian crossing improvements? answers focused most commonly on a need for tools and resources to help with prioritization, including better crash- based (CMFs) evidence, and perhaps the need for better “pack- aging” of the information that (may) already exists to enhance cost-effective and context-appropriate decision making. A few answers from states are as follows: A guideline that address(es) [sic] when to use various devices (as to what to look for when assessing the problem location) relative to the realistic expectations each device or treatment can address. It should provide some uniform basis for making the decision to install it and when to augment it with another treatment/device or when to consider something different. A tool that analyzes the network and conditions where crossing improvements should be installed and then suggests a countermeasure and prioritizes the locations, such as USRAP [United States Road Assessment Program], cloud-based public comments, etc. In some ways it appears that much of the information is available but not very conducive to being readily compiled. Better organization of the knowledge and tools would be helpful as would more decision support/expert- type tools. Tools that better focus efforts toward locations with greatest need and most potential for improvement/ public benefit from investment of limited safety resources. Additional warrants or guidance for countermeasures that are not signals was mentioned, but the respondent clari- fied that warrants were not necessarily what was needed: For example, in our developing guidance document we are considering a “warrant” for marked and signed, nonsignalized crosswalks based on sight distance and the ability of the pedestrian to cross the roadway and clear the roadway based on available sight distance of the approaching vehicles. Signal installations have clear thresholds currently and hopefully simple signed/ marked crosswalks will also have clear thresholds. Then we have room for flexibility in countermeasures for crossings that don’t meet either that range from prohibiting crossings using barriers (for locations where a significant generator is not too far from a designated crossing), to moving generators (such as bus stops or sidewalks leading from developments, to installing some type of enhanced sign or signal treatment. [Many of the pedestrian treatments are determined based on engineering judgment. It would be helpful to provide some more guidance but not necessarily warrants.] More information or guidance about when to use HAWKS and PHBs was mentioned several times, including ways to assign the right type of project to the right level of risk. The respondent went on to address the need for incorporating latent demand: “Provide warrants that account for latent demand (many pedestrian situations are like the chicken and the egg … pedestrians can’t cross so they don’t try), clearer warrants for large investments such as signals and HAWKS [PHBs].” Others pointed out this need as well: We are always in need of tools to help predict usage on facilities that don’t currently exist. We have many corridors that do not currently have pedestrian facilities (such as sidewalks, shared use paths, etc.) so it is difficult to either justify where a facility is needed or determine what type of usage may occur to determine which type of facility would be most appropriate. A local jurisdiction also mentioned the need for a latent demand predictive model.

34 A state and a local jurisdiction also mentioned the need for improvements in guidance or improvements in practice relating to the balance of vehicular LOS with pedestrian needs. Responses include the following: Sometimes pedestrian crossings are not installed because movement along the roadway is prioritized. Some people in the community feel a Complete Streets policy could overcome this bias in mode choice (driving versus walking). I don’t think pedestrian crossings should be as onerous to install as they are. Traditional engineering practice places a lower value on walking than on driving and maintaining an effective LOS. Guidance related to acceptable reductions in vehicular LOS in order to benefit nonmotorized mobility and safety. One state mentioned the need for “low-cost methods to count pedestrians or determine exposure.” A local jurisdiction expressed the need for more infor- mation on effects of FHWA’s proven countermeasures in an urban environment. NACTO, in addition to local and other design guidance, has become a useful tool for Charlotte, North Carolina. One jurisdiction mentioned the needs for the following: Widespread general pedestrian engineering education. Hands-on field studies. Knowledge of the availability of documents and training in some of the primary resources being made available locally, and without excessive costs.