Improving Pedestrian Safety at Unsignalized Intersections

Kay Fitzpatrick and Shawn Turner and Marcus A Brewer
Institute of Transportation Engineers. ITE Journal

Apr 30, 2007 20:00 EDT


Increasing vehicle volumes and congestion cause changes in traffic conditions. These changes also affect pedestrians' ability to safely cross many roadways. Although numerous treatments exist at unsignalized crossings, there is growing concern about their effectiveness. There is a need to identify and study selected treatments to determine their effectiveness.

A recent research project jointly sponsored by the Transit Cooperative Research Program (TCRP) and the National Cooperative Highway Research Program (NCHRP) was initiated to address this particular need. The research was conducted by the Texas Transportation Institute.

One objective of the research effort was to recommend selected engineering treatments to improve safety for pedestrians crossing at unsignalized locations, particularly those served by public transportation. Another objective was to examine the Manual on Uniform Traffic Control Devices (MUTCD) pedestrian signal warrant.1 There is concern that the existing traffic signal pedestrian warrant may need to be modified.

The research team felt that it was necessary to provide more than just a menu of possible treatments. The team developed quantitative guidelines to help engineers and transit agencies determine which treatments are recommended for different street environments and traffic conditions. The team evaluated a selection of pedestrian crossing treatments and documented the evaluation results.

Guidelines for Pedestrian Crossing Treatments, which is included in the research report, was developed using the results from the evaluations.' An example is included in this feature. Details on the field studies and other research efforts are contained in the appendices of the research report.3

In accomplishing the two main study objectives, the research team also developed useful supporting information on various aspects of pedestrian safety at unsignalized roadway crossings, including pedestrian characteristics (walking speed, gap acceptance and treatment activation behavior); motorist yielding; and traffic engineering and transit agency perspectives. Following is a summary of the findings for the two major objectives and key supporting efforts.


Pedestrians have a wide range of needs and abilities. The 2003 edition of MUTCD includes a walking speed of 4.0 feet per second (ft./sec), or 1.2 meters per second (m/sec.), for calculating pedestrian clearance intervals for traffic signals. It also includes a comment that where pedestrians who walk slower than normal or pedestrians who use wheelchairs routinely use the crosswalk, a walking speed of less than 4.0 ft./sec. (1.2 m/sec.) should be considered in determining pedestrian clearance times. Other research studies have identified pedestrian walking speeds ranging from 2.2 to 4.3 ft./sec. (0.6 to 1.3 m/sec.).4

One of the pedestrian characteristics collected during field studies was the time for pedestrians to cross to the middle of the street or median and then to the other side of the street. Using the distances being traversed, the walking speeds of pedestrians were determined. To permit comparisons with other studies, the data were grouped to reflect the following:

* Younger, including pedestrians between the ages of 13 and 60

* Older, including pedestrians older than 60 or elderly

The following conclusions were developed for walking speed:

* The 15th-percentile walking speed for younger pedestrians was 3.77 ft./sec. (1.15 m/sec.) (sample size of 2,335); the 15th-percentile walking speed for older pedestrians was 3.03 ft./sec. (0.92 m/sec.) (sample size of 106).

* The older pedestrian groups (male, female and both genders) had 15thpercentile walking speeds that were statistically different from the 15thpercentile walking speeds of younger pedestrians.

* Two studies with databases containing more than 2,000 pedestrian crossings were the 1996 Knoblauch et al. study (data collected in 1993) and the TCRP/NCHRP study.5 The data collected in 2003 for the TCRP/ NCHRP study identified a slower walking speed for the younger group (3.77 ft./sec. [1.15 m/sec.]) as compared to 4.02 ft./sec. (1.23 m/sec.) found in the 1993 data collected for the Knoblauch et al. study.

* When both older pedestrians and younger pedestrians were considered using the Knoblauch et al. data (sample size of 4,459), the 15th-percentile value of 3.53 ft./sec. (1.08 m/sec.) was determined.

* Using population projections and the 15th-percentile walking speeds for each population group older than 15 years old, the proportionally weighted 15th-percentile walking speed for 2,045 was 3.56 ft./sec. (1.09 m/sec.).

Comparing the findings from the TCRP/ NCHRP study with previous work resulted in the following recommendations:

* 3.5 ft./sec. (1.1 m/sec.) walking speed for general population

* 3.0 ft./sec. (0.9 m/sec.) walking speed for older or less able population

These values were endorsed by the National Committee on Uniform Traffic Control Devices (NCUTCD) as guidance for use in signal timing for the upcoming revision to MUTCD.


The research team chose motorist compliance (yielding or stopping where required) as the primary measure of effectiveness for engineering treatments at unsignalized roadway crossings. Motorist compliance data were collected at 42 study sites that included nine different types of pedestrian crossing treatments. Table 1 shows examples of the treatments.

In addition to collecting motorist yielding behavior for general population pedestrians, data collection personnel also staged street crossings to ensure consistency among all sites as well as adequate sample sizes. The research team conducted a thorough analysis of motorist compliance. The findings were used to support development of the Guidelines for Pedestrian Crossing Treatments. The following conclusions were developed for motorist compliance:

* The crossing treatment does have an impact on motorist compliance. Treatments that showed a red indication to the motorist had a statistically significant, different compliance rate from devices that did not show a red indication. These red signal or beacon devices had compliance rates greater than 95 percent and included mid-block signals, half signals and HAWK signal beacons. Nearly all of the red signal or beacon treatments evaluated were used on busy, high-speed arterial streets. Pedestrian crossing flags and in-street crossing signs also were effective in prompting motorist yielding, achieving 65 and 87 percent compliance, respectively. However, most of these crossing treatments were installed on lower-volume, two-lane roadways.

* The measured motorist compliance for many crossing treatments varied considerably among sites. For example, treatments that were active when the pedestrian was present or would fit in an enhanced and/or high-visibility category had a wide range of compliance rates, as shown in Figure 1. In fact, a statistical analysis could find no significant differences between many of the crossing treatments, even though the difference in average compliance rates appeared practically significant (30 to 40 percent greater). The research team concluded that there were other factors (such as roadway width, speed limit and street environment) affecting compliance rates.

* The number of lanes being crossed influenced the effectiveness of the crossing treatment. All but one of the treatments on the two-lane roadways performed at better than a 75-percent compliance rate. On four-lane roadways, compliance ranged from below 30 percent to 100 percent.

* The posted speed limit influenced aie effectiveness of the crossing treatment. Flags, refuge islands and high-visibility markings all had higher compliance rates on lower-speed roadways. On a 35 mile-per-hour (mph), or 55 kilometer-per-hour (km/hr.), roadway, the best compliance rate observed for a treatment that was not showing a red indication to the motorist was approximately 58 percent. Compliance rates for the devices on 25 mph (40 km/hr.) streets all were above 60 percent. Compliance rates were as low as 15 percent for streets with a 35 mph (55 km/hr.) speed limit.

The research team recommended the addition of red signal or beacon devices to the engineer's toolbox for pedestrian crossings. The study results indicated that all red signal or beacon devices were effective at prompting high levels of motorist compliance on high-volume, high-speed streets. However, only a traffic signal is currently recognized in MUTCD, and the current pedestrian signal warrant is very difficult to meet.

In the current situation, engineers are unable to easily employ traffic control devices that appear most effective for pedestrians on wide, high-speed streets. The research team recommended the inclusion of a new type of highway traffic signal in MUTCD called "pedestrian beacon." Pedestrian beacons would have different signal operation modes than traditional traffic control signals and would include the red signal or beacon devices that this study found most effective on high-volume, high-speed roadways.


In the 2003 edition of MUTCD, the pedestrian warrant for a traffic control signal considers several factors in determining the need for a signal: pedestrian volume, gaps in vehicular traffic and walking speed (which may be used to reduce pedestrian volume). Previous studies have documented the difficulty in meeting this warrant at intersections with pedestrian crossing needs. In reviewing all traffic control signal warrants, the research team noted several inconsistencies between the pedestrian warrant and other vehicle-based warrants.

For example, the pedestrian warrant provides a single pedestrian volume criterion regardless of the major street vehicle volume being crossed. Other vehicle-based warrants provide a "sliding scale" such that fewer minor street vehicles are required as the major street vehicle volume increases. Additionally, other vehicle-based warrants permit a vehicle volume reduction to 70 percent when major street vehicle speeds exceed 40 mph (70 km/hr.).

The research team also conducted workshops to gather engineering judgment about proposed revisions to the pedestrian warrant. The findings from these efforts were used in developing the research team's proposed recommendations for a revised pedestrian warrant for traffic control signals.

The net effect of the proposed revisions is as follows: The pedestrian warrant will be slightly easier to meet with lower pedestrian volumes on streets with high vehicle volumes and slightly more difficult to meet on streets with low vehicle volumes.

In addition to traffic signal warrant revisions, the research team identified two other MUTCD sections that could be revised. In MUTCD section 4B.04, the research team suggested a minor addition to an enumerated list of alternatives to traffic control signals. The recommendation suggests median refuge islands or curb extensions as other alternatives to traffic control signals that could improve pedestrian safety.

The final recommended revision is the inclusion of a new type of highway traffic signal in MUTCD called "pedestrian beacon." The pedestrian beacon represents devices that this study found most effective on high-volume, high-speed roadways.

All three recommendations were presented to the Signal Technical Committee and then to the full NCUTCD. The full NCUTCD endorsed both the proposed changes to the signal warrant and the change to section 4B.04. More research was suggested before proceeding with a vote on the inclusion of a pedestrian beacon.



The research team developed guidelines that can be used to select pedestrian crossing treatments for unsignalized intersections and mid-block locations. Guidelines for Pedestrian Crossing Treatments is included in the research project's final report as Appendix A.6 Quantitative procedures in the guidelines use key input variables (such as pedestrian volume, street crossing width and traffic volume) to recommend one of four possible crossing treatment categories:

* Marked crosswalk

* Enhanced, high-visibility, or "active when present" traffic control device

* Red signal or beacon device

* Conventional traffic control signal

The guidelines include supporting information for these treatment categories as well as examples and pictures of traffic control devices in each treatment category. The supporting information for one of the crossing treatments is shown in Table 2. The summaries reflect the more common treatments currently being used and do not include every available treatment; summaries are based on observations of installed treatments and discussions with traffic engineers who have used or have considered using one or more of the components.

Several traffic engineers tested the Guidelines for Pedestrian Crossing Treatments and provided feedback that was incorporated into the current version. Additionally, the research team tested the guidelines using actual field data from field study sites as well as other marked crosswalks without treatments. The results of these tests indicated that the guidelines provide appropriate recommendations of pedestrian treatments that substantially agree with engineering judgment.


The Guidelines for Pedestrian Crossing Treatments are intended to provide general recommendations on pedestrian crossing treatments to consider at unsignalized intersections; in all cases, engineering judgment should be used in selecting a specific treatment for installation. The guidelines are built upon the recommendations of several studies and focused on unsignalized intersections. They do not apply to school crossings. Considerations (in addition to the procedure provided) should be employed at locations where the installation of a pedestrian treatment could present an increased safety risk to pedestrians, such as where there is poor sight distance, complex geometries, or in close proximity of traffic signals.

Note that the installation of a pedestrian crossing treatment alone does not necessarily result in more vehicles stopping for pedestrians unless that device shows a red indication to the motorist. Therefore, treating a location to improve pedestrian access or safety should include several components.

For example, in addition to traffic control devices such as signs or markings, geometric improvements (such as a refuge island, roadway narrowing, or curb extensions) may be used to shorten the crossing distance (and, hence, the exposure time for the pedestrian). Traffic calming may be used in the vicinity to slow vehicle speeds near the pedestrian crossing.

Example Using Guidelines

Citizens have requested a pedestrian treatment at the 2700 block crossing of Elm Street. Known characteristics of the site include the following:

* Four-lane road with no pedestrian refuge median

* 56 ft. (17 m) crossing distance

* 35 mph (55 km/hr.) speed limit

* During the peak pedestrian hour, 50 pedestrians counted when the major road volume was 1,000 vehicles per hour

* During the peak vehicle hour, 20 pedestrians counted when the major road volume was 1,500 vehicles per hour

* Motorists observed stopping for pedestrians, showing a "high" compliance

The following assumptions were made:

* Walking speed is 3.5 ft./sec. (1.1m/ sec.)

* Start-up time is 3 sec.


Figure 2 is a flowchart that provides an overview of the procedure. Following are the procedures for this example:

Step 1: Select worksheet. Worksheet 1 is the applicable worksheet for a speed limit of 35 mph (55 km/hr). Figure 3 shows the worksheet with appropriate values for the example.

Step 2: Check minimum pedestrian volume. The next step in the evaluation is to determine if a minimum number of pedestrians is present at the site. Because more than 20 pedestrians are crossing the roadway during the peak hour, some form of a pedestrian treatment is suggested.

Step 3: Check signal warrant. The minimum number of pedestrians needed on the minor road approach crossing a four-lane roadway with 1,000 vehicles in the peak hour is 271. The number of crossing pedestrians (50) is less than 271 ; therefore, a signal is not warranted under the pedestrian volume warrant. Checking the peak vehicle hour provides the same result: A signal is not warranted under the pedestrian volume warrant.

Step 4: Estimate approach pedestrian delay. The average pedestrian delay equation was used to determine the total pedestrian delay. A total pedestrian delay value of 9.8 pedestrian-hours was calculated.

Step 5: Select appropriate treatment. The motorist compliance observed at the site is "high." With a total pedestrian delay value of 9.8 pedestrian-hours and a motorist compliance of high, the worksheet indicates that an "enhanced/active" device should be considered.

As an alternative method to the worksheet calculations, the Guidelines for Pedestrian Crossing Treatments contain charts showing graphical illustrations for selected sets of conditions. Regions of the chart are shaded/color-coded to correspond to the appropriate treatment categories. These charts may also be used to verify the results obtained from the worksheet calculations.

Figure 4 shows the solution to the example problem using the major roadway volume of 1,000 vehicles per hour and the pedestrian volume of 50. The intersection of these two lines on the graph (see the circle in Figure 4) results in the same finding: "enhanced/active" device. Guidelines for Pedestrian Crossing Treatments includes specific suggested treatments within the enhanced/active category, such as in-street pedestrian crossing signs and pedestrian crossing flags.


A recent research project jointly sponsored by TCRP and NCHRP had two main objectives: 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 to recommend modifications to the MUTCD pedestrian traffic signal warrant.

The research team developed guidelines that can be used to select pedestrian crossing treatments for unsignalized intersections and mid-block locations (Guidelines for Pedestrian Crossing Treatments). Quantitative procedures in the guidelines use key input variables (such as pedestrian volume, street crossing width and traffic volume) to recommend possible crossing treatment categories. The research team developed and presented recommendations to revise MUTCD to NCUTCD.

The research team recommended that the Guidelines far Pedestrian Crossing Treatments be widely distributed. The audience and potential users of these guidelines include state, county, and city traffic engineers, transit agencies, roadway designers and urban planners, as well as consultants for these groups and agencies.


The final report, Improving Pedestrian Safety at Unsignalized Crossings (TCRP Report 112/NCHRP Report 562), can be purchased from the Transportation Research Board Bookstore at bookstore/ or can be downloaded from nchrp_rpt_562.pdf. The appendices, Improving Pedestrian Safety at Unsignalized Crossings: Appendices B to O (TCRP WebOnly Document 31 /NCHRP Web-Only Document 91) can be downloaded from tcrp_web_documents.

© 2007 Institute of Transportation Engineers Provided by ProQuest LLC. All Rights Reserved.

Source: Institute of Transportation Engineers. ITE Journal