National Academies Press: OpenBook

Improving Pedestrian Safety at Unsignalized Crossings (2006)

Chapter: Chapter 6 - Field Studies

« Previous: Chapter 5 - Findings From Surveys
Page 32
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 32
Page 33
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 33
Page 34
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 34
Page 35
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 35
Page 36
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 36
Page 37
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 37
Page 38
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 38
Page 39
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 39
Page 40
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 40
Page 41
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 41
Page 42
Suggested Citation:"Chapter 6 - Field Studies." National Academies of Sciences, Engineering, and Medicine. 2006. Improving Pedestrian Safety at Unsignalized Crossings. Washington, DC: The National Academies Press. doi: 10.17226/13962.
×
Page 42

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.

32 This chapter summarizes the field data collection approaches used in this project to evaluate pedestrian cross- ing treatments. Details on the study sites as well as the data collection techniques are provided below. Background A field study approach was developed to provide insight into the actual behavior of motorists and pedestrians at loca- tions with existing pedestrian crossing treatments. The specific measures of effectiveness (MOEs) used for the pedestrian crossing evaluations are listed in Table 15. Also, the research team collected data on site conditions at existing crossing treatment locations, which were used to help explain the vari- ation in MOE results for similar treatments at different loca- tions. Essentially, the team conducted observational and operational studies at existing crossing treatments with special consideration given to site conditions that ultimately influ- enced the effectiveness of crossing treatments. The primary MOEs focused on motorist and pedestrian behavior, conflicts, and delays at existing crossing treatment locations. The research team believed that this combination of behavioral and operational data analysis provided the best insight into the effectiveness of pedestrian crossing treat- ments. Table 15 summarizes the MOEs along with the method of calculation or the categories used to classify them. In addition to these MOEs, other crossing characteristics were desired in order to gain a thorough understanding of pedestrian movements at each site. Some of these character- istics included the gender of the pedestrian, the direction of the crossing movement, and the number of vehicles that did not stop when the treatment was activated. Site Selection The initial goal for the project was to collect data at 40 sites. Potential sites were identified during the project’s Phase I travels and contacts with cities, states, and transit agencies. States in this initial list were Texas, Utah, Washington, Ore- gon, California, and Arizona. Comprehensive evaluation data were collected at 40 sites, and video for an additional two sites was provided to the research team for analysis. Two of the three additional sites were in Maryland, while the remaining site was a midblock crosswalk in Arizona that had been selected for data collection by the research team. There- fore, 42 unique locations are represented in the evaluation dataset. The sites were selected to represent various treatment types and site conditions. Specific site selection was based on several factors so that the research team could obtain data across a representative range of treatment types, street envi- ronments, and traffic conditions. The primary factors were as follows: • Proximity to transit stop – sample of sites near or at a tran- sit stop, • Roadway type – moderate to high traffic volumes, • Proximity to driveways – locations where turning traffic conflicts from nearby driveways are nominal, • Area type – suburban and urban, and • Pedestrian age and ability – sample of sites with a range of pedestrian ages represented including the elderly and pedestrians with disabilities. Study Sites In total, 42 study sites were selected in seven different states (see Table 16). The study sites were chosen in an effort to dis- tribute the different types of crossing treatments in certain regions, such that the data for a particular treatment is not collected from a single city. This could not be avoided for two treatments (i.e., HAWK and in-street crossing sign) were each only installed in a single city. The sites were chosen to focus on arterial streets, with a range of operational and design C H A P T E R 6 Field Studies

characteristics (e.g., number of lanes, presence of median refuge island, and speed limit). Although not by design, 40 of the 42 study sites were in the western United States. However, the sites still included a wide range of climate and urban design features that were important to represent (e.g., snow- fall, cold winters, pedestrian-friendly versus less-than- friendly street design, and aggressive drivers). Descriptions of Crossing Treatments The research team categorized the crossing treatments into three basic types according to function and design: • Red signal or beacon – devices that display a circular red indication to motorists at the pedestrian crossing location. Examples (see Figure 12) include a midblock traffic signal, half signal, or HAWK signal beacon. • Active when present – devices that display a warning only when pedestrians are present or crossing the street.Examples (see Figure 13) include flashing amber beacons (both push- button and passive detection) and pedestrian crossing flags. • Enhanced and/or high visibility – devices and design treat- ments that enhance both the ability of pedestrians to cross the street and the visibility of the crossing location and pedestrians waiting to cross. Warning signs and markings in this category are present at the crossing location at all times. Examples (see Figure 14) include in-street pedes- trian crossing signs, high-visibility signs and markings, and median refuge islands. The treatment abbreviations as shown in subsequent tables and figures are as follows: • Half signals (Half); • HAWK signal beacon (HAWK); • Midblock pedestrian signal (Msig); • Smart pedestrian warning, where an overhead pedestrian sign and two yellow flashing beacons are passively activated by an approaching pedestrian (OfPa), • Overhead flashing beacons, where an overhead pedestrian sign and two yellow flashing beacons are activated when a button is pushed by the pedestrian (OfPb); • Pedestrian crossing flags (Flag); • High-visibility markings and signs (HiVi); • In-street pedestrian crossing sign (InSt); and • Pedestrian median refuge island (Refu). Figure 15 shows the number of sites in the seven states rep- resented in the study. Table 17 lists the 42 sites included in the study along with their characteristics. 33 MOE Description Categories or Method of Calculation Pedestrian visual search Percent of crossing events in which pedestrians use proper visual search 1. Looked in both directions 2. Looked in only one direction 3. Did not look in either direction Pedestrian crosswalk compliance Percent of crossing events in which pedestrians cross within 10 ft (3 m) of marked crosswalk 1. Crossed within 10 ft (3 m) of crosswalk 2. Crossed between 10 and 50 ft (3 and 15 m) from crosswalk 3. Crossed greater than 50 ft (15 m) from crosswalk Pedestrian activation Percent of crossing events in which pedestrians activate treatment (where applicable) 1. Activated treatment and waited for proper time to cross 2. Activated treatment but did not wait 3. Did not activate treatment 4. Treatment malfunctioned (no activation) Pedestrian- vehicle conflicts Percent of crossing events having a pedestrian-vehicle conflict 1. Conflict with first direction of major street vehicular traffic 2. Conflict with second direction of major street vehicular traffic 3. Conflict with left-turning side street traffic 4. Conflict with right-turning side street traffic Pedestrian delay Seconds of delay per pedestrian while waiting to cross Difference in time between these two events: 1. The pedestrian approached within 3 ft (0.9 m) of the crossing and indicated intent to cross the street. 2. The pedestrian began to cross the street. Pedestrian walking speed Average walking speed (ft/s) of pedestrian groups 1. Elderly and/or with physical disabilities 2. School children (ages 0-12) 3. Teenagers (ages 13-18) 4. Young adult (ages 19-30) 5. Middle aged (ages 31-60) 6. Older (age above 60 but not in Group #1) Motorist compliance Percent of motorists yielding or stopping for pedestrians Number of cars that stopped for (or yielded to) the staged pedestrian divided by the number of cars that should have stopped Table 15. Summary of pedestrian crossing treatment measures of effectiveness.

Protocol for Data Collection The data collection and analysis protocol for evaluating the effectiveness of treatments used two approaches for collect- ing the data: • General population pedestrians–use on-site and video- taped observations to record various pedestrian behavior and operational characteristics, and • Staged pedestrians–use staged pedestrians to measure motorist compliance at existing pedestrian crossing treat- ments. General Population Observation studies were used to record numerous pedes- trian behaviors and operational characteristics. A video recording was made of the crossing to permit review and data reduction after the actual crossing event occurred. It was necessary to observe actual pedestrian behavior (rather than simply using staged pedestrians) to measure a part of the crossing treatment’s effectiveness. In general, the following protocol was used in the observa- tion studies: • A minimum of 100 pedestrian crossing events or 4 hours of data (whichever occurred first) were recorded at each location, where each crossing event consisted of one or more pedestrians crossing the entire width of the street. • Two members of the project team were positioned at inconspicuous locations near the pedestrian crossing to make anecdotal notes of the crossing events. These anec- dotal notes did not include quantitative data on the MOEs in Table 15 but instead focused on qualitative observations about vehicle and motorist behavior. • The on-site field observers counted the number of pedes- trian crossing events as they occurred to ensure that the minimum sample size of 100 crossing events was achieved as time allowed. 34 City Number of Study Sites Range in Through Lanes Range in Speed Limit (mph) HAWK signal beacon 5 4 to 6 30 to 40 Tucson, AZ High-visibility markings and signs 2 4 4 6 4 4 4 25 to 35 Overhead flashing beacon (passive) 4 2 to 4 30 to 35 Los Angeles, CA Midblock signal 4 4 to 5 25 to 35 Santa Monica, CA Median refuge island, high- visibility signs 2 30 Capitol Heights, MD Overhead flashing beacon (continuous) 1 35 Towson, MD Overhead flashing beacon (pushbutton) 1 35 Half signal 3 4 35 Portland, OR Median refuge island, high- visibility signs 3 2 to 4 25 to 35 Austin, TX High-visibility signs and markings 1 35 College Station, TX Median refuge island, high- visibility signs 1 35 Overhead flashing beacon (pushbutton) 3 4 30 to 35 Salt Lake City, UT Pedestrian flags 3 4 to 6 30 to 35 Kirkland, WA Pedestrian flags 3 2 to 4 25 to 35 Redmond, WA In-street crossing sign 3 2 to 3 25 to 30 Seattle, WA Half signal 3 3 to 4 35 Crossing Treatment Table 16. Summary of study sites.

• The observers and the video recording devices were posi- tioned to be, as much as possible, inconspicuous to both the pedestrians and motorists. Staged Pedestrian Tests Staged pedestrian tests were used to measure motorist compliance at existing pedestrian crossing treatments. Staged pedestrians were used in the belief that consistent presenta- tion of a pedestrian intent to cross was critical for comparing motorist compliance results from different locations or areas of the country; in other words, pedestrian positioning, stance, and aggressiveness affect a motorist’s decision to stop or yield at a pedestrian crossing. For example, motorists are less likely to stop or yield when pedestrians stand several feet behind the curb line (e.g., the pedestrian may appear as though they are waiting instead of intending to cross). The following protocols were used in the staged pedestrian tests: • A minimum of 40 staged crossings (i.e., 20 crossings in each direction) were performed at each location. • The staged pedestrian was a male dressed in blue jeans and a neutral-colored shirt (e.g., gray, blue, tan, or white). 35 Midblock Traffic Signal Characteristics ■ A midblock signal is a standard traffic signal that is not located at an intersecting cross street. ■ The pedestrian phase for a midblock signal is typically activated by a pushbutton and can consist of a steady red indication or a sequence of steady red and then flashing red indications for drivers. ■ A midblock signal typically dwells in steady green (or green arrow) for vehicles. ■ A supplemental sign is typically used to indicate the signal is for pedestrians. ■ The signal is subject to requirements specified in the MUTCD. Half signal (Intersection Pedestrian Signal) Characteristics ■ A half signal is a standard traffic signal (with red, yellow, and green indications) for the major road. When located at an intersection, the minor cross street has Stop sign control. ■ The pedestrian phase for a half signal is typically activated by a pushbutton and consists of a steady red indication. ■ In the United States, most installed half signals dwell in steady green for vehicles, whereas most half signals in British Columbia dwell in flashing green. ■ This is an experimental traffic control device not currently included in the MUTCD. HAWK Signal Beacon Characteristics ■ A HAWK signal beacon resembles an emergency vehicle beacon and only provides yellow and red indications for drivers. Pedestrians see standard pedestrian control features. ■ The pedestrian phase for a HAWK signal beacon is typically activated by a pushbutton. Drivers see a sequence of flashing yellow, steady yellow, steady red, and flashing red indications. ■ The HAWK signal beacon, used exclusively in Tucson and Pima County, Arizona, dwells in a dark mode for vehicles. ■ This is an experimental traffic control device not currently included in the MUTCD. Stop sign on minor Signal heads on major approaches Figure 12. Red signal or beacon devices.

• The staged pedestrian approached the crossing, activated the crossing treatment (where applicable), and stood fac- ing oncoming traffic within 1 ft (0.3 m) of the curb line (even when parking or bike lanes are present). Where no curb was present, the staged pedestrian stood within 1 ft (0.3 m) of the outside edge of the curb lane. • The staged pedestrian could approach the crossing at any time when vehicles were within sight distance of the crossing. Vehicles that were too close to comfortably stop—estimated as being inside the stopping sight dis- tance per AASHTO Green Book—were not counted in the test (56). • Staged pedestrians avoided attempting to cross while other pedestrians were attempting to cross. • Motorists who did not stop were counted as not comply- ing. Motorists who slowed down without passing through the crosswalk to permit the staged pedestrian to safely cross were considered yielding vehicles. • For multi-lane approaches, the staged pedestrian took one or two steps into the street if the curb lane motorist stopped/yielded but a motorist in the inside lane was still approaching. • For divided roadways with a median refuge island, the staged pedestrian paused as necessary within 1 ft (0.3 m) of the island curb line (or inside lane line) before crossing the second direction of traffic. • Staged pedestrians aborted the crossing attempt after 1 minute if no vehicles stopped or yielded. • A second research team member and a video recording device were always present but inconspicuous to motorists during the staged pedestrian tests. Collection Approaches To obtain the general population and staged pedestrian data, the following data collection approaches were used: • Videocameras were used to provide a permanent record of pedestrian and motorist behavior. • Palmtop computers were used on site to record certain aspects of pedestrian and motorist behavior. • Site condition sheets were used to document geometric characteristics of each site. The specific protocol for each of these activities is described in more detail in the following sections. 36 Flashing Amber Beacons Characteristics ■ Overhead flashing amber beacons are mounted on mast arms that extend over the roadway at or in advance of the crossing location. Flashing amber beacons can also be mounted on signposts at the roadside. ■ The flashing beacons can be activated by either a pushbutton or a passive detection technology, such as bollards, video, or microwave sensors. ■ Continuously flashing beacons are not included in this category; they are included in the enhanced and/or high-visibility category. ■ This traffic control device is subject to requirements specified in the MUTCD. Pedestrian Crossing Flags Characteristics ■ Pedestrian crossing flags are square flags (of various colors, typically orange, yellow, or fluorescent yellow-green) mounted to a stick that is held by pedestrians waiting to cross or while crossing the street. ■ The flags are typically stored in sign-mounted holders on both sides of the street at crossing locations. ■ This is an experimental device not currently included in the MUTCD. Figure 13. “Active When Present” devices.

Video Data Collection Approach Originally the research team intended to collect the neces- sary data by videotaping the crossing movements at the vari- ous study sites. Using video data would provide a record of the events that took place during the study period. In the data reduction process, it was possible to review these events numerous times to consistently interpret and record needed information about pedestrian crossing events. The video recording of pedestrian activities primarily used one of TTI’s two camera trailers, which can raise a camera 30 ft (9 m) in the air to record a bird’s-eye view of the study area. The video trailer is normally outfitted with a single videocas- sette recorder (VCR) and a monitor, along with a hydraulic 37 In-Street Pedestrian Crossing Signs R1-6 Characteristics ■ In-street crossing signs are regulatory signs placed in the street (on lane edge lines and road centerlines, or in medians) to remind road users of laws regarding right-of-way at an unsignalized pedestrian crossing. ■ This traffic control device is subject to requirements specified in the MUTCD. High-Visibility Signs and Markings Characteristics ■ High-visibility signs and markings are warning devices placed at or in advance of the pedestrian crossing. ■ These include fluorescent-yellow-green pedestrian crossing signs, other pedestrian crossing signs, high-visibility crosswalk markings, and other devices that attempt to draw attention to the pedestrian crossing. ■ Many of these high-visibility signs and markings are included in the MUTCD and are subject to requirements specified in the MUTCD. Median Refuge Islands Characteristics ■ Median refuge islands are a design treatment that permits pedestrians to cross one direction of street traffic at a time. ■ Median refuge islands are typically raised above the roadway surface with an accessible pedestrian path, typically offset to direct the view of crossing pedestrians at the second direction of street traffic. ■ Two-way left-turn lanes and other median treatments that vehicles routinely enter are not considered appropriate refuge for pedestrians. ■ Curb extensions, roadway narrowing, raised crosswalks, and other design treatments or traffic- calming elements can also be used to improve the safety of unsignalized pedestrian crossings. R1-6a Figure 14. “Enhanced and/or High-Visibility” devices. 9 6 10 6 7 2 2 Figure 15. Location and number of study sites.

lift system on the roof of the trailer that raises an attached videocamera. For this project, a second VCR and camera were added to the trailer. One camera provided a wide-angle view of the area around the crossing under observation, while the second camera was zoomed in for a more detailed view of the crossing itself and the pedestrians using the crossing. The entire outfit in the trailer was powered by a portable genera- tor or, if available, a nearby fixed electrical outlet. A drawback of the video trailer was that it required the trailer to be driven to the study sites, and a place to park the trailer at each site had to be found. An issue at each study site was whether there would need to be a trade-off between the necessary viewing angle of the trailer and the need to be relatively inconspicuous. At selected sites, it was not possible to position the trailer in an inconspicuous location and still obtain an unobstructed view of the entire crossing and approaches. In these instances, supplemental battery-powered camcorders were used in conjunction with the video trailer to complete the necessary visual record. Observation Data Collection Approach Despite the advantages of video data collection, the use of only video data would have had some significant drawbacks. First, the amount of time necessary to pull all pedestrian and motorist characteristics from the video for each crossing would be immense. Although rewinding and fast-forwarding the video to review specific characteristics of crossing events would be possible, this would have to be done multiple times for each crossing. Second, the images obtained from the video 38 Si te N um be r Tr ea tm en t Sp ee d Li m it La ne s M ed ia n W id th (f t) C ur b to C u rb (f t) Pa rk in g La ne Bi cy cl e La ne C ur b Ex te ns io n N ea r Si de w al k Fa r Si de w al k D ist an ce to Tr an sit (f t) D ist an ce to Si gn a l (f t) M id bl oc k C ro ss in g AZ-TU-1 Hawk 40 6 20.00 100.17 N N N Y Y 45 1600 N AZ-TU-2 Hawk 35 6 20.00 103.50 N Y N Y Y 105 1255 N AZ-TU-3 Hawk 35 6 20.00 100.50 N Y N Y Y 125 1225 N AZ-TU-4 Hawk 35 4 13.00 72.00 N Y N Y Y 0 975 N AZ-TU-5 Hawk 35 4 12.00 63.75 N N N Y N 108 1274 N AZ-TU-6 HiVi 35 4 20.00 72.50 N N N Y Y 123 1280 N AZ-TU-7 HiVi 25 4 12.00 57.33 N N N Y Y 69 197 Y CA-LA-1 OfPa 35 4 11.00 75.00 Y N N Y Y 20 562 N CA-LA-2 OfPa 35 4 11.00 83.00 Y N N Y Y 325 684 N CA-LA-3 OfPa 30 2 12.50 60.50 Y N N Y Y 285 350 N CA-LA-4 Msig 25 4 10.00 70.00 Y N N Y Y 250 320 Y CA-LA-5 Msig 25 5 0.00 57.00 Y N N Y Y 90 295 Y CA-LA-6 Msig 35 4 0.00 60.00 Y N Y Y Y 200 360 Y CA-LA-7 Msig NP 4 10.00 80.00 N N N Y Y 60 850 Y CA-LA-8 OfPa 35 4 10.00 60.00 N N N Y Y 55 720 N CA-SM-2 Refu 30 4 6.00 66.00 Y N N Y Y >2600 420 N CA-SM-3 Refu 30 4 16.00 76.00 Y N N Y Y 20 370 N MD-P1 Refu 35 6 22.00 92.00 N N N Y Y 0 500 Y MD-TO-1 OfPb 35 4 0.00 46.00 N N N Y Y 400 950 N OR-PO-1 Refu 35 4 8.50 60.67 N Y Y Y Y 188 675 N OR-PO-2 Half 35 4 12.00 77.00 Y Y N Y Y 153 500 N OR-PO-3 Half 35 4 0.00 50.00 N N N Y Y 0 535 N OR-PO-4 Half 35 4 0.00 51.00 Y N N Y Y 0 780 N OR-PO-5 Refu 25 2 10.00 44.00 N N Y Y Y 0 785 N OR-PO-6 Refu 35 4 9.00 75.50 Y Y Y Y Y 110 1000 N TX-AU-1 HiVi 35 4 10.00 50.00 N N N Y Y 185 190 N TX-CS-1 Refu 35 4 10.00 54.50 N N N Y Y 1056 530 Y UT-SL-1 Flag 30 6 5.50 96.83 Y N Y Y Y 0 242 N UT-SL-2 OfPb 30 4 0.00 54.00 N N N Y Y 208 360 N UT-SL-3 OfPb 35 4 0.00 70.00 N N N Y Y 2600 800 N UT-SL-4 OfPb 35 4 0.00 56.75 N N N Y Y 0 1848 N UT-SL-5 Flag 35 4 9.08 68.00 N N N Y Y 71.5 1320 N UT-SL-6 Flag 30 4 12.33 83.33 Y N Y Y Y 600 600 N WA-KI-1 Flag 35 2 11.42 57.92 Y Y N Y Y 50 2640 N WA-KI-2 Flag 30 4 0.00 60.00 Y N Y Y Y 2600 215 N WA-KI-3 Flag 25 2 0.00 45.00 Y N Y Y Y 97 473 N WA-RD-1 InSt 25 2 0.00 38.25 N N N Y N >2600 2700 Y WA-RD-2 InSt 30 3 8.00 49.50 N N N Y N 2600 439 N WA-RD-3 InSt 30 2 10.50 43.50 N N N Y Y 234 345 N WA-SE-1 Half 35 4 0.00 54.50 Y N N Y Y 61 447 N WA-SE-2 Half NP 2 12.00 54.00 Y Y N Y Y 104 534 N WA-SE-3 Half 35 4 0.00 42.00 N N N Y Y 123 291 N Table 17. Study sites and their characteristics.

did not always provide the detail and resolution necessary to record certain characteristics. Gathering characteristics such as pedestrian age and gender would not be possible if video was the only data available for each crossing. Figures 16 and 17 illustrate video records with insufficient resolution; the crossing pedestrians are circled for emphasis. Because of these limitations, it was decided that members of the research team should make personal visual observa- tions on site during the intervals of video data collection. The first approach was to record these observations on a printed data sheet. However, initial tests revealed that, especially in cases of high pedestrian volumes, it was extremely difficult to write down the necessary information in a timely manner and still ensure the accuracy of the data. The next attempted method of collecting observation data was a personal digital assistant (PDA), also known as a palmtop computer. A program was written for the PDA to collect information on various types of pedestrian crossing characteristics and save them to a database. Through a series of drop-down menus and radio buttons, the observation data could be collected quickly and accurately. Initial tests of the PDA revealed that the PDA was more efficient, so much so that more items were added to the PDA program. One of the screens of the PDA program is shown in Figure 18. This screen was used to collect specific information on the pedestrian crossing. At the top of the screen, the observer would note the type of pedestrian crossing using the radio buttons at the top. In the two columns below, the observer could then record 10 characteristics about the pedes- trian and the crossing maneuver: crossing direction, lighting conditions, gender, age, activation conditions, looking behav- ior, start-of-crossing behavior, crosswalk compliance, the number of vehicles not stopping, and the distance between the pedestrian and the nearest vehicle that did stop or yield. Finally, there was a button for the observer to indicate whether the record was complete with all the items for the crossing recorded. Occasionally, especially during periods of high pedestrian traffic, the observer might be unable to record one or more items; noting “Incomplete” would be a signal to look for the unrecorded items on the video during the reduction process. Tapping the “Save” button would save the information for that crossing record into the database; tapping the “Back” button would return the observer to the previous screen. The primary benefit of using the PDA to collect this por- tion of the data was that a large amount of data could be stored and easily downloaded later. The information gathered on each screen was saved to a database file for more detailed calculations and analysis. The downloaded data was already formatted and ready for integration with the data collected from the video, which further improved the efficiency of the reduction process. Both video data and observation data were post-processed manually to determine pedestrian volumes, pedestrian gap acceptance levels, pedestrian delay threshold levels, and behavior as pedestrians wait for an adequately sized gap. Site Characteristics Data Collection Approach In addition to motorist and pedestrian behavior, the research team also collected data on the characteristics of the study site that included roadway, overall location, and pedes- trian characteristics. These site characteristics data would be used for explaining the variation in treatment effectiveness at different locations (e.g., were posted speed limit and average daily traffic volume strong predictors of treatment effective- ness within a particular group of treatments?). Most of these 39 Figure 16. Video record of pedestrians crossing in twilight. Figure 17. Video record of pedestrian partially obscured by background clutter.

site characteristics were recorded by hand on a pre-printed data sheet and supplemented by a hand-drawn sketch of the geometric configuration of the site. The sketch contained key geometric dimensions of the study site, which were recorded by hand measurements, as shown in Figure 19. The remain- ing site characteristics were collected from the video. Finally, observers also took multiple pictures of each site using a digital camera. These pictures illustrated the various approaches to each crosswalk, relevant traffic control devices, other conditions at the site, and any unusual characteristics that might have been present. The pictures supplemented the sketch and the recorded video for use in reviewing character- istics of each site. Data Reduction Both video and palmtop computer data were post- processed manually to determine pedestrian behavior at each site. PDA data were downloaded and stored in a database file at the conclusion of the study period for each site. These data- base files were converted to spreadsheet files to expedite cal- culations for various emphases in the data reduction: motorist behavior, conflicts, pedestrian gap acceptance and crossing times, vehicle counts, and group/cluster information. The group/cluster information was simply copied from the handwritten sheet completed on site to the spreadsheet file 40 Individual Group Cluster Direction W/SB E/NB Light Day Twilight Nite Gender M F ? Age Child (0-12) Elderly/Disabled Teen (13-18) Y. Adult (19-30) Middle(31-60) Over 60 Activate Act, Waited Act, No Wait No Act, Wait No Act, No Wait Malfunction Look Both ways One way No ? not sure Start Normal Hesitate Abort Wait in roadway Compliance <10 ft 10-50 ft >50 ft (Jaywalk) #Cars Not Stop [Keypad] Dist to Stop <5 ft 5 to 30 ft >30 Complete Incomplete Save Back Figure 18. Screen #3 of PDA program. Figure 19. Measuring dimensions at a study site.

for storage. The vehicle counts were completed by watching the videotapes and counting the vehicles traveling through the study site. These counts were divided into 5-minute peri- ods in each direction of vehicle travel. The remaining items were reduced with a combination of the PDA data and the video recordings. In the spreadsheet file, worksheets were created for manu- ally recording motorist behavior, conflicts, and gap and cross- ing data. Technicians would then review the video for a site and record the pertinent information for each crossing event, correlating each event to a recorded event in the PDA data when possible. At each site, however, there were crossing events not recorded in the PDA on site; these events were added to the spreadsheet. The motorist behavior information collected for a particular crossing event included the PDA record number (if applicable), the crossing number (a count of both PDA records and events not in the PDA), the number of pedestrians (if a group or clus- ter), the motorist compliance behavior in both directions of travel, the stopping distance for vehicles in both directions of travel, and comments about the behavior of the pedestrian. Conflict information was a determination of whether a conflict occurred during the crossing event and, if so, what type of conflict it was. Conflicts were defined by one of four general categories, as shown in Figure 20. Most crossing events occurred without conflicts. Reducing gap and crossing data involved observing each crossing event with respect to the time the pedestrian arrived at certain points along the crossing route. The crossing behav- ior was also recorded, as were any comments by the observer. Five distinct times were recorded for each pedestrian, as defined in Figure 21. For vehicles that passed through the site during a crossing event, their time of arrival and travel lane were recorded. Summary The research used observational studies of motorist and pedestrian behavior to evaluate the effectiveness of pedestrian crossing treatments at 42 sites in seven states. Several measures of effectiveness were used as surrogates for safety perform- ance, because the timing and duration of the study did not 41 A) Dart/Dash The pedestrian walks into the roadway in front of an approaching vehicle, requiring that vehicle to make an avoidance maneuver (e.g., severe braking or sudden lane change). B) Multiple Threat The pedestrian enters the traffic lane in front of stopped traffic and conflicts with a vehicle traveling in the same direction as the stopped vehicle. C) Walking Along Roadway The pedestrian walks along the roadway looking for a gap through which to cross and steps into the path of a vehicle. D) Other The pedestrian’s path conflicts with a vehicle’s path and cannot be classified by any of the other scenarios. Figure 20. Definition of conflict categories (Adapted from Pedestrian Facilities User Guide–Providing Safety and Mobility [27]).

permit the collection of before-and-after pedestrian crash data at several promising study sites. Several criteria were used to select the study sites, chief among them: presence of a marked crosswalk, pedestrian activity, proximity to transit stops, and high-volume, high-speed streets. The study sites were grouped into three categories according to function and design: red signal or beacon devices (e.g., half signals and HAWK); “active when present” warning devices (e.g., flashing beacons or crossing flags); and enhanced and/or high-visibility signs and markings. Videocameras were placed at inconspicuous locations to provide a permanent record of pedestrian and motorist behavior at each study site. Members of the research team also staged crossings at each site to provide a consistent reference point for comparison among all sites. Palmtop com- puters were used on site to record certain pedestrian and motorist behaviors not easily extracted from video. The research team also gathered data about site conditions (e.g., geometry and dimensions in vicinity of crossing). These field studies provided a comprehensive, multi-faceted dataset that permitted a wide variety of analyses. 42 KEY: A = pedestrian approaches curb, indicates interest to cross B = pedestrian steps off of near side curb C* = pedestrian reaches middle of median D* = pedestrian departs middle of median E = pedestrian steps onto far side curb *C and D are omitted for sites with no median. C A B D E Figure 21. Crossing diagram.

Next: Chapter 7 - Findings From the Field Study »
Improving Pedestrian Safety at Unsignalized Crossings Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

TRB's Transit Cooperative Research Program (TCRP) and National Cooperative Highway Research Program have jointly produced and published Improving Pedestrian Safety at Unsignalized Crossings. The product, which can be referred to as TCRP Report 112 or NCHRP Report 562, examines selected engineering treatments to improve safety for pedestrians crossing high-volume and high-speed roadways at unsignalized locations. The report presents the edited final report and Appendix A. TCRP Web-Only Document 30/NCHRP Web-Only Document 91 (Pedestrian Safety at Unsignalized Crossings: Appendices B to O) contains the remaining appendixes of the contractor's final report.

A summary of TCRP Report 112/NCHRP Report 562 as published in the July-August 2007 issue of the TR News is available online.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!