U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590
Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations
|This report is an archived publication and may contain dated technical, contact, and link information|
Publication Number: FHWA-HRT-06-125
Date: November 2006
Pedestrian and Bicyclist Intersection Safety Indices
PDF Version (2.25 MB)
PDF files can be viewed with the Acrobat® Reader®
CHAPTER 7. CONCLUSIONS AND DISCUSSION
The pedestrian and bicycle intersection safety indices developed in this study are intended to prioritize intersection crossings (Ped ISI) or intersection approaches (Bike ISI) according to the relative level of safety for pedestrians or bicyclists given macro-level site characteristics. The analysis incorporated behavioral data in the form of conflicts and avoidance maneuvers and subjective data in the form of expert safety ratings. The final models are shown below in Table 28 and Table 29. For an explanation of the variables, see Table 17 on page 35 and Table 22 on page 39.
Table 28. Bicycle intersection safety index (Bike ISI).
Table 29. Pedestrian intersection safety index (Ped ISI).
APPLICATION OF THE PED ISI AND BIKE ISI
The Ped ISI and Bike ISI are intended to be used to give relative rankings of intersections according to pedestrian and bicyclist safety. The intent of this tool is not to dictate a predetermined index value that would warrant safety improvements. Rather, the Ped ISI and Bike ISI provide the practitioner with a way of prioritizing a group of intersections according to the relative likelihood of safety for pedestrians and bicyclists. This prioritization approach will allow practitioners to target the most hazardous sites, but also work within the confines of budgetary restrictions.
The authors envision practitioners using the Ped ISI and Bike ISI to evaluate each approach or pedestrian crossing at all intersections in their jurisdiction or a select group of intersections. The tool and accompanying instructions is laid out in an easy-to-use format in the accompanying User Guide, which provides several real-world examples and Quick Reference Tables for safety index values. Once safety index values are assigned to each site, the practitioner would then select the sites with the highest index values and conduct more detailed reviews of those sites (using other tools and methods) to determine whether any geometric or traffic control treatments are needed to improve the safety of the intersection. The User Guide recommends resources such as PEDSAFE and BIKESAFE to aid with countermeasure selection (Harkey and Zegeer, 2004; Hunter et al., 2005).
GEOGRAPHICAL RELEVANCE OF THE MODELS
The Ped ISI and Bike ISI were each developed based on sites selected from three cities representing three geographic areas, including West Coast (Oregon and California), Northeast (Philadelphia), and Southeast (Florida). Data were not collected in other States or regions of the United States, since the scope and resources for this study were limited because of the large amount of data collected at each site. Also, sites were selected to represent some of the more common characteristics of intersections, and it was not practical to include sites covering all possible site conditions; State practices; regions of the United States; or demographics of drivers, pedestrians, and bicyclists. However, this study does include consideration of hundreds of hours of video data collection at approximately 150 intersections in several States, as well as intersections ratings by pedestrian and bicycle professionals throughout the United States. It is expected that pedestrian and bicycle safety information from future studies might be useful to refine the models and index procedure developed herein.
LIMITATIONS OF THE RESEARCH
Although the sites used in this study varied in their geometric and traffic characteristics, there is concern that the site selection did not include the most hazardous intersections in the study cities. The results of the safety survey seem to indicate that the sites in the study did not cause the evaluators to use the full range of the 6-point scale (very few 5’s and 6’s). This result is probably a result of the site selection process. This study depended on finding sites with at least moderate amounts of existing pedestrian or bicyclist traffic in order to collect sufficient conflict and avoidance maneuver data. In general, users are more likely to choose easier, safer sites to walk or bicycle rather than difficult ones, and, therefore, it was difficult to find high-hazard sites (i.e., ratings of 5 or 6) that also carried many travelers on foot or bike. The development of the 6-point scale in this study still allows for those sites with higher hazard levels (5’s and 6’s) to be found and rated when this safety rating is applied to urban and suburban intersections. Additionally, since the avoidance maneuvers, conflicts, and safety ratings were all collected during daylight, the Ped ISI and Bike ISI may not accurately identify sites that would be particularly hazardous at night.
Once pedestrian crossings and bicycle approaches to intersections have been prioritized for safety improvements, the practitioner will have many options of analysis and treatment. The authors recommend PEDSAFE and BIKESAFE as excellent tools to assist in the selection of appropriate countermeasures. PEDSAFE is available from FHWA (Harkey and Zegeer, 2004). The online version can be accessed at www.walkinginfo.org/pedsafe. BIKESAFE is in its final stages of review and is due to be released in 2006 (Hunter, et al., 2005). The online version can be accessed at www.bicyclinginfo.org/bikesafe.
PEDSAFE and BIKESAFE are designed to recommend treatments for specific safety problems. In order to make full use of the information provided in these tools, the practitioner will need to gather knowledge of the most common safety problems at each site to be addressed. This step can be done through examining the types of crashes that occur at the site or through observational analysis of pedestrians, bicyclists, and motorists at the site.
The PEDSAFE Guide provides details on 49 different types of safety treatments that can be used to improve pedestrian safety and/or mobility. This Guide also includes information on the specific types of countermeasures that may be appropriate for addressing such objectives as:
A listing of pedestrian-related treatments for each of these eight performance objectives is given in Figure 13 and Figure 14 by "categories" of treatments, including pedestrian facility design, roadway design, intersection design, traffic calming, traffic management, and signals and signs. For example, to reduce the speed of motor vehicles, some of the possible roadway design treatments include adding bike lane or shoulder, road narrowing, reducing the number of lanes, driveway improvements, curb radius reduction, or adding a right-turn slip lane.
The PEDSAFE Guide also gives a description of 12 specific pedestrian crash types (e.g., dart/dash, walking along roadway, turning vehicle, multiple-threat), with corresponding countermeasure options for each crash type. The Guide also contains write-ups for 71 case studies of pedestrian improvements that have been implemented in the United States. Also, the expert system software is provided to allow a user to input the type of pedestrian safety problem, along with the location or roadway section characteristics, such as intersection or midblock, type of control devices (e.g., traffic signal, stop sign, no control), number of lanes, and traffic volume. The software then will generate a "short list" of countermeasure options based on the type of pedestrian safety problem and site characteristics.
The BIKESAFE Guide also gives similar types of information on countermeasures for bikerelated crashes. For example, countermeasure options are given for the following objectives:
There are nine categories of bicycle treatments given in Figure 15 and Figure 16, including those involving shared-roadway treatments; on-road bike facilities; intersection treatments; maintenance measures; traffic calming; trails/mixed-use paths; markings, signs, and signals; education and enforcement; and support facilities and programs. For example, potential measures to improve bike safety at intersections include curb radii revisions, roundabouts, intersection markings, sight-distance improvements, turning restrictions, and the redesign of the bike/motor vehicle merge area. BIKESAFE also provides a matrix of potential bike safety treatments that correspond to 13 different types of bicycle crashes.
The BIKESAFE Guide also provides details of more than 50 case studies from the United States and abroad related to past safety improvements. As with PEDSAFE, the BIKESAFE Guide includes a CD-ROM that allows an engineer, planner, or other safety professional to enter the basic crash or information or performance objectives for a location or section, along with site characteristics. The expert system software will then give a short list of candidate countermeasures that are appropriate for those conditions.
Figure 13. Matrix of pedestrian safety countermeasures associated with various objectives.
Figure 14. Matrix of pedestrian safety countermeasures associated with various objectives (continued).
Figure 15. Matrix of bicyclist safety countermeasures associated with various objectives.
Figure 16. Matrix of bicyclist safety countermeasures associated with various objectives (continued).
RECOMMENDATIONS FOR FUTURE RESEARCH
Expansion of Scope
As discussed, the safety ratings did not extend to the full range of possible ratings (i.e., very few 5’s and 6’s). This was due to the fact that the study sites were selected to include those locations with high volumes of pedestrian and bicyclist activity. These high volumes were necessary for the collection of behavioral data. However, obtaining expert safety ratings is not dependent on having high volumes of pedestrian or bicyclist activity. Future research that uses expert safety ratings may consider including sites that would be considered more hazardous (e.g., heavier and faster traffic, fewer pedestrian or bicycle facilities, etc.). This type of study design could yield a model that would give prioritization of a wider range of intersection types.
The Ped ISI and Bike ISI would benefit from a large-scale field validation effort in one or more cities. The intended field validation would consist of selecting a group of intersections, independently rating them with the safety index tool and ped/bike safety experts, and comparing the two ratings. The effort could also compare the ratings with safety data, such as crashes and conflicts. Probable outcomes of this procedure would be a validation of the type and magnitude of the variables in the safety index models, as well as possible modifications to the models based on feedback from the safety experts.
The models developed in this study should be considered for future validation with more extensive pedestrian and bicyclist crash-based models. Specifically, as future studies are able to better quantify pedestrian and bicyclist crash effects of various intersection features, such information should be used to modify the safety index models accordingly. The inclusion of a greater number of sites may lead to a more sensitive model that would reflect the effects of smaller factors, such as median type and width.