Investigators:
Dr. Michael Hunter, Associate Professor, Georgia Tech
Dr. Michael Rodgers, Principal Research Scientist, Georgia Tech
Project Overview:
This research is evaluating safety treatments proposed at two high speed rural intersections with high numbers of crashes and fatalities. The full evaluation period for this study will be approximately 6 months prior to the installation of the safety treatments and at least 3 years after the implementation of the proposed non-standard striping at the two intersections and also at various other control intersections. This safety study will include a traditional analysis of incident data prior to and approximately 3 years after the implementation of safety treatments. To complement the traditional approach to evaluating safety treatments this study will largely focus on evaluation of surrogate measures that will be available throughout the project time frame. This will allow for interim results and measures of effectiveness for the installed treatments in a shorter time frame. This approach offers the potential for adjustments or expansion of the mitigation measures prior to the end of the full 3-year evaluation period.
This research study will consider both crash data and surrogate safety measures. A surrogate safety measure is an indirect measure of safety that attempts to assess the safety of a facility through means other than crash data. For example, conflicts (i.e. near misses, erratic maneuvers, and evasive actions), speed, and available gaps are potential surrogate safety measures. Surrogate safety measures are often useful when crash data is limited and/or there is a desire to assess safety in a shorter time frame, or where there may exist unique facility or operational characteristics impacting facility safety.
The initial findings of the project indicate that surrogate measures represent a potentially useful means for rapid evaluation of safety treatments, although potentially significant effort and resources may be required to gather surrogate measure information, depending on the selected surrogate measure. In addition, significant uncertainty exists in the application of surrogate safety measure analysis to actual field conditions. Often the analyst must interpret the results of a surrogate safety study with limited guidance on the implications of the measured surrogate measure values. In an attempt to address these concerns initial project findings seem to support Post Encroachment Time (PET) as a simple and cost efficient surrogate measure. Thus this project is also exploring the general use of PET as a surrogate safety measure. The collection of surrogate data over a wide range of high, median, and low incident rate signalized and unsignalized intersections will be conducted. Given the known safety performance of the intersections based on crash records, this data to will then be used to determine the most effective form of the surrogate measure for identifying the safety performance. The determined surrogate analysis may then be applied to studies of future treatments with significantly increased confidence.
Sponsor: Georgia Department of Transportation