The Pennsylvania State University
University of Maryland
University of Virginia
Virginia Polytechnic Institute and State University
West Virginia University

LTI HOME

A Safety Evaluation of Photo-Red Enforcement Programs in Virginia (Full Text)

Authors: Rahul Khandelwal and Dr. Nicholas J. Garber

Abstract

A photo-red enforcement system entails the use of cameras that photograph vehicles entering an intersection after the signal has turned red; citations are then mailed to the vehicle’s registered owner. The purpose of the research was to identify the safety impacts of photo-red enforcement programs in Virginia.

An empirical Bayes approach was used to examine the impact of the program on crashes while controlling for mainline traffic volume, yellow interval, truck percentage, number of lanes, and speed limit. The use of the cameras were correlated with decreased red light running crashes (25% to 34%), increased rear-end crashes (45% to 65%), increased total crashes (5% to 13%), decreased injury crashes attributable to red light running (23% to 34%), and increased total injury crashes (4% to 20%).

Analysis of variance (ANOVA) and generalized linear models (GLMs) were used to control for confounding factors (such as average daily traffic, the yellow interval, and intersection geometry) and to pinpoint locations where use of photo-red enforcement can have a positive safety effect. ANOVA was used as an innovative screening tool to delineate the factors (including second order interaction terms) that potentially affect the crash frequency, and GLMs were used to quantify how these factors affect the crash frequency. The analysis illustrates the utility of selecting largest and most heterogeneous group of sites possible subject to the constraints 1) the geometric characteristics can be explicitly modeled and 2) the sites are homogenous in all other aspects not included in the model. Such sites can only be identified by detailed manual examination. The results suggest that photo red enforcement may have a positive impact on safety at intersections where the yellow interval is excessively higher than that recommended by ITE standards.

The crash results presented herein suggest that Virginia’s program will realize a net safety gain if the severity of the eliminated red light running crashes is substantially greater than the severity of the induced rear end crashes. A detailed study of injury severity, therefore, is needed to determine if the cameras have a net safety benefit.