The conclusion of this study was that RLCs are not an effective countermeasure and that they may increase the number of rear end crashes.
A Long Term Study of Red Light Cameras and Accidents
David Andreassen
Australian Road Research Board
February, 1995
Summary
This study has examined the long term effect on accident-types of redlight cameras (RLC) at 41 signalised intersections in Melbourne. The RLC were installed in 1984, and reported accidents for the period 1979 to 1989 were used in the detailed analysis.
The analysis was addressed in several ways. The first was a grouped analysis taking the predominant accident-types for all the RLC sites taken together and comparing the changes over time with the changes in the same accident-types in Metro Melbourne, in the rest of the State, and at signalized intersections in Melbourne. The second was to separately examine each accident-type for the 41 sites and look for changes over the whole period. The third was to classify the accidents at individual RLC sites according to whether it involved the approach on which the camera was installed. The fourth was to consider the frequency of each accident-type before the RLC installation and stratify the frequencies to ascertain if there was any difference in effect by initial frequency. The fifth was by considering both the camera approach and initial frequency. The sixth was to compare the changes at the RLC sites with changes in accidents at signalized intersections.
The original choice of the RLC sites must be questioned. Three-quarters of the sites had initial annual frequencies of two or less reported “adjacent approaches” accidents. Low frequency sites are not good candidates for testing the effectiveness of accident countermeasures.
The results of this study suggest that the installation of the RLC at these sites did not provide any reduction in accidents, rather there has been increases in rear end and adjacent approaches accidents on a before and after basis and also by comparison with the changes in accidents at intersection signals.
There has been no demonstrated value of the RLC as an effective countermeasure.
To download the full copy of this study, click here.