There is an established consensus among traffic safety researchers that a nonlinear relationship exists between traffic exposure and safety. This relationship is reflected by the safety performance functions (SPFs) calibrated for various classes of roads and intersections. One of the main uses of SPFs is to identify locations with potential for accident reduction. While this application is certainly important, the use of SPFs provides no information related to the nature of the accident occurrence. Without being able to relate accident frequency and severity to roadway geometrics, traffic control devices, roadside features, roadway condition, driver behavior, or vehicle type, it is not possible to develop effective countermeasures. A methodology was developed to provide guidance in diagnostics of safety problems, recognition of accident patterns, and development of appropriate countermeasures. Considering that traffic accidents can be viewed as random Bernoulli trials, it is possible to detect deviation from the statistical process by computing observed cumulative probability for each of the accident characteristics. Detection of an accident pattern at an intersection suggests the presence of an element in the roadway environment that triggered a deviation from a random statistical process in the direction of reduced safety. Identification of such an element always provides a critical clue to accident causality.
Abstract