Accident prediction models for signalized crosswalks.

In Poland, there is an obvious need for effective countermeasures to improve traffic safety, especially pedestrians' safety. This paper presents the results of a study into pedestrian safety on signalized crosswalks. Firstly, a regression technique was used in the investigation into accident-related factors. Secondly, the accident predictive models have been developed. The models have been used to highlight, and quantify the effects of traffic and signals impacts on the pedestrian safety. The study shows that short-term measurements carried out specifically for a safety analysis or conducted by traffic authorities on a regular basis provide sufficient data for developing accident regression models. This paper proposes the accident prediction method applicable in non-stationary conditions, as faced in Poland and other East European countries.