Multilevel Analysis of Road Accident Frequency: The Impact of the Road Category

Abstract

This work attempts to develop a universal model for predicting micro-and macro-level accident frequency. The study implies that the number of accidents may vary depending on the type of roadway and the characteristics of its small segments, both at the group and individual levels. A multilevel model has been designed to address the nested link between individual road segments and multiple road classifications. The multilevel analysis allows one to investigate the hierarchical nature of road accident parameters at the micro and macro levels to comprehend the risk of specific road segments within distinct roadway categories. To accomplish this, a case study of fifty-seven roads has been selected, covering five main categories: motorway, expressway, primary arterial, secondary main roads, and minor roads. In addition, each roadway has been subdivided into several flexible-length segments, each of which is identified by its traffic and geometrical characteristics. The outcome confirms that the accident frequency per road segment is more likely to vary between roads with distinct categories. Moreover, road types with low design requirements are more hazardous than well-designed roads, even with lower speed restrictions and traffic numbers.