Long downgrade road sections are dangerous parts for heavy-duty truck accidents. Most of current researches on heavy-duty truck accidents are focused on the speeding of truck, which ignore other factors (weather, vehicle types, driver attribute, etc.). Through analysis of accident data in long downgrade sections of Luofu Freeway in Yunnan Province, the effects of weather, vehicle types, driver attribute, etc. on heavy-duty truck accidents are studied in this paper. First, descriptive statistical analysis method is used to study the accident distribution of different factors. Then using K-mean clustering method, the factors are divided into three clusters based on their effects on accident severity (accident number, occupant injuries, and deaths). The study results show that non-local trucks, 6-axle trucks and 00:00-07:59 are the most important factors on accident severity in long downgrade road sections; fog, rain and low temperature have little effects on accidents in long downgrade road sections. The study results may be used as a reference basis to improve heavy-duty truck transportation safety in long downgrade road sections.

traffic accident  /  heavy-duty truck  /  long downgrade  /  clustering analysis