Accidents are considered as one of the leading causes of death around the world. Investigating the causes of various accidents, identifying the primary factors and accident blackspots can be used to prevent severe injuries and the associated costs and casualties. Many studies are conducted around the world to investigate the impacts of accidents and to identify the main factors. This study investigates the accidents and the severity of their injuries at intersections and squares. Accident data related to Zanjan is used to assess the severity of accidents at urban squares and intersections, which are modeled by SAS software. For this purpose, two models of OP and cumulative Probit are used. The results show that the cumulative probit model is better able to interpret the dependent variable based on the independent variables than the OP model.
Based on the overall results of the models, it can be seen that the occurrence of an accident in daytime compared to nighttime, on holidays compared to non-holidays and in winter compared to other seasons, reduces the injury severity. Heavy vehicles reduce the injury severity compared to cars, while two-wheeled vehicles, including motorcycles and bicycles, increase the injury severity. Accidents at intersections also increase injury severity.

injury severity, intersections, squares, Ordered Probit model