Multi-Vehicle Crash Severity Prediction Model in Different Weather Conditions (Case Study of the East Azerbaijan Roads)

Please enable javascript in your browser.

Yazdani, Mirbahador | 2015

843 Viewed

Type of Document: M.Sc. Thesis
Language: Farsi
Document No: 46888 (09)
University: Sharif University of Technology
Department: Civil Engineering
Advisor(s): Nassiri, Habibollah
Abstract:
Traffic accidents, is an inevitable event. There are various reasons for accidents and almost not possible to identify all of them, but a careful study of these phenomena can identify the main causes of this phenomenon. This study estimates Multi-Vehicle severity in terms of different weather. Accidents data the Including 2,949 accidents recorded during 1388 to 1392 in six axes of the East Azerbaijan roads in three levels of accident severity: damage, injury and death. For modeling, sequential Logit model was selected because accidents severity have Sequential nature. This model used the Stata software, version 12, for evaluation and sensitivity analysis. The obtained model consists of independent variables temperature below 7° C, temperature above 19° C, rainfall exceeds 5 mm, working and non-working day, gender, road machinery, autumn and winter, all of them significant amounts of less than 0.05. The model also shows that the possibility of damage accident is 77.5 percent, injury 18.9 percent and death 3.6 percent
Keywords:
Crash Severity ; Ordered Logit Model ; Different Climate ; Multivehicle Accidents

No TOC