Identification of effective factors on severity of road crashes and prioritization of countermeasures by Structural Equation Modeling (SEM) and Fuzzy Inference System (FIS)

Population growth, increased travel demand and, consequently, increased motor vehicle use has led to concerns about road safety in today’s society. Various factors such as driver characteristics, environmental conditions, road surface conditions, and type of vehicles affect the severity of accidents. In this paper, by using structural equation modeling (SEM) method and relying on crashes data, it has been tried to identify the factors that have the greatest effect on the severity of suburban accidents in Kerman province. Then, the results obtained from structural modeling are used to develop a fuzzy inference system (FIS) in order to prioritize solutions in relation to each factor. According to the results obtained from the structural equation modeling, human factor (estimate -0.47) has shown the greatest effect on occurence of accidents with high injury compared to other latent factors in the model. Among the parameters explaining the human factor, using the safety belt variable (estimate 0.85) was the stronger predictor for this factor. Finally, in order to determine the priority of implementing strategies to reduce the severity of crashes by human, environmental, and road factors, the fuzzy inference system has been used due to its acceptable accuracy in solving problems that face uncertainty. The results of the outputs of the fuzzy inference system have been shown that the priority of solutions related to human factors is far more than solutions related to road and environmental factors. Overall, the results of this study can help transportation professionals to prioritize countermeasures to reduce the severity of crashes on the suburban roads in an efficient and optimal manner.