Identification and Prioritization of Accident Prone Segments based on Wavelet Theory and Cause-oriented Method

Accident prone segments have a significant role in the occurrence and the numbers of road accidents. And they impose negatively social and environmental effects on the performance of transport system. Thus, identification and prioritization these segments play positively a role in reducing accidents, costs, and improvement of safety level on roads. Due to the importance of determining the accident prone segments, the aim of this study is to use dynamic segmentation and prioritization methods including wavelet theory and cause-oriented methods. Therefore, results from the segmentation and wavelet signal theory on the Kermanshah-West Islamabad Road indicated that accident prone segments are classified as main and local segments. Then, the cause-oriented prioritization method based on the analytical hierarchy process method for main and local segments showed that the S2 section, which ranges from about 2.5 km to 11.5 km from the beginning of the road is in the higher priority. However, S3 section which ranges from about 7.5 km to 10.5 km from the beginning of the road is the lower priority of the improvement of the road safety. In the future, this paper may help researchers in order to examine combination of arithmetic’s functions with artificial intelligence methods, logical reason methods, and SVM (Support Vector Machines) algorithm for segmentation accident prone segments as dynamic segmentation methods.