Hazardous Road Location Identification and Prioritization by means of Segmentation and Data Envelopment Analysis

Systematic reduction of accidents needs a comprehensive road safety management. The determination of black spots referred to as hazardous road location (HRL) is the first step for the highway safety management process. Identification of the HRLs needs to quantification of the hazard measures. Employing simple measures such as frequency, rate and cost of accidents may cause many errors because of random changes in accident, annually. Another method to identify HRLs is use of regression models applied for computing of accident frequency and their severities in different temporal and special patterns. The requisite of regression models is a statistic function calibrated by explanatory and dependent variables, also special assumptions have to be considered for data distribution and model limitation. The aim of the present study is representation of a novel method to pronounce accidents with regard to traffic, geometric and environmental circumstances of road which can consider the interaction of accidents as well as their casual factors. This approach segments the road to homogeneous sections; therefore, decisions about road safety condition would be made for a length of road with the specified parameters instead of a spot which is called accident prone section (APS). This method has been done with data envelopment analysis technique. Proper rating can be possible by obtaining inefficiency values from this method for road sections. A case study has been done in 144.4 km which resulted in identifying 154 sections with different relative hazard scores. Accordingly APSs have been identified and prioritized with the proposed method which is a new experience based upon the definitions of inputs and outputs according to data envelopment analysis.