Development of Accident Modification Factors in Two-Lane Highways

Base model and accident modification factors play an important role in the accident predictive algorithm for two-lane highways used in the interactive highway safety design model (IHSDM). One of the main approaches of obtaining the accident modification factor associated with each geometric or traffic feature of the highway is the utilization of before and after studies. If there exists no sufficient statistics of before and after studies in the region, the available approved resources in the aforementioned algorithm may be used. In this research, the non-related-intersection accident statistics within the timeframe of 2014-2015 are implemented for 18 two-lane highways in Guilan province, Iran. Comparing the sensitivity analysis of the obtained results from the regression models, one is able to analyze the final achievements. The functions of accident modification factors obtained via Poisson regression model of the present study are compared with the available accident modification factors. Our achievements indicate the probable improvement of accident modification factor related to the roadside hazard rating (RHR) and calibrated base model in the flat and rolling areas of Guilan province, Iran