Alternative Approach for Combining Multiple Crash Modification Factors Using Adjustment Function and Analytic Hierarchy Process

Abstract

The safety effects of multiple treatments have recently emerged as an important issue of validation for the Highway Safety Manual procedures to improve performance of the predictive process. To estimate the combined safety effects of multiple crash modification factors (CMFs) more reliably, several combining approaches have been suggested. However, several critical issues still exist for the combining methods, such as overestimation, region-specific method, and a nonscientific approach. Therefore, this study suggests a novel adjustment method to combine multiple CMFs and to enhance the reliability of combining the safety effects of multiple treatments. Various combinations of CMFs for single and multiple treatments were estimated or obtained from previous studies and used for an exploratory analysis. Moreover, an alternative combining approach with the development of an adjustment function was suggested through comparison with the existing combining methods by using a multicriterion decision-making process. The results show that the proposed alternative combining method provides better estimates than the existing methods and can account for different roadway types and severity levels. Thus, it can be recommended that the safety effects of multiple treatments be estimated with the proposed new combining approach to overcome the overestimation issue and produce results that are more reliable.