Road Safety Assessment under Uncertainty Using a Multi Attribute Decision Analysis Based on Dempster–Shafer Theory

Abstract

Road Safety Assessment (RSA) is an integral part of governmental programs that are meant to reduce road traffic accidents. RSA programs are periodically conducted by a committee of road safety experts based on subjective judgments. Hence, there is a degree of uncertainty inherent in almost every safety analysis. In this regard, this paper aims to employ an Evidential Reasoning (ER) approach for RSA modeling as a complex Multi Attribute Decision Analysis (MADA) problem to deal with unavoidable uncertainties such as ignorance and vagueness. The ER approach, which is different from the traditional MADA approaches, uses the belief decision matrix, which describes each attribute of an alternative by a distribution assessment using a belief structure. ER improves decision making process by using a belief decision matrix for problem modeling and the Dempster-Shafer (D–S) theory of evidence for attribute combination. A case study has also been provided to demonstrate the practical applicability of the proposed method to solve decision making problems under partial information.