TOPSIS and its Extensions: Applications for Road Safety Performance Evaluation

Abstract

Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is one of the well-known classical multi-criteria decision-making (MCDM) techniques. In this chapter we illustrate the use of this method to combine individual safety performance indicators (SPIs) into an overall index of road safety performance for a set of European countries. In this respect, to deal with the subjective kind of uncertainty on data (such as linguistic variables given by experts) which are usually adopted to assess the weights of criteria/indicators, we explore an extension of the classical TOPSIS method to fuzzy environments. Moreover, due to the ever increasing number of SPIs used to reflect each road safety risk factor in a more comprehensive way, we consider a hierarchical structure of the indicators in this study. Accordingly, a hierarchical fuzzy TOPSIS model is realized and applied to combine the multilayer indicators into one overall index. Comparison of the resuls based on the three models (i.e., the classical TOPSIS, the fuzzy TOPSIS, and the hierarchical fuzzy TOPSIS) demonstrates the effectiveness of applying the hierarchical fuzzy TOPSIS method to handle the problem of linguistic expression instead of crisp values given by experts, and to take the layered hierarchy of the indicators into account which is seldom considered in the current road safety index research.