Comparison of Several Aggregation Techniques for Deriving Analytic Network Process Weights

Abstract

Water pipelines are exposed to severe aging and deterioration challenges. The weights of importance of deterioration factors shall be estimated to predict the condition of underground water pipelines. Analytic network process (ANP) is one of the most common multiple criteria decision making techniques. The fuzzy set theory has been employed to tackle the vagueness and imprecision in human judgments. Accordingly, the overarching aim of this research is computing the weights of importance of these factors using fuzzy ANP (FANP). Chang’s extent analysis method is one of the most frequently used algorithms in FANP. For group decision making, two aggregation approaches are considered, namely geometric mean (GEO) and minimum-maximum (MIN-MAX). The performance of the modified extent analysis method is evaluated against the original method using four evaluation measures. These measures are satisfactory index, group minimum violations, group Euclidean distance, and distance between weights. Most of the evaluation measures indicate that the modified extent analysis (MIN-MAX) method exhibits better results than other methods. A water distribution network in Shaker Al-Bahery, Egypt is used to demonstrate the application of the proposed model.