Supplier selection for development of petroleum industry facilities, applying multi-criteria decision making techniques including fuzzy and intuitionistic fuzzy TOPSIS with flexible entropy weighting

Abstract

Generic supplier selection from the perspective of multi-criteria decision making (MCDM) methodologies including crisp, fuzzy and intuitionistic fuzzy analysis of decision matrices has received much attention, but less so specifically for the gas and oil industry, and in terms of comparing performance of a number of available techniques. A set of 30 criteria are identified for assessing supplier selection for facilities and field development projects across the petroleum industry. Bidders are assessed in terms of these criteria, with varying degrees of uncertainty and subjectivity, using linguistic scoring terms that are then transformed into crisp and fuzzy numerical sets. Eight MCDM scoring methods are described mathematically and applied to a facilities-procurement scenario in order to analyze a linguistic-assessment matrix for five alternative bidders using the 30 recommended criteria. These scoring methods are: linear; non-linear; the order of preference by similarity to an ideal solution (TOPSIS); Fuzzy TOPSIS (with and without entropy weighting); and, intuitionistic fuzzy TOPSIS (IFT) with three alternative methods for calculating entropy weighting (We). Performance of the eight methods is assessed by comparing calculated rankings for the five bidders in relation to the defined supplier selection scenario for a base case and ten sensitivity cases. The results of the analysis suggest that entropy weightings applied to fuzzy sets provide more consistent bidder selection, and led to the proposal of a new intuitionistic-fuzzy-TOPSIS-method-with-flexible-entropy-weighting method that enables the entropy weighting scale to be tuned to suit the circumstances of specific scenarios using equation 30 to flexibly normalize the entropy weighting scale.