A statistical approach for improvement of Best Worst Method (BWM)

Abstract

This paper endeavors to measure the consistency of a decision-making tool, popularly known as Best Worst Method (BWM), which is one of the latest developments in multiple-criteria decision analysis (MCDA). BWM is finding a vast arra of applications in the literature. Several investigators have extended this tool. BWM measures the weight of decision-making criterion and is recognized as a subjective decision tool. The first step in this method is to find the best and worst criterion, while we suppose several experts are asked to present their evaluation over set of criteria. The aim is to measure how these judgments are consistent and reliable. So, we statistically (using χ2 distribution) add a pre-evaluation to experts’ opinion and analyze whether the agreement of experts’ opinions is satisfactory and group opinion is established. This action improves the quality of the decision-making process by incorporating the reliability evaluation of experts’ idea. This extension for BWM helps decision makers in facilitating and getting results that are more consistent for criteria evaluation. We present examples in sustainable construction and architecture project.