A multiple attribute relative quality measure based on the harmonic and arithmetic mean

Abstract

In this paper a relative quality measure is presented that is applicable to rank alternatives characterized by multiple attributes or performance measures. The quality measure proposed is based on the harmonic and arithmetic mean, and allows for a simple and quick analysis of the alternatives with respect to their attributes. An alternative ranked by this method and having the maximum relative quality of one can be considered as an extreme efficient unit according to the method of data envelopment analysis. The proposed method of the relative quality measure is compared with different multiple attribute decision making approaches that apply simple additive weighting, the MADM methods based on OWA operator, maximizing deviations, and information entropy, and the PROMETHEE II method.