A Novel Owa Weights Determination Based On Ordinal Dispersion

One of the most common techniques to find the adequate weights in OWA operators is based on the orness concept. The weights are determined by maximizing the entropy (variation) for a fixed orness value. The entropy can be viewed as a dispersion measure for nominal variables. Taking into account that the weights in OWA operators can be viewed as an ordinal measure (instead of a nominal measure), in this paper we propose a novel way to determine these weights based on ordinal dispersion measures instead of the entropy measure. We find an explicit formula for the weights in this case and we show the differences in some multicriteria decision making examples.