Density‐clusters ordered weighted averaging op erat or based on generalized trapezoidal fuzzy numbers

Abstract

To the problem of multi-attribute decision making with fuzzy numbers, this paper proposes a new type of operator called the density-clusters ordered weighted averaging (OWA) operator based on generalized trapezoidal fuzzy (GTF) numbers. This operator is abbreviated as the GTF-DOWA operator. A primary characteristic of the GTF-DOWA operator is that it considers the implicit structure of the GTF numbers to be aggregated by grouping the numbers to various local clusters. We discuss the grouping methods of the GTF numbers using the centroids of the numbers. The cluster weights are determined by the combined consideration of the decision maker's attitude and the scale of each local cluster. In addition, we discuss the primary properties of the GTF-DOWA operator. Finally, a numerical example regarding the selection of optimal alternative is provided. The aggregations of the GTF-DOWA operator are compared with those of the weighted arithmetic averaging (WAA) operator and the OWA operator based on GTF numbers to illustrate the validity of the GTF-DOWA operator. K E Y W O R D Scluster weights, generalized trapezoid fuzzy number, GTF-DOWA operator, OWA operator, uncertain multi-attribute decision making

Competitive behavior induced OWA operator and the weighting method

“…Since the appearance of the OWA operator, it has been widely used in various applications, 9–11 and many extensions have been developed about it 12–15 . The induced ordered weighted averaging (IOWA) operator is an interesting and important extension of the OWA operator, first proposed by Yager and Filev 16 .…”

“…Since the appearance of the OWA operator, it has been widely used in various applications, [9][10][11] and many extensions have been developed about it. [12][13][14][15] The induced ordered weighted averaging (IOWA) operator is an interesting and important extension of the OWA operator, first proposed by Yager and Filev. 16 The primary characteristic of the IOWA operator is that the reordering step is induced by the order-inducing variables rather than the argument values.…”