Efficiency Performance and Dominance in Influential Subsets: An Evaluation using Fuzzy Clustering and Pair-wise Dominance

Abstract

The identification of influential subsets in efficiency studies is an ongoing research concern. Seaver et al. (1999) have demonstrated the value of using a fuzzy clustering strategy to identify influential subsets without knowing which production plans are outliers a priori. From a fuzzy clustering perspective, dominance is assessed in terms of the degree of belonging of a production plan in an influential subset. From a production point of view, dominance is assessed in terms of pair-wise comparisons. These concepts of dominance within influential subsets are illustrated using the preprint insertion data set (Girod, 1996).