A fuzzy clustering methodology for linguistic opinions in group decision making

The objective of this paper is to automate a decision-aid tool, which provides homogeneous clusters from a set of heterogeneous opinions for a particular criterion to evaluate an item under group decision scheme. In such situation, decision-making by a group of experts becomes more realistic and consistent while they provide more or less homogeneous responses. But in real practice, the homogeneity among the opinions for a specific criterion to evaluate an item is not maintained due to the diversity among the experts’ several cognitive factors as well as biasness. As a result, the group's overall effectiveness is suffered and making a true decision becomes difficult as well as sometimes confusing. In order to avoid the heterogeneity among the opinions (fuzzy numbers), we propose here a fuzzy clustering methodology based on a fuzzy distance measure. Also a ranking index is introduced on the basis of Ordered Weighted Average (OWA) operator. Finally, a fuzzy multi-criteria decision-making problem on a flight simulator software development project is considered here to employ the proposed technique. The results are discussed and compared.