Clustering Z-information based on a system of fuzzy reference requirements

The paper develops a clustering model of multi-criteria object evaluations, taking into account the reliability results obtained. Clustering is based on system fuzzy reference requirements about importance evaluated characteristics objects for each clusters. Object evaluations are formalized basis linguistic Z-numbers, both numbers which values variables. Information and cluster presented as set pairs (according to number characteristics) consisting (importance characteristic corresponding cluster) Z-number (an evaluation within this its reliability). Using information, ratings determined in accordance with cluster. Fuzzy objects, defined numbers, reflect compliance requirements. comparative analysis all one proposed makes it possible identify best representative (or representatives) under consideration determine degree belonging remaining carried out A numerical example given, shows effectiveness developed method Z-information.