A multiple criteria decision making problem involving a discrete set of alternatives A is investigated. Each alternative is characterized by a fuzzy evaluation to a given number of criteria weighted by fuzzy numbers.The classical mean aggregation procedure is exploited in order to(i) isolate the smallest subset (called set of "best" elements)A included in A liable to eliminate all alternatives belonging tooA \ Ao(ii) rank the alternatives in various classes (being or not exclusive)using an antisymmetric fuzzy preference graph.