Fuzzy set and cluster analysis in qualitative multicriteria evaluation
This paper presents a new multicriteria decision aid model, dealing with discrete evaluation problems in which the impact matrix may include cardinal, ordinal and fuzzy criteria. It refers to the "beta problem formulation" (the aim is to choose all "good actions"), and its main feature is the treatment of ordinal information on the basis of statistical homogeneity indices in the context of a fuzzy cluster analysis. Fuzzy sets are also used to represent ordinal information and to supply the decision-maker with information on the compensation introduced in the model. In order to increase the equivalence between the procedures used for the different kinds of evaluation criteria, a new distance metric is introduced. This model does not use any classical weighting of criteria and is particularly useful when there are many alternatives. The procedure is illustrated by means of an empirical example regarding the evaluation of cars.
Bibliographic Reference: Paper presented: XV Convegno Annuale A.M.A.S.E.S., Grado (IT), Sept. 26-29, 1991
Availability: Available from (1) as Paper EN 36407 ORA
Record Number: 199111457 / Last updated on: 1994-12-02
Original language: en
Available languages: en