摘要：An optimizationaly motivated learning-oriented interactive method for solving a class of discrete multicriteria choice problems with a large number of alternatives and a small number of quantitative criteria is proposed in the paper. At each iteration the decision maker (DM) can set his local preferences in terms of desired improvements or relaxations of the criteria. On this basis a discrete optimization scalarizing problem is constructed. A small ranked set of relatively close alternatives is defined with the help of this scalarizing problem. The ranked set is presented to the DM for selection of the most preferred alternative or for entering his/her new local preferences.
