To improve the operating efficiency and performance of modern technical and economic control systems, decision making must be based on an ever expanding set of diverse interconnected factors. This has led to rapid development of decisionmaking theory and wide use of mathematical decision methods. Applied decision-making problems usually involve multiple criteria. In economics they arise in the context of production planning or in automatic design of complex system outline [1]. One of the fundamental concepts in multicriterion optimization theory is the notion of Pareto-optimal or efficient decisions. A decision is Pareto-optimal if the value of any of the criteria can be improved only by accepting a worse value of the other criteria. Let the compactum X C E ~ be the initial set of alternatives (decisions); the continuous function F(x) = (Fl(X) ..... Fro(x)) is the vector efficiency criterion. For decision making it is sufficient to consider only the set of efficient decisions; if the decision x is completely characterized by its criterion value y = F(x) C E m, then the set of Pareto-optimal criterion values is called the Pareto set and is denoted by P(Y). It is relevant to consider the problem of approximation of the Pareto set, which is computationally costly and is characterized by instability of the Pareto set in the presence of errors in initial data specifications and computation errors [2]. One of the techniques for constructing an approximation of the Pareto set is to reduce the original multicriterion problem to approximate solution of a parametric programming problem with finitely many parameter values, specifically: find y(2) ~ N(/',, s, X),