I. Introduction
The notion of a fuzzy set that was introduced in 1965 by Zadeh (see [17]) has proved to be a powerful tool for modeling imprecise or linguistic data. One of the issues that appears in Decision making theory regarding the modeling of uncertainty is the issue of evaluation of fuzzy quantities. Generally, the most popular method consists of defuzzification and, then, ordering of fuzzy values. If the measure theory is used as a background, the evaluation of fuzzy values can be transferred to integration of an estimation of the corresponding -cuts with respect to a probability measure. The roots of this type of approach to the estimations of fuzzy values can be found in [11] and [2]. The working assumption was that the lower -levels are less important than the upper ones. Further investigation of this subject contains the concept of the interval-valued expectation of fuzzy numbers done in [4] and the possibilistic mean values (lower, upper and interval-valued) that are investigated in [7] and [20].