“Softer” optimization and control models via fuzzy linguistic quantifiers

A “softening” of a basic formulation of multicriterion optimization and control (multistage decision making) is presented. For optimization, instead of seeking an optimal solution that best satisfies all the fuzzy objectives as has been done so far, we seek an optimal solution that best satisfies most, much more than 50%, etc. (a linguistic quantifier, in general) of the fuzzy objectives. For control, we seek in turn an optimal sequence of controls that best satisfies the fuzzy constraints and fuzzy goals at most, much more than 50%, etc. of the control stages. A calculus of linguistically quantified statements based upon fuzzy sets and possibility theory is used. Some applications to softer evidence aggregation in expert systems are also indicated.