RECURRENT FUZZY RULES (TEODORESCU'S FUZZY SYSTEMS) IN ECONOMIC PROCESS MODELLING

摘要

In this paper we model the decision-making and control processes using a new type of fuzzy logic rules, proposed by H.N. Teodorescu. These rules implied to modify the confidence degree of the rules using a recurrent formula. In the classical models, the certainty associated by the experts to the rules is well determined, but this kind of knowledge is not so proper to describe the complex human processes. With this more general fuzzy models (Teodorescu model), the description of the human iterative process of decision-making is improved. Practically, all the fuzzy applications may be modified, by replacing the oversimplified classical fuzzy rules with the recurrent fuzzy rules, obtaining a new class of fuzzy control systems. We analyze the behavior of the single-input, single-output Sugeno recurrent systems of type 0. We intend to extend in the future the analysis to more complex fuzzy systems (multiple inputs) in different fuzzy logics.