Abstract. A method of fuzzy inference for higher-order Takagi–Sugeno systems using the operation of decreasing the order of rules is proposed. It is shown that higher-order hierarchical fuzzy systems can be used to reduce the number of rules and parallelize fuzzy inference. The method is implemented in an intelligent software system to speedup computing in the problem of evaluating investment projects.

Keywords: Takagi–Sugeno fuzzy systems, fuzzy system order, order reduction, hierarchical dependencies, intelligent software systems, parallelization.

¹ V.M. Glushkov Institute of Cybernetics of National Academy of Sciences of Ukraine, Kyiv, Ukraine

 sershv@ukr.net

² V.M. Glushkov Institute of Cybernetics of National Academy of Sciences of Ukraine, Kyiv, Ukraine

 ponomarenko_roman@ukr.net