Modeling Based on Retrospective Quantitative Information Processing Via Virtual Expert Resource

Abstract

In modeling of complex systems, including objects with inhomogeneous characteristics, the effectiveness of the obtained models is strongly influenced by the applied methods of identification, focused on the similar system components. In this regard there is a need for the improved methods of identification of objects with inhomogeneous characteristics and the development of approaches to the quantitative evaluation of their quality characteristics. To solve this problem in the retrospective quantitative information processing, the use of expert and virtual resource, based on the interaction of an expert (s) and a computer system, is provided. An algorithm for building a mathematical model based on the combined use of expert and computer resources, is introduced. The approximate models obtained in that way are recommended to use in filling up the missing information at the preliminary stage of investigation and thinking of plenty of system development alternatives.