Authors

A.A. GONCHAROV, N.A. Seeds

Abstract

The active use of expert systems in various industries is dictated by their ability to solve problems of data interpretation, diagnostics, monitoring, designing, forecasting, planning and training. Each expert system is based on a knowledge representation model, while the production model is most widely used. The article describes the proposed method for increasing the efficiency of the process of inference in production systems, based on the use of aspect-oriented approach. An aspect-oriented approach allows you to identify overlapping functional elements and ensure their consolidation during the creation of the architecture and implementation of the system. First introduced by Gregor Kiczales in 1997, this approach is still popular today. As an example, the article provides a set of production rules of an expert system for selecting requirements for a given level of control [1]. In this set of production rules, the facts are the values ​​of the control levels, and in the form of actions the requirements for the selected level of control are presented. The aspect-oriented approach to the organization of production systems proposed in the article made it possible to increase the speed of logical inference in expert systems. Reducing the number of operations when searching for a solution and eliminating the need for enumerating facts and actions was made possible by highlighting the facts and actions from the set of production rules as aspects of overlapping.

Keywords

aspect-oriented programming, expert system, production rule, artificial intelligence.