International Scientific Journals
of Scientific Technical Union of Mechanical Engineering "Industry 4.0"

The paper presents standard statistical procedures for determining expert assessments for the influence of magnesium alloy components on the controlled mechanical properties: tensile strength and relative elongation. There are obtained neural models describing mechanical characteristics from the amount of aluminum, manganese, nickel, iron and silicon directly related to the exploitation properties of the product. The applied approach makes it possible to determine of compositions ensuring relatively optimal values of the explored quality indicators.

In the present paper is proposed a classification of the problems and the mathematical models for choosing optimal structural variant while designing a technical system. Algorithms and software models were developed based on known methods for determining the priority of criteria functions. The algorithms and models are implemented in a dialog system for structural optimization. Results are shown from using the software for solving a particular problem.