OPTIMIZATION OF RELIABILITY OF COMPLEX NON-REPETITIVE SYSTEMS

Abstract

Context. In this paper we consider problems of optimizing the reliability of complex non-repairable systems. Such systems consist of a set of interrelated elements. Optimizing the reliability of such systems is a complex computational problem and requires the development of new methods. Objective. Construction of mathematical models of complex non-repetitive systems and development of effective methods for optimization their reliability. Method. We use the method of exact quadratic regularization to solve problems of optimizing the reliability of complex systems. Precise quadratic regularization allows us to transform multiextremal problems of optimizing the reliability of complex systems to the problem of maximizing the norm of a vector on a convex set. We use the effective primal-dual interior point method and the dichotomy method to solve the transformed problem. The method of exact quadratic regularization made it possible to significantly expand the classes of solvable optimization problems for the reliability of complex systems. This is confirmed by comparative numerical experiments. Results. Comparative numerical experiments show that the method of exact quadratic regularization is more efficient than existing methods for solving this class of problems. This method allows you to extend classes of problems optimizing the reliability of complex systems for which it allows you to find optimal solutions. Conclusions. We proposed an effective method for optimizing complex non-repetitive systems, this method showed the best numerical results.