RESEARCH OF MATHEMATICAL MODELS AND METHODS OF MULTI-CRITERIA OPTIMIZATION OF WORK DISTRIBUTION IN THE FIELD OF IT SERVICES

Abstract

The article addresses the problem of optimizing task distribution in the field of IT services using mathematical models and multi-criteria optimization methods. The main goal of the research is to develop and evaluate effective approaches for distributing tasks under conditions of limited resources and conflicting performance criteria such as execution time, energy consumption, and service quality. The study formulates mathematical models and applies three optimization methods: the Pareto Method, Weighted Sum Method, and Genetic Algorithm. The results of computational experiments revealed significant differences in efficiency: the Pareto Method showed the best balance between speed and quality (0.000194 sec), the Weighted Sum Method yielded average performance (0.010478 sec), and the Genetic Algorithm demonstrated high solution quality but at the cost of slower execution (0.013633 sec). These results highlight the trade-offs inherent in method selection. The research concludes that the Pareto Method is most suitable for real-time and resource-constrained systems, while the Genetic Algorithm is more appropriate for solving complex, large-scale optimization problems.