Search academic texts

Information × Registration Number 0221U102336, 0119U100305 , R & D reports Title Development of algorithms and software for optimization of modern communication systems and inventory management systems. popup.stage_title Head Lebedev Yevhen O., Доктор фізико-математичних наук Registration Date 01-02-2021 Organization Taras Shevchenko National University of Kyiv popup.description2 Object of research: service networks, random processes, telecommunication systems, evolutionary models. Purpose: to create effective algorithms for calculating the performance of multidimensional stochastic systems with controlled parameters. construction of mathematical models of information struggle against migration, research of living conditions and construction of algorithm for finding the optimal strategy in the model of inventory management at different types of functions of total costs. Research methods: optimization of stochastic and deterministic systems; modern methods of discrete optimization; analysis of multidimensional Markov chains with continuous time. The conditions for the existence of a stationary regime are determined and explicit formulas of vector-matrix type for stationary probabilities for different classes of stochastic systems with repeated calls are obtained. For the Markov model of a system with repeated calls with one service device, one place in the queue and an infinite orbit with variable intensity of the input flow, which is controlled by a threshold strategy, the conditions of existence and formulas for ergodic distribution are found. The rate of convergence of the stationary distribution of a finite system with repeated calls to the stationary distribution of an infinite system with one service device and one place in the queue is obtained. A mathematical model and algorithm for finding the optimal solution have been developed for typical problems that arise in the study and optimization of inventory management systems. The asymptotic properties of random evolutions given by stochastic differential equations with Markov switches and momentum perturbations under conditions of aggregation and nonclassical approximation schemes are analyzed. Estimates of pulse transient functions based on observations of input signals and system response to these signals are constructed. Product Description popup.authors Grygor'eva Nataliya M. Knopov Pavlo S. Lukovych Olga V. Ponomarov Vadym D. Rozora Iryna V. Samoylenko Ihor V. Semenova Nataliy V. Usar Iryna Ya. popup.nrat_date 2021-02-01 Close