Development of a Discrete Optimization Operation Solution Information Technologies Based on Swarm Intelligence

The object of this research is the procedure of building information technologies, the functioning of which is based on the methods of swarm intelligence, for solving problems of discrete optimization.To solve any optimization problem in the plurality of swarm algorithms, there will surely be at least one algorithm that will give at least satisfactory results. However, there is not and can't be an algorithm that could provide high efficiency in solving all optimization problems. Therefore, for each of the swarm algorithms, classes of problems that it solves can be distinguished: algorithms are better than others; something like other algorithms; worse than other algorithms.In the course of the research, information technologies were used to solve discrete optimization problems based on swarm algorithms. Methods for applying various classes of swarm intelligence algorithms for solving discrete optimization problems are obtained. Methods of swarm intelligence to solve a specific class of problems re combined. The optimal values of the parameters of certain methods of swarm intelligence are determined.An information technology is developed to use swarm algorithms depending on the class of the discrete optimization problem, based on the characteristics of swarm algorithms (type of input parameters, neighborhood of populations, type of population formation, type of iteration processes). This makes it possible to choose the relevant swarm algorithm for solving applied problems and to classify these tasks depending on the characteristics of the swarm algorithms that are used to solve it.An information technology is developed using a combination of different methods of swarm algorithms for solving a certain class of problems, which, unlike other approaches, is based on a hybrid approach using swarm algorithms depending on their characteristics. This allows to take advantage of a specific swarm algorithm and thereby increase the efficiency of solving certain classes of applied discrete optimization problems