×

You are using an outdated browser Internet Explorer. It does not support some functions of the site.

Recommend that you install one of the following browsers: Firefox, Opera or Chrome.

Contacts:

+7 961 270-60-01
ivdon3@bk.ru

  • The use of various criteria in solving an inhomogeneous minimax problem

    The article deals with the solution of the inhomogeneous minimax problem using genetic algorithms, as well as using several variants of the Plotnikov-Zverev algorithm. Three types of criteria for setting the function of assessing the fitness of individuals are described. The efficiency of genetic algorithms is compared with the Plotnikov-Zverev algorithm, using different criteria of the fitness function of individuals. According to the results of the computational experiment, it was concluded that the use of the quadratic criterion for the modified Goldberg model, using a two-point crossover, increases the efficiency of the genetic algorithm, and the accuracy of this solution is higher compared to the solutions obtained using modifications of the Plotnikov-Zverev algorithm.

    Keywords: schedule theory, inhomogeneous minimax problem, modified Goldberg model, genetic algorithm, minimax criterion, quadratic criterion, minimax criterion, cubic criterion, Plotnikov-Zverev algorithm

  • An approach to reducing the operating time of the modified Goldberg model in solving the inhomogeneous minimax problem

    The article deals with the problem of solving the inhomogeneous minimax problem typical for the theory of schedules. This problem is NP-complete and for it there is no exact algorithm of the solution having polynomial time for problems of big dimension. A modified Goldberg model is considered as a method of solving this problem. Godberg's model is considered with several crossovers and the most effective mutation. Under certain parameters (a large number of individuals and repeats), the modified Goldberg model receives a solution for a long time, so the article analyzes in detail one of the approaches to reduce the operating time without loss of accuracy. Since it is extremely difficult and practically impossible to make calculations analytically, a computational experiment was put into operation. As a result of the computational experiment, the tables provide a comparison of the efficiency of the modified Goldberg model after the use of HT technology. The use of HT technology leads to a significant reduction in time costs.

    Keywords: single-point crossover, two-point crossover genetic algorithm, modified Goldberg model, mutation, minimax problem, scheduling theory, individual, generation, hyper-threading

  • Hybrid algorithm application for inhomogeneous minimax problem solution by using strong mutation

    The article deals with the problem of the minimax solving that are a typical problem for the theory of schedules. As a possible method for solving this problem, we consider a hybrid model, which is one of newest of genetic algorithms. A effency this model based on comprasion result accuracy, obtained by using two-point crossover, hybrid algorithm, simple and strong mutations.

    Keywords: two-point crossover, hybrid algorithm, modified Goldberg model, mutation, minimax problem, scheduling theory, strong mutation, individual, generation

  • Modifications of the tournament selection Investigation in the inhomogeneous minimax problem solution by using modified Goldberg model

    In this paper was discuss, for the first time in detail, many ways of forming a new generation of tournament selection using a modified Goldberg model with the most common crossover and the original mutation. A computational experiment was carried out, which revealed the dominant advantage of using the "left individual and mutation" strategy for solving the heterogeneous minimax problem. It has been experimentally established that an increase in the number of individuals and repetitions leads to an increase in the time to obtain a solution to the inhomogeneous minimax problem when using any strategy, but at the same time to improve the accuracy of the solution.

    Keywords: single-point crossover, genetic algorithm, modified Goldberg model, mutation, minimax problem, scheduling theory, elite individual, individual, generation

  • Investigation of the single-point crossover apply in the inhomogeneous minimax problem solving

    In the article is considered the minimax problem solving. This is a characteristic problem of the schedules theory. As a possible method for solving this problem, a modified Goldberg model is considered, which is one of the varieties of genetic algorithms. The efficiency of this model is described on the results accuracy estimate, obtained by using a standard crossover for various types of mutations and parameters of the genetic algorithm.

    Keywords: single-point crossover, genetic algorithm, modified Goldberg model, mutation, minimax problem, scheduling theory, elite individual, individual, generation