Analysis of trends in scientific research development of the industry on the basis of mathematical modeling methods

Бесплатный доступ

Modern global trends cause a revolutionary transition in the development of industry to new high-tech industries with innovative infrastructure, which are controlled by artificial intelligence as part of the creation of cyber-physical systems. All these complex transformations are impossible without a powerful research component, which is the foundation for a quick and high-quality transition to an innovative development that can ensure the competitiveness of the domestic industry at the global level. The most important components of research development are the development, implementation and use of the latest production developments, the number of personnel involved in research and development, as well as state support in the form of financing and implementation of various incentive grant programs. The purpose of the article is to predict the trends in the research development of the industry. To achieve this goal, the following tasks were solved in the work: a review of Russian and foreign literature in the field of application of mathematical modeling methods to identify industry development trends is given; determination of the range of indicators characterizing the level of research development; built trend lines to predict indicators for the future period; the possibilities of applying the methods of neural network modeling using modern information resources are considered. To solve the tasks in the article, the following methods were used: methods of descriptive statistics, correlation and regression analysis, polynomial trend line, training of neural networks based on the Wolfram Mathematica software package. Substantiation of the possibilities of using modern methods of mathematical modeling to assess trends in research development can become a promising tool for improving the state regulation of the country›s industry.

Еще

Research development, industry, correlation and regression analysis, forecasting, neural network modeling, wolfram mathematica

Короткий адрес: https://sciup.org/148325305

IDR: 148325305   |   DOI: 10.37313/1990-5378-2022-24-4-68-74

Статья научная