Using Neural Networks to Optimizing Industry Economic Chains (Using the Example of Nuclear Energy)

The article substantiates the use of neural networks as an innovative tool for planning interindustry balance. The author’s previously proposed concept of the economic cross is adapted to the development of the nuclear energy sector. An assessment is made of the potential for introducing neural networks into the process of forming intra- and inter-industry links in the two-component nuclear energy sector. The needs of the nuclear energy economy for tools that ensure the implementation and full realization of the economic potential of neural networks at the industry level are assessed. Based on the provided assessments, a forecast system has been developed regarding the structure of economic counterparties of the Rosatom State Corporation in terms of the formation of neural network support for the industry. Separate groups are allocated to those participants in the proposed economic cross contour whose creation seems economically feasible in the process of designing the economic cross. Separate blocks highlight elements that are recommended for inclusion in economic chains as structural elements of the Rosatom State Corporation and as independent participants in commercial relations operating in a competitive market. The potential for economic results that can be achieved through the introduction of neural networks in the process of designing the economic sectoral cross is examined using the example of nuclear energy. Based on the analysis conducted in this article, an economic model of industry counterparty interaction chains is proposed. Its development is technologically feasible given the availability of industryspecific big data analysis using neural networks. An adaptation of the economic cross of two-component energy is proposed based on the proposed model, outlining the financial and strategic advantages of the proposed format for organizing the flow of information, resources, and risks within the industry.