Problems of forecasting output

This article shows the need to revive the scientific interest and practical use of science-based planning methods in solving the problems of effective development of the economy of the country as a whole, and its individual regions and industries. The analysis of works devoted to methods of forecasting and planning, research and application of science-based planning method—program-target method is presented, as a result of which the conclusion is made about the necessity of using the capabilities of artificial intelligence to solve the problems of strategic planning. The results of machine learning (ML) of the neural network SARIMAX are presented. A comparative analysis of the use of neural networks SARIMAX and LST for output forecasting is made. Conclusions are drawn about the possibilities of using neural networks for the development of regional socioeconomic systems and branch systems in the conditions of mobilization economy.