DEVELOPMENT OF FORECASTING AND MANAGEMENT MODELS FOR AUTOMATED PRODUCTION SYSTEMS

Sun Hui, Zhuang Xintong

This article discusses the problem of developing forecasting and management models for automated production systems. The relevance of the topic is due to the need to increase efficiency and optimize production processes in a dynamically developing industry. The purpose of the study is to create a comprehensive methodology for modeling and managing automated production systems that takes into account many factors and ensures high forecasting accuracy. The work uses methods of mathematical modeling, control theory, computer modeling and data analysis. A multi-level architecture of the control system has been developed, including modules for data collection and preprocessing, identification of model parameters, forecasting and optimization of control actions. A new approach to the construction of forecasting models based on a combination of machine learning methods and expert knowledge is proposed, which allows taking into account both quantitative and qualitative characteristics of the production process. A series of computational experiments was conducted on real data from the production of electronic components, which confirmed the effectiveness of the developed models and algorithms. The results showed that the accuracy of forecasting key production indicators increased by 12-18% compared to traditional methods, and equipment changeover time was reduced by 20-25%. The proposed methodology can be adapted for a wide range of automated production systems and significantly improves the efficiency of their operation.

Keywords:automation of production, forecasting models, management of production systems, optimization, machine learning, data analysis