Design and evaluation of an artificial intelligence-based digital twin for industrial manufacturing optimization

In this work, analyzed the role of artificial intelligence–based digital twins in improving the efficiency and stability of industrial manufacturing processes under dynamic operating conditions. Investigated the integration of a virtual manufacturing process model with artificial intelligence methods to enhance predictive accuracy and adaptive decision-making. Identified key technological parameters influencing production efficiency, energy consumption, process stability, and output quality within a generalized manufacturing environment. Studied the structure and functional interaction of the physical manufacturing system, digital twin model, artificial intelligence module, and optimization algorithm within a unified framework. Determined the effectiveness of supervised machine learning models in capturing nonlinear relationships between process inputs and performance indicators using simulated manufacturing data. Established that the proposed AI-based digital twin framework enables systematic optimization of manufacturing parameters without direct intervention in physical production systems. Formulated a validation approach based on comparative analysis of baseline and optimized operating scenarios to assess performance improvements. Proposed an AI-driven digital twin architecture that demonstrates measurable improvements in production efficiency, reduced energy consumption, enhanced process stability, and improved output quality, confirming its applicability for intelligent manufacturing and industrial process optimization.