Comparison Criteria and Efficiency of Thermomechanical Process Control Methods

The article presents a comparative study of process control methods for thermomechanical treatment of steels, emphasizing their relevance under the trends of digitalization and automation in metallurgical industries. Ten key performance criteria are identified and used to assess five major control method groups: technological, mathematical, machine learning-based, automated systems (ACS), and statistical. A comparative matrix highlights strengths, weaknesses, adaptability, interpretability, and cost factors. The results demonstrate that hybrid methods integrating automation, mathematical modeling, and machine learning offer superior performance, with intelligent neural network systems showing the greatest promise due to adaptability and high predictive accuracy.