Ontology-based equipment resource control in knowledge-intensive manufacturing

Modern knowledge-intensive manufacturing requires high precision, reliability, and efficiency of equipment operation. The failure of critical components can lead to significant financial losses, disruption of technological processes, and safety risks. In this context, equipment management and residual life prediction become key tasks to ensure uninterrupted production. This study analyzes the influence of technological parameters and steel chemical composition on the residual life of continuous casting machine molds. A comprehensive ontological model integrating data on steel temperature, mechanical loads, alloying elements, and equipment geometric characteristics has been developed. Objective: to design an ontological model for semantic integration of heterogeneous data and improving equipment residual life prediction accuracy using a hybrid approach combining ontological engineering and machine learning methods. Materials and methods. An OWL ontology was developed, including classes such as “Mold,” “Chemical Composition,” and “Technological Parameters.” SPARQL queries were implemented to identify dependencies between operational parameters and mold residual life. Machine learning methods were integrated for prediction and anomaly detection. Results. Key influencing factors were identified: steel temperature, copper content, and billet geometry. A prediction accuracy of R² = 0.85 was achieved, surpassing traditional statistical methods. Logical inference rules were developed for automatic detection of critical equipment conditions. Conclusion. The study demonstrated the effectiveness of a comprehensive approach to predicting the residual life of continuous casting machine molds, combining analysis of technological parameters, steel chemical composition, and equipment geometry. The integration of machine learning and ontological engineering into high-tech equipment management enables a shift from reactive to predictive maintenance, reducing costs and improving reliability. This is particularly crucial in industries where downtime costs are extremely high, and safety and precision requirements are critical. Further development of these technologies, including integration with digital twins and cognitive systems, opens new opportunities for Industry 4.0 and smart manufacturing.