Intelligent configuration management in modular production systems: Integrating operational semantics with knowledge graphs

This paper presents an innovative approach to integrating data-driven strategies into intelligent manufacturing systems, specifically targeting the challenges of configuration management in modular production environments. To address the distinct and evolving requirements of customized products, we propose a dynamic configuration management methodology that automatically adjusts system settings in real-time. This approach utilizes operational semantics to formalize the interactions between production modules, capturing essential operational information for intelligent decision-making. A novel control mechanism is developed, using knowledge graphs to semantically represent and manage the relationships between production system components and settings. By mapping these, the system can determine optimal configurations based on real-time data and specific operational requirements. The interaction between the control mechanism and the knowledge graph ensures continuous adaptability, enabling the system to reconfigure dynamically in response to changes. This method was validated in an industrial dry-air leak testing scenario, demonstrating its effectiveness in adaptability. • A semantic approach is proposed to manage leak testing in modular production systems. • Real-time reconfiguration is enabled by linking system requirements with live sensor data. • A knowledge graph is utilized to support adaptive control and flexible decision-making. • The approach is validated through an industrial leak testing scenario. • Improved responsiveness and scalability are demonstrated in dynamic manufacturing environments.