The integration of formal modeling and data-driven analysis is crucial for addressing complex challenges in Manufacturing Operations Management (MOM) systems under Industry 4.0. While our previously proposed refinement calculus of Object-Oriented Event-Graph (rCOE) offered a preliminary syntactic and semantic foundation for MOM modeling, its operational semantics were only briefly outlined. This paper extends our prior work by providing a complete and rigorous formal definition of rCOE’s operational semantics, enabling precise executable specification and dynamic analysis. Furthermore, using a synthetic yet industrially realistic dataset, we apply the Isolation Forest algorithm to identify key high-risk processes—such as “Matching Drill” with excessively long duration and near-zero pass rate, and “Straightening” with high execution frequency—that threaten production efficiency and product quality. To demonstrate the practical synergy between data-driven discovery and formal modeling, we model these critical processes using rCOE and conduct a simulation based on the operational semantics. The simulation successfully replicates the possible resource contention and scheduling conflicts, validating its ability to replicate anomalies and evaluate mitigation strategies. This work establishes an integrated, empirically-grounded formal methodology for enhancing the robustness of MOM systems, moving from anomaly detection toward explainable diagnosis and mitigative analysis.
Chen et al. (Wed,) studied this question.