Abstract: The manufacturing process of sub-assemblies is complex and susceptible to various ‌anomalies, making production scheduling difficult. At the same time, ‌manufacturing systems generate multi-source heterogeneous data‌, ‌hindering information systems from acquiring real-time comprehensive information‌ and consequently ‌impairing scheduling plan formulation‌. Cyber-physical ‌integration can achieve data interaction between the physical domain and the cyber domain, but existing research ‌remains inapplicable to shipbuilding and has not resolved unified data semantics. ‌Additionally‌, ontology-based production scheduling research ‌faces limitations‌. To address this challenge, this paper focuses on ship sub-assembly process and proposes a semantic integration and dynamic scheduling decision architecture based on ontology. Firstly, a method is proposed to construct a workshop scheduling ontology model, a data model, and a rule model, for establishing a unified data semantic system, and ensuring the rapid integration of multi-source and heterogeneous data in the information domain. Secondly, a real-time data integration platform based on Flink CDC technology is established to ensure timely acquisition of assembly process data in the physical domain. Subsequently, an assembly process control system of ship sub-assembly is developed, which includes production monitoring, disturbance detection, and dynamic scheduling functions. Through real-time interaction of data in the cyber and physical domains, the manufacturing process of the ship sub-assembly is dynamically controlled. The effectiveness of the proposed system is verified by comparative analysis before and after implementation.