Construction of dynamic knowledge graph for the production process of manufacturing enterprises

In the heavy production process of manufacturing enterprises, it is difficult to effectively share resource data, which leads to relatively scattered production resources and low utilization rates. This study combined product lifecycle knowledge with time series data and put forward the Framework of Dynamic Manufacturing Knowledge Graph Construction (FDMKG) for the production process of manufacturing enterprises. Specifically, the production process data was divided into two dimensions: static resources and dynamic data stream. The static resources and the production data stream were semantically associated to generate a dynamic knowledge graph. Then, taking the production line of a packaging company as an example, the FDMKG was applied to construct the dynamic knowledge graph. The results show that FDMKG achieves query performance improvements ranging from 1.3 to 26.4 times and attains high diagnostic accuracy with a precision of 0.92, a recall of 0.90, and an F1-score of 0.91. Furthermore, FDMKG efficiently organizes and reuses knowledge throughout the production process, enhancing manufacturing knowledge traceability and reasoning. This study provides valuable practical guidance for knowledge management applications in manufacturing enterprises. Nevertheless, this study focuses on manufacturing production processes, future research should incorporate real-time data from diverse industries to construct a cross-industry dynamic knowledge graph.