To address the difficulty of quality traceability in the field of automotive foundries using sand molds, data of different types, formats, structures, and qualities that affect the quality of castings during the production process were collected. Based on the manufacturing execution system (MES), the collected data were processed using multi-source heterogeneous fusion technology to generate standardized quality-related data of the casting manufacturing process, which was linked to the castings in real-time and stored in a database or data lake for sharing. It was then used for production planning and scheduling, production process control, overall quality management of the production process, and internal logistics management systems to achieve traceability of individual casting product quality and batch quality. The processed data play an important role in quality traceability and guiding the optimization of quality parameters in the casting production process by comparing them with process quality standards. After the system is implemented and a large amount of data is accumulated, big data analysis and quality model algorithms are applied to achieve quality prediction and feedback control. This study explores the path for a foundry enterprise to enhance product quality through data empowerment in the field of production manufacturing.