To cope with the challenges of monitoring dynamic and variable quality variation into supply chain, diagnosing the abnormal variation at the right moment, is a difficult problem that a enterprise in supply chain faces in process quality control. A dynamic process quality control method, which integrated quality prevention, analysis and diagnosis, was all put forward. This method integrated several enabling technologies such as the theory of similarity manufacturing, Statistical Process Control (SPC), neural network. Furthermore, some key enabling technologies were studied in detail, including process quality analysis on-line based on similarity process and process quality diagnosis based on Elman. It is basis of realizing network, intelligent and automatic process quality control.