Abstract: As the "mother machine" of high-end manufacturing, high-end CNC machine tools have intelligent upgrading as a key development direction in the field of intelligent manufacturing. To address the problems of complex error sources, difficult accurate traceability and compensation in five-axis CNC milling, a set of error traceability and trajectory optimization methods based on machining process data was proposed. Firstly, the process data model of the entire CNC milling process was sorted out to clarify the composition system of machining errors. Secondly, an error traceability analysis framework was constructed, and machining errors were decoupled into three core sources: CAD/CAM programming errors, CNC system decoding and interpolation errors, and servo drive system errors. Finally, for errors caused by the servo drive system, a pre-compensation optimization method based on contour error was designed, and simulation verification was completed. The research results show that the proposed method can realize accurate traceability of machining errors, and the pre-compensation strategy can effectively reduce servo drive-related errors and improve the machining accuracy of machine tools.