Abstract: A simulated loading test was conducted on a horizontal lathe over 12 cycles totaling 288 hours based on national standards for reliability testing and accuracy inspection. During the test, 15 geometric and 20 positioning accuracy parameters were measured. A linear regression model was used to decouple the contributions of loading force, torque, rotational speed, and their combinations to accuracy degradation. The method applies to both scalar and array-type accuracy indicators. After identifying key influencing factors, further analysis explored how specific loads affect accuracy degradation. The proposed model predicts accuracy retention trends, evaluates degradation under specific stresses, and guides users in selecting cutting parameters to control accuracy loss during operation.