Figure 2
The bar graph displays the top important features governing milling anomaly detection. The x-axis represents the feature importance score, ranging from 0.00 to 0.16. The y-axis lists the feature names, including smarms_std, xmarms_mean, smarms_mean, ymarms_max, ymarms_mean, S_Axis_Load_std, smarms_max, energy_mean, N, and S_Axis_Load_mean. The bars are horizontal and indicate the importance score for each feature. The highest importance score is for smarms_std, followed by xmarms_mean, smarms_mean, ymarms_max, and ymarms_mean. The graph highlights that vibration-related variables dominate the prediction of machining anomalies, indicating that milling instability is governed primarily by the dynamic response of the machine. All values are approximated.Feature importance ranking obtained from the XGBoost model using interval-wise MTConnect features. Vibration-related variables dominate anomaly prediction, demonstrating that machining instability is governed primarily by the dynamic response of the machine rather than nominal cutting parameters alone