Abstract: The stability and reliability of equipment operation are pivotal determinants of production efficiency and product quality in modern industrial manufacturing. To ensure the reliability of industrial equipment, fault classification prediction is particularly critical. Existing fault classification models proposed in current research are often limited to specific industrial equipment, resulting in poor model generalizability. Therefore, to address common industrial equipment fault types such as cooling system failures, power supply faults, excessive strain failures, and wear-related failures, a universal fault classification model based on the random forest algorithm is proposed, incorporating characteristic data prevalent in industrial equipment including torque status, air temperature, tool wear, process temperature, and rotational speed.Feature importance is calculated by the model using the Gini index to precisely determine the contribution of each feature to fault prediction.Validation is conducted using the AI4I 2020 Predictive Maintenance Dataset. To comprehensively evaluate model performance, metrics including accuracy, precision,recall, and F₁-score are adopted, with macro-average and weighted-average methods applied for holistic assessment. It is indicated by the results that torque status serves as the critical fault feature.An accuracy of 85.4% in fault diagnosis is achieved by the model, with particular excellence demonstrated in predicting normal equipment states at 95.8% accuracy, outperforming three classical algorithms: support vector machine (SVM), gradient boosting decisiontree (GBDT), and logistic regression (LR).