Machine Fault Diagnosis Using Random Forest with Recursive Feature Elimination and Cross Validation

In modern industrial environments, early and accurate machine fault diagnosis is crucial for minimizing downtime, reducing maintenance costs, and ensuring operational safety. This research presents a robust fault classification framework that combines Recursive Feature Elimination with Cross-Validation (RFECV) and Random Forest classifiers to address the challenges of high dimensionality, overfitting, and limited model generalization. The proposed approach begins with comprehensive data preprocessing, followed by RFECV to identify and retain the most relevant features, thereby enhancing model efficiency and accuracy. Subsequently, a Random Forest classifier is trained on this optimized feature set to classify four fault types: No Failure, Power Failure, Tool Wear Failure, and Overstrain Failure. By integrating feature selection with ensemble learning, the framework effectively mitigates high variance and improves robustness under varying operational conditions and data distributions. Experimental results demonstrate that the proposed methodology achieves a high predictive accuracy of 99.2% along with improved computational efficiency, making it highly suitable for real-time fault diagnosis applications in smart manufacturing systems.