Explainable AI for Predictive Maintenance App...

Explainable AI for Predictive Maintenance Applications

Author Name : Explainable AI for Predictive Maintenance Applications

DOI: https://doi.org/10.56025/IJARESM.2023.111223109

ABSTRACT In the ever-evolving industrial landscape, companies must continuously adapt and improve to stay ahead of the curve. Predictive maintenance (PdM) has emerged as a critical strategy to minimize downtime and boost manufacturing efficiency. This paper investigates the performance of various machine learning models using a synthetic dataset: Support Vector Machine (SVM), Random Forest (RF), Gradient Boosting, eXtreme Gradient Boosting (XGBoost), and Multi-Layer Perceptron (MLP). A Voting Classifier ensemble is employed to generate predictions. The article compares the performance of these models based on metrics such as accuracy, precision, recall, F1 score, and Matthews correlation coefficient (MCC). As machine learning models become more intricate, explainable AI (XAI) plays a crucial role in comprehending their predictions and decision-making processes. This paper utilizes eXAI techniques like Partial Dependence Plot (PDP), Local Interpretable Model-agnostic Explanations (LIME), and SHapley Additive exPlanations (SHAP) to shed light on the predicted behavior