In industries, artificial intelligence (AI) has transformed diagnostics and predictive maintenance (PdM), increasing operational efficiency, lowering downtime and promoting sustainability. This study investigates AI-based techniques, such as machine learning, statistical quality control (SQC) and hybrid models, which use historical and real-time data to predict equipment failures and optimize maintenance plans. A strong emphasis is placed on SQC methods, specifically cumulative sum (CUSUM) and exponentially weighted moving average (EWMA) control charts, which are critical for sustaining industrial quality. These statistical methods detect tiny changes in process performance, enabling early defect identification and stability. The AI4I 2020 Predictive Maintenance Dataset uses machine learning classifiers such as Random Forest to anticipate machinery failures, while CUSUM and EWMA charts supplement AI models for quality monitoring. Using GridSearchCV for hyperparameter tuning enhances the model's performance even further. The results demonstrate a 99.8% accuracy, indicating that merging AI with SQC methodologies such as CUSUM and EWMA gives an accurate, cost-effective and scalable approach to modern industrial maintenance.