Purpose <p>Industrial rotating machines are critical assets in sectors such as power generation, manufacturing, transportation, oil and gas, and air conditioning systems. Ensuring their reliable operation requires effective fault detection, diagnosis, and accurate prediction of remaining useful life (RUL). Although existing review studies address fault diagnosis and RUL estimation separately, an integrated perspective within a unified predictive maintenance (PdM) framework remains limited. The purpose of this review is to bridge this gap by providing a comprehensive overview of fault diagnosis and RUL estimation methods for industrial rotating machinery.</p> Methods <p>This paper systematically reviews the literature on common faults in rotating machines, categorizing them based on their causes and impacts on machine performance. Diagnostic techniques, including conventional vibration-based methods and advanced approaches leveraging machine learning and deep learning are analyzed. For RUL estimation, the review examines key stages such as data acquisition and preprocessing, degradation modelling, RUL prediction, model training and validation, and deployment with continuous monitoring.</p> Results <p>The study compares physics-based, data-driven, and hybrid prognostic models, highlighting their strengths and limitations. Emerging techniques, including transfer learning, are discussed as promising solutions for improving generalization and robustness in real-world applications.</p> Conclusion <p>The review identifies open challenges and future research directions, emphasizing the need for intelligent, integrated, and reliable PdM frameworks to enhance the safety, efficiency, and sustainability of industrial rotating machinery.</p>