Literature Review on the Current State-of-the-Art in Research and Technological Advancements in the Field of Machine Learning Applied to Predictive Maintenance

This literature review examines the technologies of Machine Learning (ML) in Predictive Maintenance (PdM), highlighting the necessity for industries to boost production efficiency for competitive advantage amid growing global demands. It underscores the pivotal roles of technological advancements, especially the Internet of Things (IoT) and Big Data, in enabling smarter automation and intelligent production processes. By leveraging equipment sensor data, Predictive Maintenance serves as a crucial proactive maintenance strategy, enhancing reliability and accuracy through its diagnostic and prognostic stages. However, challenges in financial, organizational, data, and repair complexities hinder its full-scale implementation, necessitating a deeper understanding of ML techniques in PdM. The review further discusses AI/ML’s role in enhancing predictive maintenance, detailing applications of AI, ML, and Deep Learning (DL) in predictive analytics and identifying emerging trends like transformers and self-supervised learning, which promise to improve PdM outcomes. Through a structured analysis, this report underscores the evolving landscape of ML applications in PdM, highlighting both challenges and opportunities, and suggesting further exploration of ML algorithms in this field.