SMART DIAGNOSTICS IN INDUSTRIAL MAINTENANCE: A SYSTEMATIC REVIEW OF AI-ENABLED PREDICTIVE MAINTENANCE TOOLS AND CONDITION MONITORING TECHNIQUES

Abstract

This systematic review explores the integration of artificial intelligence (AI) in predictive maintenance and condition monitoring systems, with a focus on smart diagnostics across industrial applications. The study examines how AI technologies—including machine learning, deep learning, and hybrid models—are transforming traditional maintenance strategies by enabling automated fault detection, anomaly classification, and remaining useful life (RUL) estimation based on real-time sensor data. Drawing on a comprehensive review of 156 peer-reviewed articles published between 2015 and 2025, the research synthesizes current advancements in AI-driven maintenance tools and their applications in diverse industrial sectors such as manufacturing, energy, transportation, and utilities. Special emphasis is placed on the role of sensor technologies, including vibration, acoustic, thermal, and ultrasonic sensors, in providing the high-frequency data necessary for effective AI integration. Furthermore, the study discusses key theoretical underpinnings—including Reliability-Centered Maintenance (RCM), Prognostics and Health Management (PHM), Spatial Decision Support Systems (SDSS), and Information Systems Success Models—that frame the adoption and impact of smart diagnostics. Despite clear benefits, challenges such as data quality, model interpretability, and integration complexity remain persistent. This review contributes to both academic discourse and industrial practice by offering a comprehensive understanding of how AI technologies are redefining maintenance strategies and by identifying research gaps and opportunities for future exploration.