Although machine learning methods for industrial maintenance systems have been extensively studied in recent years, their practical adoption remains slow. One key challenge is the lack of comparative analyses of machine learning systems. To address this, we conducted a systematic literature review (2012–2021) of 104 monitoring and prediction systems. Additionally, we identified five design patterns—high-level construction frameworks—through k-means cluster analysis. Our findings indicate that while monitoring and prediction systems vary in their operational choices, they typically employ similar learning strategies (e.g., supervised learning) and tasks (e.g., classification, regression). This study aims to provide researchers and practitioners with insights into common characteristics, contextual applications, and emerging trends.

M L, Classification, regression