Traditional maintenance strategies, typically reactive or schedule based, often fail to prevent water infrastructure failures and operational inefficiencies. In response, machine learning–driven predictive maintenance has emerged as a promising approach for enhancing water infrastructure reliability through early fault detection and data-driven decision-making. This study presents a bibliometric review of 29 peer-reviewed, Scopus-indexed publications to examine global research trends, thematic developments, and geographical contributions within this domain. The findings reveal a strong dominance of BRICS countries, particularly India and China, in both publication output and citation impact. India emerges as the most influential contributor with eight publications, followed by the United States (four) and China (three), indicating significant advancements in predictive maintenance and machine learning applications. In contrast, African representation remains limited, with only South Africa and Sudan appearing in the dataset, highlighting a critical regional research gap. The most prolific journals include Water (Switzerland) and Sensors, reflecting a strong focus on water infrastructure and sensor-based technologies. Thematic analysis identifies four dominant research streams: advanced artificial intelligence techniques, water infrastructure applications, IoT-enabled decision-support systems, and infrastructure management systems. Furthermore, density visualisation reveals a highly centralised knowledge structure characterised by distinct clustering patterns, suggesting both specialisation and limited cross-cluster integration. The study emphasises the need to expand research and implementation efforts in African regions to support the adoption of machine learning–driven predictive maintenance, thereby enhancing the sustainability, resilience, and operational efficiency of water infrastructure systems.

Keywords: Machine Learning, Predictive Maintenance, Operational Efficiency, Water Infrastructure Systems, Developing Countries