A review of machine learning and internet-of-things on the water quality assessment: Methods, applications and future trends

Clean and safe water is fundamental to human health and environmental sustainability, yet increasing pollution due to urbanisation, industrialisation, and climate change poses significant risks. Traditional water quality monitoring relies on manual sampling and laboratory analysis, which are often costly, time-intensive, and lack real-time insights. This review critically examines the integration of machine learning (ML) and the internet of things (IoT) for real-time water quality monitoring and predictive analytics. The study evaluates peer-reviewed research published between 2016 and 2024, focusing on advancements, limitations, and future trends in automated water quality assessment. ML models, including random forest, extreme gradient boosting, support vector machines, and neural networks, have been more frequently used in water quality research and have achieved high accuracies (R²=0.99 in regression and 0.99 accuracy metric in classification). Explainable AI (XAI) which can explain the decision making process of ML, is underutilised, appearing in only a few recent studies. While IoT significantly improves real-time contamination detection, persistent challenges remain in sensor fouling, data continuity, data privacy, network reliability, and cybersecurity. Such challenges can hinder the scalability and effectiveness of long-term IoT implementations. Integrating IoT with machine learning enhances water quality monitoring by enabling real-time data collection, remote tracking, and predictive analytics. This synergy improves efficiency, addresses monitoring challenges, and supports sustainable water management.