Application of machine learning for real-time water quality monitoring in developing countries: A review
Access to clean water is important in ensuring environmental sustainability and public health. However, water quality monitoring in developing countries faces significant challenges, which include; inadequate infrastructure, limited funding, and inadequate skilled personnel. Traditional water qualit...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-12-01
|
| Series: | Sustainable Futures |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2666188825005489 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Access to clean water is important in ensuring environmental sustainability and public health. However, water quality monitoring in developing countries faces significant challenges, which include; inadequate infrastructure, limited funding, and inadequate skilled personnel. Traditional water quality assessment methods are effective but are time-consuming and require extensive resources, limiting their applicability in developing countries. The emergence of the Internet of Things (IoT) technologies and Machine Learning (ML) present opportunities that can be exploited for monitoring water quality in real-time. Traditional reviews on ML-based water quality monitoring focus primarily on general ML models without considering real-time, low-power implementations in developing countries. To fill the gap, this study systematically reviews the application of Machine Learning (ML) and the Internet of Things (IoT) for water quality monitoring, with a special focus on TinyML’s potential for resource-constrained environments. Unlike previous reviews, this study explores model optimization techniques, data limitations, and challenges in scaling ML-based solutions. The key gaps identified include (1) the limited systematic reviews on ML for real-time water quality monitoring, (2) limited integration of TinyML into water quality solutions, and (3) insufficient exploration of the availability of localized datasets. A conceptual model for real-time water quality monitoring is also proposed to address these gaps. This work aims to guide the development of scalable, low-power ML models for water monitoring, especially in developing countries. |
|---|---|
| ISSN: | 2666-1888 |