New Study Shows Safe Water Optimization Tool Nearly Triples Effectiveness in Refugee Camps

A recent groundbreaking study has demonstrated that using a machine learning-powered tool to optimize chlorination levels can significantly improve the safety of drinking water in humanitarian settings. Developed by researchers at York University, the Safe Water Optimization Tool (SWOT) is an open-source, evidence-based application designed to tailor water chlorination practices based on routine monitoring data.

The study, published in BMJ Global Health, analyzed water quality data from the Kutupalong-Balukhali refugee settlement in Cox's Bazar, Bangladesh, where SWOT was implemented. Findings revealed that households following the tool's guidelines experienced safe water approximately 90% of the time, compared to only 35% with traditional universal guidelines. This indicates nearly a threefold increase in effective safe water supply.

Syed Imran Ali, the lead author and director of the Humanitarian Water Engineering Lab, emphasized the significance of this development. "Our research shows that SWOT can deliver safe drinking water at nearly three times the rate of standard practices, reinforcing its value as a vital tool in emergency response situations," Ali explained. He also highlighted that the success of SWOT depends on integrating water monitoring teams with water treatment operations to optimize chlorination levels effectively.

The study underscores the need for support systems that enable water system operators to adapt and respond quickly to changing conditions. Collaborators included experts in machine learning and public health, among them Professors and researchers from York University, as well as members of Médecins Sans Frontières.

SWOT was created after Ali’s experience working in South Sudan with Médecins Sans Frontières. It aims to replace faulty universal chlorination guidelines with customized, data-driven solutions. Since its development, the tool has helped provide safe water to over 700,000 people globally, including challenging environments in Yemen, Gaza, and Uganda.

Enhanced by machine learning, SWOT utilizes routine water monitoring data to dynamically adjust chlorination levels, ensuring water safety and reducing the risk of waterborne diseases in humanitarian crises. As resource constraints and operational challenges intensify, evidence-based solutions like SWOT are increasingly vital to maintaining health standards in vulnerable populations.

More details about this promising approach can be found in the full study: Proof-of-concept evaluation at Cox’s Bazar of the Safe Water Optimization Tool in BMJ Global Health.