Global estimates indicate that more than 800 million people live without access to safe drinking water.
Many of these people live in rural areas and rely on groundwater – one of the safer potable forms of water – extracted using handpumps. With neighbouring handpumps potentially hours away, broken ones often result in people turning to other water sources, such as surface water from rivers or lakes, which may bring significant but avoidable health risks.
A cross-disciplinary Oxford project, involving researchers from the School of Geography and the Environment, and the Department of Engineering Science, has developed a technology that addresses the challenges caused by the unreliability of the pumps on which people depend. Using machine learning techniques, the team has designed a system that can analyse data to predict pump failures before they happen and accurately estimate levels of groundwater available.
The work combines scientific advances in engineering with policy, development and entrepreneurship and has led to changing national policy in Kenya and Bangladesh, as well as benefiting tens of thousands of people. The research has to date received over £3m in competitive research grants from ESRC, NERC, EPSRC, DFID, UNICEF, the Fell Fund and the Clarendon Fund.