E-Madzi Automated Water Kiosks
🤖 AI-generated summary of the sources below · 🔎 Last verified
This overview was generated with AI from the public sources listed below and checked against them before publication. It has not had a complete human fact-check — treat it as a starting point and follow the sources.
Lilongwe Water Board's E-Madzi ('water electronically') kiosks replace attendant-run communal water points with automated dispensers: residents tap an RFID smart card, credit is deducted for the exact volume drawn, and a web dashboard lets the utility monitor each kiosk remotely. Piloted at four kiosks in 2019 under the World Bank-financed Lilongwe Water and Sanitation Project and scaled up with COVID-19 response funds, automation removed the attendant's margin and cut the price of kiosk water by about 65%. LWB today lists 91 kiosks fitted with E-Madzi units and 28 more under construction, while much of its kiosk network remains attendant-run. A 2023 peer-reviewed survey of 111 peri-urban households found 98.2% satisfaction with prepaid communal points, but flagged the system's weak spot: the solar-powered dispensers lose pressure on cloudy days.
📊 Impact
Automation cut users' kiosk water price by about 65% and served roughly 8,750 customers in the first expansion phase (World Bank, 2020); a 2023 peer-reviewed survey of 111 peri-urban households (Chiumya & Gumbo, World Water Policy) found 98.2% satisfaction and no income-based affordability gap.
🎓 Lesson
Removing the human attendant is where the saving lives - but it moves the point of failure to power and hardware: a solar dispenser that loses pressure under clouds rations water by weather, with no attendant left to improvise.
Sources
Continue exploring
Compare this project
Explore the wider context
📎 Cite this project
1001 Smart Cities (2026). “E-Madzi Automated Water Kiosks” — Lilongwe, Malawi. The Smart City Atlas. https://1001smartcities.org/projects/lilongwe-emadzi-water-kiosks/ (last verified 2026-08-07). Data: CC BY 4.0.
The underlying data is free to reuse with attribution — see the open data page.
