Vain-Guy/tanzania-water-well-status-prediction

A data-driven analysis and predictive modeling of Tanzania’s rural water wells. This project cleans and explores the public “Pump It Up” dataset (Taarifa + Tanzanian Ministry of Water) to identify factors affecting well functionality and build machine-learning models to predict which wells are at risk of failure. #### **AUTHOR:** NORMAN # FLOW STATE: MAPPING TANZANIA'S WATER WELLS Access to safe and reliable water remains a daily uncertainty for millions of Tanzanians. This project analyzes the **Tanzania Water Wells dataset** to clean, explore and model rural water point data to support proactive maintenance, smarter investments and improved community resilience. ## BACKGROUND - **Population at risk:** ~70.5 million people in 2025; about one-third rely on wells and boreholes. - **Problem:** Many wells fail within a few years because of poor construction, environmental stress, limited maintenance or lack of spare parts. - **Impact:** Health risks (waterborne diseases), economic loss, and disproportionate burden on women and children fetching water. The Tanzanian Water Sector Development Program (WSDP) and SDG-6 (Clean Water and Sanitation) recognize that **predictive insights into well functionality** are key to sustainable access. ## OBJECTIVES - **Predict well status:** Functional vs. non-functional (binary classification after removing “needs repair”). - **Optimize resource allocation:** Prioritize high-risk wells for maintenance. - **Improve investment efficiency:** Focus funds where they have the greatest impact. - **Support data-driven planning:** Provide actionable insights to policymakers, NGOs and donors. - **Promote transparency and accountability:** Track well performance over time. ## STAKEHOLDERS The beneficiaries of this project are: 1. Tanzanian Ministry of Water 2. Local Communities 3. Non-Governmental Organizations (NGOs) 4. Donors and Development Partners 5. Local Governments (District & Regional Councils) 6. Private Sector & Contractors 7. ommunity-Based Water Committees (CBWCs) 8. Data Scientists and Researchers 9. Monitoring & Evaluation Teams 10. Policy Makers & Legislators At the center of these stakeholders are Tanzanian households, schools, and clinics, whose health, dignity, and productivity depend on reliable access to safe water. ## DA …