Photo credit: Stellamaris Nakacwa With contributions from Stella Nacakwa, M.S. Candidate, West Virginia University & Courtney Clark, YouthMappers. The Gulu University chapter of YouthMappers, in
Updated version expected in 2022
This tool harnesses the power of machine learning to make predictions about the status of water points based on the past performance of similar water points in the country.
- Predict which water points are at a higher risk of failure in order to carry out preventative maintenance
- Identify high-risk water points in order to increase monitoring where it is most needed
- Determine which districts have relatively more high-risk water points to more effectively match maintenance budgets with likely need
- Select target country from the drop-down menu
- Select target district(s) from the drop-down menu
- Select whether you want the points on the map colored by the “Last Known Status” (when the last data was collected), or “Today’s Prediction”
- Click Submit
- Access data by clicking “Download Data”
This tool uses available WPDx attributes, such as #water_tech, #water_source, #pay, and others as training data for developing a classification machine learning model. The target variable is #status_id. The models are tuned to optimize the precision (percent of water points that are actually broken) and the recall (percent of all broken water points that are identified as high risk). Predictions are based on adjusting calculating the age of each water point based on #install_year and the current year. A priority for each water point (high/medium/low) is assigned based on the relative number of water points within 1 kilometer and the population within 1 kilometer.
Like all predictions, these predictions are based on probabilities and may not reflect the reality of the status of water points at a given point in time.
The Challenge: Dramatic acceleration needed to achieve SDGs Today, 2 billion people lack access to safely managed drinking water services, including: 1.2 billion people with