Inspiration
Water pollution is a global problem, but solutions like constructed wetlands are often deployed without clear evidence of where they would have the highest impact. I wanted to build a simple, data-driven way to guide real-world environmental decisions instead of relying on intuition or fragmented reports.
What it does
This project ranks countries based on where water-cleaning wetlands should be built first. It combines environmental stress (water stress, sanitation, CO₂) with human impact (population exposure) to generate a Priority Score, then classifies countries into High, Medium, and Low priority tiers.
How we built it
We built an end-to-end Python pipeline using World Bank and Our World in Data datasets. The process includes data cleaning, normalization, and feature engineering. We compute: Need Score=f(water stress, sanitation, CO₂) Impact Score=f(population, density) Priority Score=0.5⋅Need+0.5⋅Impact
Challenges we ran into
The main challenge was balancing “need” vs “impact.” Early models over-prioritized either population-heavy countries or high-stress regions. We also dealt with missing and inconsistent global data across indicators.
Accomplishments that we're proud of
We built a fully reproducible, transparent ranking system that turns raw environmental data into actionable global insights. The model is interpretable and stable across sensitivity tests.
What we learned
We learned how difficult it is to design fair multi-factor scoring systems and how normalization choices strongly affect real-world conclusions.
What's next for Water-Cleaning Wetlands Priority Model
We plan to add geographic constraints (river basins, rainfall, land availability) and improve feasibility modeling to move from “priority ranking” to true deployment recommendations.
Log in or sign up for Devpost to join the conversation.