This project identifies which Pittsburgh neighborhoods should be prioritized for drainage facility improvements or temporary installations to minimize flood impacts on residents and infrastructure under a limited city budget.
The model integrates:
- FEMA flood hazard data
- Population and infrastructure exposure
- Facility and hydrology characteristics
- Optimization models for investment prioritization
The outcome supports data-driven resilience planning for the City of Pittsburgh.
equal_weights_map and pop_heavy_map Python files produce maps to show top 10 high risk neigborhoods.
Implemented in Risk Ranking Scoring.ipynb
This notebook takes all the data in the Data folder as inputs, processes, cleans, and normalizes them, and outputs risk_rank_equal.csv and risk_rank_popheavy.csv in the Outputs folder.
Implemented in optimization_results.ipynb
This is the main notebook for the project. It combines:
- Risk ranking data (
risk_rank_equal.csv,risk_rank_popheavy.csv) - Optimization models (MIP and Greedy)
- Visualization outputs (maps and PNGs for presentation)
- Mixed-Integer Programming (MIP)
- Greedy Heuristic