rennykefs/Bank-Geospatial-Network-Optimization

A comprehensive Geospatial Data Science study designed to optimize Kenya's banking network across Kenya's three major urban centers. # Bank Africa: Banking Network Optimization **Strategic Site Selection for Kenya's Major Urban Centers** ## Project Overview This project utilizes Geospatial Data Science (GWR, p-Median Optimization) to identify the top 9 optimal banking locations across Nairobi, Mombasa, and Kisumu. ## 🛠️ Tech Stack - **Language:** Python 3.12 (uv package manager) - **Frontend:** Streamlit - **Database:** PostGIS (PostgreSQL) - **Models:** Geographically Weighted Regression (GWR) & Huff Gravity Model ## 📊 Key Results - **98.2%** Predictive Accuracy (GWR) - **18%** Reduction in average customer travel time. - **14%** Increase in captured market demand. ## 🛡️ Setup 1. Clone the repo. 2. Install requirements: `pip install -r requirements.txt` 3. Add your `secrets.toml` to the `.streamlit` folder.