bmuhwezi/git-commit--m-Initial-commit-SME-electricity-geospatial-analysis-pipeline-

Estimates the effects of complementary infrastructure on grid connected SMEs in Kenya # Effects of Complementary Infrastructure on SME Electricity Consumption in Kenya A comprehensive geospatial analysis pipeline for examining Small and Medium Enterprise (SME) electricity consumption patterns in Kenya using multiple spatial features. ## Overview This repository contains code and tools for analyzing the relationship between SME electricity consumption and various geospatial factors including population density, road access, financial service providers, building structures, and nighttime lights. The analysis uses 5+ years of electricity billing data from grid-connected SMEs across Kenya. ## Key Features - **Geospatial Feature Extraction**: Extract features within 500m buffers around each SME location - **Parallel Processing**: SLURM-based cluster computing for large-scale data processing - **Multiple Data Sources**: Integration of satellite imagery, census data, and utility records - **Panel Data Regression**: Perfomes panel data Fixed Effects regression ## Repository Structure ``` Kenya-sme-electricity-geospatial-analysis/ ├── README.md ├── requirements.txt ├── config.py # Configuration and file paths ├── main.py # Main pipeline orchestrator ├── main.R # Main pipeline orchestrator │ ├── feature_extraction/ │ ├── fsp_extraction.py # Financial service providers │ ├── population_extraction.py # Population density (WorldPop) │ ├── roads_extraction.py # Road access and length │ ├── nightlights_extraction.py# VIIRS nighttime lights │ └── slurm_scripts/ # SLURM job submission scripts │ ├── fsp_array.sh │ ├── population_array.sh │ ├── roads_array.sh │ └── nightlights_array.sh │ ├── utils/ │ ├── data_processing.R # Data preprocessing utilities │ ├── utility_functions.R # Data preprocessing utilities │ ├── data_cleaning.py # Data preprocessing utilities │ ├── spatial_utils.py # Spatial calculation fun …