Settlement-Level Least-Cost Electrification (Benin, 2025-2040)

Open-source toolkit combining geospatial analytics with Multi-Tier demand modelling and strict LCOE costing.

Why it stands out

• Spatial resolution: every known settlement evaluated.
• Technology-agnostic: grid, mini-grid, and SHS cost curves in one model.
• Transparent parameters: all assumptions centralized in parameters.py.
• Reproducible: single CLI command or Python API call.
• Policy-ready outputs: investment and technology map directly usable by planners, financiers, and researchers.

Headline numbers

Technology	Settlements	Population	Investment
Grid	1,523 (9%)	11.5M (81%)	$1.58B
Mini-Grid	7,552 (44%)	2.1M (15%)	$0.67B
SHS	8,130 (47%)	0.5M (4%)	$0.01B
Total	17,205	14.1M	$2.26B

Workflow

flowchart TB
    subgraph INPUT["INPUT"]
        A[Geospatial Settlement Data]
    end

    subgraph CLASSIFY["CLASSIFICATION"]
        B1[Urban / Rural]
        B2[Energy Access Tier]
        B3[Household Count]
    end

    subgraph DEMAND["DEMAND ESTIMATION"]
        C1[Residential]
        C2[Commercial]
        C3[Agricultural]
        C4[Public Services]
        C5[Growth Projection]
    end

    subgraph COST["TECHNOLOGY COSTING"]
        D1[Grid Extension]
        D2[Mini-Grid]
        D3[Solar Home Systems]
    end

    subgraph SELECT["OPTIMIZATION"]
        E1[LCOE Comparison]
        E2[Least-Cost Selection]
    end

    subgraph OUTPUT["OUTPUT"]
        F[Technology & Investment<br/>per Settlement]
    end

    A --> B1 --> B2 --> B3
    B3 --> C1 & C2 & C3 & C4
    C1 & C2 & C3 & C4 --> C5
    C5 --> D1 & D2 & D3
    D1 & D2 & D3 --> E1 --> E2 --> F

    style INPUT fill:#e3f2fd
    style CLASSIFY fill:#fff3e0
    style DEMAND fill:#f3e5f5
    style COST fill:#e8f5e9
    style SELECT fill:#fce4ec
    style OUTPUT fill:#e0f2f1

Loading

Approach:

1. Classification - Categorize settlements by urbanization level and energy access tier using demographic and wealth indicators
2. Demand - Estimate sectoral electricity demand (residential, productive, public) and project growth over planning horizon
3. Costing - Calculate Levelized Cost of Electricity (LCOE) for each technology option based on infrastructure distances and load characteristics
4. Optimization - Select least-cost technology per settlement considering technical constraints

Data

Required: geometry, population

Optional: num_buildings, mean_rwi, has_nightlight, dist_to_substations, distance_to_existing_transmission_lines, dist_main_road_km, dist_lake_river_km, num_health_facilities, num_education_facilities

Installation

git clone https://github.com/BadibossPy/settlement-level-electrification-optimization-benin.git
cd settlement-level-electrification-optimization-benin
pip install -e .

Usage

CLI:

python run_model.py --input data/settlements.geojson --output results.geojson

Python API:

import geopandas as gpd
from benin_least_cost.parameters import ProjectConfig
from benin_least_cost.demand import run_demand_model
from benin_least_cost.lcoe import run_lcoe_model

gdf = gpd.read_file("data/settlements.geojson")
config = ProjectConfig()
gdf = run_demand_model(gdf, config)
gdf = run_lcoe_model(gdf, config)
print(gdf["optimal_tech"].value_counts())

Output Fields

• projected_demand - Annual demand 2040 (kWh/year)
• projected_peak - Peak load (kW)
• lcoe_grid, lcoe_mg, lcoe_shs - LCOE per technology (USD/kWh)
• optimal_tech - Selected technology (Grid / MiniGrid / SHS)
• investment - CAPEX for selected technology (USD)

Parameters

All parameters in benin_least_cost/parameters.py:

• Planning: discount rate 8%, horizon 15 years, population growth 2.7%
• Grid: MV $14,000/km, LV $5,500/km, transformer $8,000, losses 18%
• Mini-Grid: PV $700/kW, battery $300/kWh, CF 18%
• SHS: $80-$600 per household by tier

Structure

├── benin_least_cost/      # Core Python package
│   ├── demand.py          # Demand estimation
│   ├── lcoe.py            # LCOE and selection
│   ├── parameters.py      # All parameters
│   └── schema.py          # Data validation
├── run_model.py           # CLI entry point
├── tests/
│   └── test_logic.py      # Unit tests
├── data/
│   └── settlements.geojson
├── notebooks/             # Optional visualization
│   └── electrification_analysis.ipynb
└── requirements.txt

Next steps

Input preparation

• Settlement points — any GeoJSON/GeoPackage with at least geometry and population. Optional columns (buildings, RWI, distances, facilities) improve accuracy.
• Infrastructure distances — compute dist_to_substations, distance_to_existing_transmission_lines, and dist_main_road_km with standard GIS nearest-line tools (PostGIS ST_Distance, QGIS ‘Distance to nearest hub’).
• Wealth proxy (RWI) — download raster from WorldPop Relative Wealth Index, sample mean value at each settlement.
• Night-lights — VIIRS night-time lights, threshold >0 to upgrade Tier 1 settlements to Tier 2.
• Facilities — point layers of health and education sites, aggregated per settlement.

Data sources

• WorldPop 2025 population, relative wealth index
• GRID3 settlement layers (Benin)
• World Bank night-lights (VIIRS, 2024 composite)
• OpenStreetMap roads, health, and education facilities (Geofabrik extract)
• Africa Electricity Transmission & Substation GIS (KTH / World Bank)

Roadmap

1. Scenario module — run sensitivities (cost decline, discount rate) via YAML config.
2. Web map — publish results and methodology on an interactive map.
3. Integration — link with national least-cost grid expansion tools (e.g., NetworkPlanner) for corridor optimisation.

Verification

pytest tests/