A data-driven geographic information system and machine learning based multi-criteria framework for strategic wind...

Area of Science:

• Renewable Energy Systems
• Geographic Information Systems (GIS)
• Decision Support Systems

Background:

• Strategic wind energy site selection is crucial for maximizing efficiency and minimizing uncertainty.
• Traditional GIS approaches often lack the ability to integrate expert judgment and predictive capabilities.
• Ethiopia's Amhara Region presents significant but underexplored wind energy potential.

Purpose of the Study:

• To develop and validate an integrated framework for strategic wind farm site selection in Ethiopia's Amhara Region.
• To overcome limitations of static GIS by incorporating expert uncertainty, efficiency screening, and predictive modeling.
• To identify high-priority corridors for immediate wind farm deployment.

Main Methods:

• Fuzzy Analytic Hierarchy Process (FAHP) to weight key variables (wind speed, slope, elevation, etc.) and generate a suitability surface.
• Data Envelopment Analysis (DEA) for objective efficiency screening of high-potential zones.
• Machine learning models (Random Forest, SVM, XGBoost) for predictive suitability modeling and delineation of development corridors.
• Independent validation using multiple data sources and a specific wind project pipeline.

Main Results:

• Wind speed identified as the dominant factor (0.4211 weight) in FAHP analysis.
• North Shewa zone identified as frontier-efficient by DEA, contributing 32.25% of suitable land.
• Random Forest model achieved high accuracy (R²=0.8145, F1-score=0.9207), delineating 1,698 km² of high-priority corridors.
• Validation confirmed North Shewa as the highest-potential development corridor.

Conclusions:

• The integrated framework effectively combines subjective expert weighting, objective efficiency analysis, and predictive modeling for strategic wind energy planning.
• The methodology is particularly valuable for data-scarce regions, offering a robust decision-support system.
• The findings provide actionable insights for accelerating wind farm deployment in Ethiopia's Amhara Region, specifically in North Shewa.