Deciding the Future of Energy: A Hybrid Approach with Fuzzy Logic and Explainable AI

 Article & Journal: Renewable Energy Source Ranking and Analysis Using Fuzzy MCDM, ML, and XAI Techniques Published in: Baltic Journal of Modern Computing

Study Summary: Selecting the right renewable energy source for a country isn't just about cost; it's a multi-dimensional puzzle involving environmental impact, technical feasibility, and social acceptance. In this study, we developed a comprehensive decision-support system for Türkiye. We combined Fuzzy Multi-Criteria Decision Making (MCDM) methods (AHP, TOPSIS, VIKOR) with Machine Learning (ML) algorithms. Crucially, we integrated Explainable AI (XAI) techniques (SHAP and LIME) to interpret the results, ensuring that the "why" behind the rankings is as clear as the rankings themselves.

Behind the Research: Energy planning is often stuck between subjective human opinion and opaque computer models. We wanted to bridge this gap. While Machine Learning can process vast datasets to predict optimal energy sources, decision-makers (governments, investors) need to trust these predictions. That’s where Explainable AI (XAI) comes in. By revealing which features—like "initial investment" or "CO2 emission"—drove the AI's decision, we transformed a technical optimization problem into a transparent policy-making tool. This work demonstrates that AI shouldn't just decide for us; it should explain itself to us.

Citation & DOI: Özkurt, C., Canay, Ö., Tunç, E. A., Aydın, E., & Velioğlu, B. S. (2025). Renewable Energy Source Ranking and Analysis Using Fuzzy MCDM, ML, and XAI Techniques. Baltic Journal of Modern Computing, 13(3), 656-679. DOI: 10.22364/bjmc.2025.13.3.06