AI Summary & Key Takeaways
AI InsightThe results show that wind speed, water depth, and capacity factor are the most important variables in predicting offshore wind farm suitability.
📌 3 Key Takeaways
Abstract
Offshore wind energy has become a key part of the global renewable energy transition. In this study, a machine learning–based classification model was developed for Türkiye’s marine areas by using the suitability zones previously identified through Geographic Information Systems (GIS)–based Boolean analysis and Multi-Criteria Decision-Making (MCDM) methods. The main goal is to build an understandable decision support system (DSS) that can predict site suitability for new locations by modeling these existing suitability results with explainable machine learning techniques. For this purpose, the Random Forest classification algorithm was trained on approximately 4 million data points representing 24 technical, environmental, socio-economic, and security-related criteria. The model’s decision logic was interpreted using the SHAP (Shapley Additive Explanations) method. The results show that wind speed, water depth, and capacity factor are the most important variables in predicting offshore wind farm suitability. Besides validating previously identified suitable areas, the developed model acts as an explainable DSS with both predictive and interpretive abilities for unexplored sites. This combined approach—merging traditional GIS and MCDM results with machine learning—offers a more flexible, scalable, and transparent framework for assessing Türkiye’s offshore wind energy potential.