Machine learning–integrated spatial decision framework for sustainable offshore wind and marine spatial planning: A Black Sea case study
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE, cilt.115424, ss.1-27, 2026 (SCI-Expanded, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 115424
- Basım Tarihi: 2026
- Doi Numarası: 10.1016/j.engappai.2026.115424
- Dergi Adı: ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
- Derginin Tarandığı İndeksler: Applied Science & Technology Source, Academic Search Ultimate (EBSCO), Engineering Source (EBSCO), Scopus, Science Citation Index Expanded (SCI-EXPANDED), Compendex, INSPEC
- Sayfa Sayıları: ss.1-27
- Hacettepe Üniversitesi Adresli: Evet
Özet
A newly developed integrated Fuzzy Analytic Hierarchy Process (FAHP), Entropy Weight Method (EWM), Machine Learning (ML), and Half-Quadratic Programming (HQ Programming) framework within a Geographic Information System (GIS) environment was designed for offshore wind farm (OWF) site selection. This hybrid methodology was applied to the Samsun–Ordu–Giresun continental shelf along the southern Black Sea to identify technically feasible, environmentally compatible, and policy-relevant offshore wind development zones. The scientific problem addressed in this work is the limited ability of conventional GIS–multi-criteria decision-making (MCDM) approaches to reconcile uncertain expert judgments, objective spatial variability, and data-driven criterion relevance within a single transparent weighting structure. FAHP represented expert judgments under uncertainty, while EWM provided objective entropy-based weights derived from geospatial variability. Five supervised ML algorithms—Decision Tree, Extreme Gradient Boosting (XGBoost), Ridge Classifier, Multi-Layer Perceptron (MLP), and Support Vector Machines (SVM)—were implemented using Annual Energy Production (AEP) as the target variable to quantify the empirical importance of spatial criteria. HQ Programming merged FAHP, EWM, and ML-based weights into a unified balance between subjective and objective assessments. Wind resource parameters contributed ∼18% to total suitability, infrastructural factors 16%, oceanographic–geophysical conditions 28%, and environmental–navigational constraints 30%. GIS-based analysis identified the Atakum–Ünye–Gülyalı corridor as the most favorable development zone. The framework helps decision-makers prioritize offshore wind investment areas, reduce conflicts with navigation and protected zones, and support grid- and port-oriented marine spatial planning. This study provides a transferable basis for explainable, data-calibrated spatial decision-making in offshore renewable energy planning.