Akademik Veri Yönetim Sistemi
Journal of Disaster and Risk, cilt.8, sa.3, ss.1341-1358, 2025 (Scopus, TRDizin)
This study introduces an integrated framework that combines fuzzy regression functions with the Fuzzy C-Means (FCM) clustering method for modeling coastal erosion risk. A synthetic dataset was generated using key environmental variables frequently highlighted in the literature as drivers of erosion processes, including wave height, wind speed, coastal slope, soil type, and the presence of protective structures. The dataset was partitioned into fuzzy clusters, and weighted regression models were subsequently developed for each cluster. Model predictions were aggregated based on cluster membership degrees to obtain the final erosion risk estimates. Model performance was assessed using Mean Absolute Error (MAE) and Root Mean Square Error (RMSE) metrics, with the relatively low error values demonstrating the method’s ability to predict erosion risk effectively under conditions of uncertainty. The findings suggest that the proposed approach provides a flexible and interpretable framework that can support both the modeling and risk assessment of complex environmental phenomena, particularly those characterized by multidimensional interactions and inherent uncertainties, such as coastal erosion. Furthermore, by considering the disaster related consequences of coastal erosion together with the risks of real estate value depreciation, structural damage, and loss of usability in coastal real estates, the study highlights that the model outputs can contribute significantly to disaster management, real estate valuation, and coastal planning processes, thereby enhancing the overall importance of the research.