Disaster management and emergency response are critical in building community resilience to disasters. This study specifically focuses on temporary shelter planning, an essential component of disaster management, with an emphasis on identifying safe zones in post-earthquake landslide-prone areas. The research was conducted on the European side of Istanbul, with Büyükçekmece identified as the most at-risk district due to its high postearthquake landslide susceptibility. A systematic and transparent data collection process was employed to ensure the accuracy, reliability, and validity of the analysis. Geospatial data, topographic information, soil structure, and population density were among the key variables used to assess risk and suitability. Clustering methods, including kmeans and hierarchical clustering, were applied to categorize potential gathering and shelter sites based on criteria. The clustering analysis categorized the sites into different risk levels and helped identified the most vulnerable areas, providing a basis for targeted disaster preparedness and resource allocation. Moreover, the findings support policy-makers and urban planners in prioritizing shelter infrastructure investments in critical areas. This study demonstrates a robust framework to develop temporary shelter planning in high-risk regions, offering practical insights for disaster management and improving safety outcomes in similar contexts.
Secondary Disaster Landslides Induced by Earthquake Temporary Shelter Machine Learning Clustering Analysis
| Birincil Dil | İngilizce |
|---|---|
| Konular | Üretim ve Endüstri Mühendisliği (Diğer) |
| Bölüm | Araştırma Makalesi |
| Yazarlar | |
| Gönderilme Tarihi | 31 Aralık 2024 |
| Kabul Tarihi | 1 Mayıs 2025 |
| Yayımlanma Tarihi | 11 Kasım 2025 |
| Yayımlandığı Sayı | Yıl 2025 Cilt: 10 Sayı: 2 |