Neutrosophic multi-criteria decision making and gis-based integrated model proposal for post-disaster temporary container settlement site selection: The case of Hatay

One of the most critical requirements following disasters is the provision of safe and adequate temporary shelters for affected populations. In Hatay, the province most severely impacted by the February 6, 2023 Kahramanmaraş earthquakes, temporary shelter settlements established rapidly during the emergency response phase gradually exhibited significant physical vulnerabilities, becoming focal points of emergencies and secondary disasters. This study proposes an integrated Geographic Information System (GIS)-based and Multi-Criteria Decision-Making (MCDM) framework to evaluate the site selection and suitability of post-disaster temporary shelter areas. During the two years following the earthquakes, a total of 306 incidents were recorded in temporary shelter settlements, including 155 fires, 147 flood events, and 4 storms. These incidents resulted in 16 fatalities, 67 injuries, and damage to 5,075 container units and 420 tents. Site selection criteria identified through a PRISMA-based literature review were refined to 25 final criteria using the Delphi method with the participation of 15 experts. The criteria were weighted using the Neutrosophic DEMATEL method, which identified preestablished settlement plans and evacuation accessibility as the most influential factors. By integrating the Analytic Hierarchy Process (AHP) with GIS, a suitability map for central Hatay was generated, revealing that only 36% of the existing container settlements were located in suitable areas, while several were established within floodplains. Based on the Jenks Natural Breaks classification, 65 alternative shelter sites were proposed for small-, medium-, and large-scale settlements and subsequently ranked using the Neutrosophic TOPSIS and VIKOR methods. The results of the Spearman rank correlation and Monte Carlo simulation confirmed the reliability, consistency, and robustness of the proposed framework under uncertainty. Overall, the findings demonstrate that the developed integrated model provides a method-independent, reliable, and robust decision-support system for identifying suitable temporary shelter sites across different spatial scales.