Yazar "Elbidari, Diana Sabah Nimma" seçeneğine göre listele

Listeleniyor 1 - 1 / 1
  • Küçük Resim
    Yayın
    Decision support system for earthquake disaster management
    (İstanbul Gedik Üniversitesi, Lisansüstü Eğitim Enstitüsü, 2024) Elbidari, Diana Sabah Nimma; Gümüş, Tuğbay Burçin
    The global occurrence of earthquakes has continued to be a significant area of concern due to the possible impact earthquakes can have on human lives, properties, and even a nation's economy. Earthquake has proven to be the most devastating natural hazard experienced in man's existence on Earth. Earthquake occurrences have always been related to the seismicity of a particular region. Regions with high seismicity tend to experience frequent earthquakes, while regions with low seismicity tend to have a low rate of earthquake occurrence. Turkey, China, India, Japan, and a few other countries have had a high rate of earthquake occurrence in the past years due to their high seismicity, which has led to significant setbacks in their economies at the times of the events. This thesis focuses on developing a decision support system (DSS) to enhance disaster management in the aftermath of earthquakes. Earthquakes are highly destructive natural hazards, often resulting in extensive damage to infrastructure and the loss of human lives. Effective disaster management is crucial but faces multiple challenges, like acquiring timely data, optimally allocating limited resources, and coordinating complex response efforts. A specialized DSS can aid authorities by integrating analytics into the decision-making process. The main challenge is providing decision-makers with essential earthquake impact information to coordinate emergency response and recovery. The chaotic aftermath causes delays in rescue operations, suboptimal resource allocation, and preventable secondary losses. There is a need for a DSS that can rapidly analyze disaster data and optimize post-earthquake decision-making. The methodology involves designing a modular DSS architecture focused on disaster management needs, like real-time monitoring, damage assessments, and strategy recommendations. It integrates a mathematical model using seismic, population, and building data to estimate casualties, injuries, and shelter needs. Multi-criteria decision-making (MCDM) methods prioritize response efforts. The system is validated using real earthquake data. This data is based on the 2023 Turkey earthquake, taking Kahramanmaraş province as case study. The goal is to optimize rescue operations, allocate resources effectively, rebuild critical infrastructure, and restore normalcy with minimal losses. The study focuses on the real-world application and validation of theoretical models in the aftermath of the 2023 Kahramanmaraş earthquake in Turkey. These evaluate the estimated versus actual casualties, showcasing the precision of the Python-based mathematical model used. It also extends the analysis to injury predictions and shelter needs, highlighting the model's reliability in post-disaster scenarios. Additionally, the multi-criteria decision-making approach is used to prioritize areas for resource allocation and rescue efforts. The study also introduces a DSS website, designed to enhance disaster response efficiency. The estimation of casualty's post-earthquake in the Kahramanmaraş province of Turkey was remarkably accurate, with the proposed model drawing more than 98% accuracy. Similarly, the estimated number of injuries had an accuracy of 91.386%, affirming the effectiveness of the proposed model. The study also estimates the number of people who need shelter, confirming the robustness of the proposed model with an accuracy of 98.683%. Based on the MCDM approach, the study suggests Onikişubat city as the most needed city for resources and rescue efforts, followed by Dulkadiroğlu and Elbastan. The DSS website, with a simple and effective user interface, aims to enhance the rapid action of authorities by providing essential information and prioritizing the most affected locations that need immediate attention for resources and rescue efforts. Finally, the study's methodical approach and the robustness of the proposed models demonstrate the potential for building an intelligent decision support system to effectively manage earthquake disaster response and recovery.