فهرست نشریات علمی دسترسی آزاد ایران

Efficient crisis management following natural disasters, particularly earthquakes, necessitates optimal resource allocation, appropriate facility location, and precise transportation planning. In this study, a multi-stage mathematical model is proposed for the design and optimization of a relief supply chain network comprising distribution centers, temporary shelters, temporary care centers, and hospitals. Aiming to minimize total operational costs and reduce shortages of relief items, the model simultaneously optimizes location, allocation, and transportation decisions. To achieve dynamic and realistic demand forecasting for relief items in temporary shelters, time-series-based machine learning algorithms are utilized. Furthermore, by employing fuzzy logic, uncertainty in the capacity of vehicles and medical centers is modeled and incorporated into the decision-making process. The proposed model is formulated as a Mixed-Integer Linear Programming (MILP) problem and utilizes the weighted sum method to integrate multiple objectives. Results indicate that integrating machine learning-based demand forecasting with fuzzy uncertainty management significantly enhances the efficiency of the relief network, reduces response time and total costs, and improves the service level provided to the victims. This approach presents a robust, data-driven framework for decision-making in critical conditions, which can serve as a decision support tool for relief organizations.