A Combined Two-Stage Clustering and Sequential Protective Submatrix Algorithms in Emergency Facility Coverage Sets
نویسندگان
1 Assistant Professor, Department of Industrial Engineering, Faculty of Mechanical Engineering, Semnan University, Semnan, Iran.
2 Assistant Professor, Department of Logistics and Supply Chain Management, Faculty of Humanities, Imam Hossein Guard Training and Officer University, Tehran, Iran.
3 M.Sc., Department of Industrial Engineering, University of Science and Culture, Tehran, Iran.
doi
10.22059/aie.2025.380198.1944چکیده
A major part of crisis management is logistics. Setting up an effective logistics system during emergencies and reducing damage is essential. This study first introduces a mathematical model for emergency logistics. Then, a hybrid metaheuristic algorithm is proposed to optimize how demand in affected areas is met based on this model. The focus is on emergency logistics with the goal of reducing costs and improving coverage of people in need. It also presents a model for locating distribution and relief centers using a two-stage clustering approach to form binary clusters from a distance matrix, where each cluster pair represents which distribution centers serve which demand areas. In contrast, our approach consistently matches the optimal solutions faster than GAMS, as detailed in Table 11. Notably, for larger instances (Ins9–10), OPSM reduces runtime by 30–50% while still achieving optimal solutions. This efficiency is particularly evident when GAMS fails to reach optimality, as our method outperforms its best-found solutions. Findings shows that, proposed algorithm is efficient and suitable for optimizing and solving coverage set problems.
