Abstract
This article addresses an emergency shelter and medical network design problem by integrating evacuation and medical service activities and considering diurnal population shifts to respond to large-scale natural disasters in urban areas. A multi-objective mixed-integer programming model that incorporates the characteristics of diurnal population shifts is developed to determine the configuration of the integrated emergency shelter and medical network. An accelerated Benders decomposition algorithm is then devised to solve large-scale problems in reasonable time. A realistic case study on the Xuhui District of Shanghai City in China and extensive numerical experiments are presented to demonstrate the effectiveness of the proposed model and solution method. Computational results suggest that more emergency shelters and emergency medical centers should be established when accounting for diurnal population shifts than when diurnal population shifts are not considered. The accelerated Benders decomposition algorithm is significantly more time efficient as compared with the CPLEX solver.
Acknowledgements
The authors thank the two anonymous referees and the department editors for their valuable comments and suggestions, which improved the quality of this article.
Notes on contributors
Qing-Mi Hu is a Ph.D. candidate in the Antai College of Economics and Management, Shanghai Jiao Tong University, China. His research interests include humanitarian logistics, second-order cone programming, and network design and optimization.
Laijun Zhao is a professor in the Sino-US Global Logistics Institute, Shanghai Jiao Tong University, China. His research interests include safety and environmental management, and logistics management. He has published more than 100 articles at home and abroad. He is a member of the Systems Engineering Society of China.
Huiyong Li is a postdoctoral researcher in the Sino-US Global Logistics Institute, Shanghai Jiao Tong University, China. His research interests include location theory, city logistics, and large-scale network optimization.
Rongbing Huang is a professor in the School of Administrative Studies, York University, Canada. His research interests include location theory, combinatorial optimization, service management and online auction.