Article Details

Abstract

In order to improve the accuracy of post-earthquake emergency material demand forecasting, this paper proposes a feature weight allocation method based on improved Case-Based Reasoning (CBR) algorithm, and constructs a post-earthquake emergency material demand forecasting model based on safety stock theory. The model is based on seven earthquake disaster indicators such as magnitude, focal depth, earthquake occurrence time, population density, number of house collapses, seismic fortification intensity, and seismic intensity. It can accurately forecast the demand for various types of emergency materials after the earthquake. The experimental results show that the Mean Relative Error of the forecast value obtained by the forecasting model optimized by game theory-improved genetic algorithm (SAGA) and Analytic Hierarchy Process (AHP) algorithm are 89.57% and 87.51% lower than that obtained by GA algorithm optimization and SAGA algorithm optimization, respectively. This shows that the model can provide strong technical support for the efficient allocation of post-earthquake emergency materials.

Keywords

post-earthquake emergency material; demand forecasting; game theory; SAGA algorithm; Case-Based Reasoning algorithm

References

1. [1] ZHOU L，WU X H，XU Z S，et al. Emergency Decision Making for Natural Disasters: An Overview–Science Direct [J].International Journal of Disaster Risk Reduction，2018，27：567-576.
2. [2] ZARGHAMI S A，DUMRAK J. A System Dynamics Model for Social Vulnerability to Natural Disasters: Disaster Risk Assessment of an Australian City [J].International Journal of Disaster Risk Reduction，2021，60：102258.
3. [3] HUANG D，WANG S A，LIU Z Y. A Systematic Review of Prediction Methods for Emergency Management [J].International Journal of Disaster Risk Reduction，2021，62：102412.
4. [4] Li Yunsong, Zhang Zhisheng Short term forecasting of multiple loads in integrated energy systems using Transformer graph neural networks considering comprehensive demand response [J]. Journal of Electrical Engineering, 2024, 39 (19): 6119-6128
5. [5] Kang Lijun Research on the Application of Particle Swarm Optimization BP Neural Network in Emergency Material Demand Prediction [D]. Lanzhou: Lanzhou Jiaotong University, 2013
6. [6] Cheng Qiongqiong Prediction of earthquake emergency material demand based on GA and BP neural network [D]. Chengdu: Southwest University of Finance and Economics, 2016
7. [7] Wang Qingrong, Ma Chenkun Emergency material prediction based on case-based consumption reasoning [J]. Computer Engineering and Applications, 2021, 57 (22): 281-287
8. [8] LI R. Research on Emergency Supplies Demand Forecasting Model [C]//2010 the 2nd International Conference on Industrial Mechatronics and Automation.wuhan：IEEE，2010：244-247.
9. [9] ZHANG H Q，XU J Y. Research on Emergency Material Demand Forecasting Model in Disaster Based on MLR-CBR [C]//International Conference of Logistics Engineering & Management.[S.I],2010，387：2888-2894.
10. [10] ZHOU M，LIU X B，CHEN Y W，et al. Assignment of Attribute Weights With Belief Distributions for MADM Under Uncertainties [J].Knowledge-Based Systems，2020，189：105110.
11. [11] PHAN A V，NGUYEN M L，BUI L T. Feature Weighting and SVM Parameters Optimization Based on Genetic Algorithms for Classification Problems [J].Applied Intelligence，2017，46：455-469.
12. [12] Tong Wenlin, Chen Dewang, Huang Yunhu, etc Fuzzy system optimization method combining simulated annealing and rule reduction [J]. Computer Engineering and Applications, 2021, 57 (16): 142-150
13. [13] TASKIN S，JR E J L. A Bayesian Decision Model with Hurricane Forecast Updates for Emergency Supplies Inventory Management [J].Operational Research for Emergency Planning in Healthcare，2016，1：330-352.
14. [14] SHEU J B. An emergency logistics distribution approach for quick response to urgent relief demand in disasters [J].Transportation Research Part E Logistics & Transportation Review，2007，43（6）：687-709.

About the authors

Zhou Zhanzan (1995-), male, Han ethnicity, born in Yueyang, Hunan, is an engineer and currently pursuing a master's degree in deep learning and data processing.