Study On Relief Supplies Allocation Models Based On Characteristics Of Disaster Situation Information

Emergency refers to such sudden incident that has caused or may cause heavy casualties, property damage, ecological destruction and serious harm to society and public safety. Emergencies include natural disasters, public health events, and the multitude events in the society and so on. As one of the most severe affected area in the world, China suffers natural disasters such as earthquakes, floods, typhoons which has caused serious losses every year. After the occurrence of emergency, many victims lack clothing, food, shelter, vehicles, daily necessities and medicine. It is necessary to provide timely relief supplies to victims to reduce the loss of life and property. So this paper studies how to distribute relief supplies in effectively and fairly manner to meet the needs of disaster victims and reduce the loss of them.This paper is based on such theories as the logistics system theory, the regional disaster system theory, emergency management theory and statistical decision theory. It analyzes laws of disaster evolution and characteristics of disaster situation information in the three phases of emergency management, which are emergency preparedness, emergency response and emergency recovery. And three models for relief supplies allocation are established based on distinct characteristics of disaster situation information, decision problems, decision features and decision criteria in the three phases of emergency management to mainly reduce the loss of victims. The algorithms for three models are designed respectively and examples are given to show their application. It also analyzes the characteristics of three models and the relationship among different variables. The main topics of this paper are as following.(1) Characteristics of disaster situation information in the three phases of emergency management are discussed. The dynamic and uncertain properties of disaster situation information including hazard inducing environment, hazard inducing factors, hazard bearing body, disaster prevention and reduction in the three phases of emergency management are analyzed.(2) The decision-making frame and modeling methods for relief supplies allocation problem in the three phases of emergency management are analyzed. It is proposed to establish models based on the properties of disaster situation information. And specific modeling methods for relief supplies allocation problem are proposed based on properties of disaster situation information, decision problems, decision features and decision criteria in the three phases of emergency management.(3) In the emergency preparedness phase, a stochastic programming model for relief supplies allocation is established based on disaster scenarios. To construct proper disaster scenarios, a two-stage construction method is proposed through classification and composition of disaster situation information which is taken as scenario elements. Then a two-stage stochastic programming model is established to solve emergency facilities location problem, relief supplies inventory problem and relief supplies pre-allocation in different disaster scenarios. The model is suitable for emergency preparedness for the disaster situation information in this phase is static and uncertain. An effective algorithm is designed and a numerical example is given based on the Sichuan earthquake zone.(4) In the emergency response phase, a Bayesian sequential decision-making model for relief supplies allocation is established. To determine the decision time and decision-making programs in a framework, a Bayesian sequential decision-making model for multiple depots selection is established. The model is suitable for emergency response because the relief supplies allocation decision in this phase is irreversible and the disaster situation information in this phase is dynamic and uncertain. Through minimizing total loss and emergency response time in each decision period, such variables as maximal observation period for disaster situation information, optimal decision time of allocation program, optimal allocation amount, the selected depots and their supply amount in each decision period are determined. The stochastic simulation and Bayesian analysis method are proposed to solve the model, and a numerical example with simulation is given.(5) In the emergency recovery phase, a set of efficiency-equity models for relief supplies allocation are established. In this phase the fairness of relief supplies allocation is particularly important and prominent. Firstly four types of efficiency- equity models for resource allocation are summarized. Secondly, an efficiency model without any equity object or equity constraint is established to solve relief supplies allocation problem with one depot, multiple affected areas and multiple relief supplies. This efficiency model is taken as the basic model for other efficiency-equity models. Then an efficiency model with nonlinear object function, a model with equity constraints, an efficiency-equity model with multiple objects are established respectively. All these models basically use identical object functions or constraints to facilitate comparisons.This paper may contribute to relief supplies allocation model construction in theory, explore a path for cross-over study and integration research of disaster science, management science, information science and statistical decision analysis in emergency management field, and deepen scientific understanding of laws of emergency management. In practice, this study may be conducive to strengthening the central and local relief supplies reserve network construction in China, enhancing the transport capabilities to support relief supplies, and establishing scientific and reasonable relief supplies allocation system. The ultimate goal of this paper is to help governments and other relief organizations to prevent and alleviate human suffering, protect life and health and ensure respect for the human being.