Study On Emergency Relief Optimal Delivery Considering Victims’ Psychological Risk Perception And Dynamic Scheduling

In order to reduce the loss that caused by the frequent unconventional emergencies, we need to deal with the coordination issue between emergency relief and vechile scheduling. Therefore, the distribution and scheduling problem of emergency relief are studied in this thesis. After unconventional emergencies breakout, disaster areas will receive emergency relief from the whole country. Because the damage degree and the number of victim in every affected area are different, the demand amount of relief is also not the same. That is very necessary and meaningful to distubte relief according to the different situation of affected areas. Also, the victims’psychological risk perception is complicated, and every victim wants to get emergency relief as soon as possible. The occurrence, development and evolution mechanism of unconventional emergencies are indeterminate. We need to schedule emergency relief dynamically according to real-time disaster information in order to maximize social and economic benefits.The main contents of this thesis are as follows:(1) Collecting the large amounts of data, and indicating the importance of emergency relief scheduling, as well as the principles in actual operation process. We analyze the detailed procedures of emergency relief scheduling in "5·12" Wenchuan earthquake, and sum up the characteristics of emergency relief scheduling. The key factors of emergency relief scheduling are digged out. We point out that the victims’psychological risk perception should be put into consideration and emergency relief should be real-time and dynamic scheduled.(2) Integrating the behavioral science theory into emergency relief optimal scheduling after emergency events, and proposing that the victims’psychological factors should been taken into account to make decisions about emergency relief scheduling. We depicte the victims’psychological risk perception with Prospect Theory, define and select the victims’ psychological expectation time as a time reference point, and get out the risk perception curve and function model. We formulate a mixed integer programming model to minimize victims’ psychological risk perception and dissatisfaction degree of emergency relief. The corresponding multilayer search algorithm is designed. The validity of this model and algorithm is veried through numerical experiment.(3) Focusing on emergency relief dynamic scheduling after unconventional emergencies. We quantify the loss of victims in terms of unmet needs, and formulate a mixed integer programming model in order to minimize victims’loss and vehicle scheduling expense. The main decisions are determining the delivery routes and allocating the emergency relief from relief distribution center (RDC) to save points without enough transport capacity. We use the hierarchical thinking to reduce solution space and improve the coding method in genetic algorithm. Experimental results show the dynamic scheduling scheme of emergency relief, and determine the optimal vehicle quantity in RDC through comparative analysis.This thesis focuses the distribution and scheduling problem of unconventional emergencies, and put the behavioral science and dynamic scheduling into the emergency relief optimal distribution model. Decision makers can make emergency relief distribution and scheduling plan according to the degree of victims’psychological risk perception. With the occurrence, development and evolution of unconventional emergencies, the dynamic scheduling scheme of emergency relief can be made in order to reduce the losses of the state and victims, as well as provide support for emergency rescue.