Study On Route Optimal Allocation Method Of Flood Disaster Emergency Rescue Materials

In recent years,frequent flood disasters happened all over the world,causing serious threat to the lives of people in disaster areas.Emergency relief supplies are important resource used to protect people's lives,and the timely,reasonable and fair resource optimal allocation is of great practical significance for the protection of people's lives and improving the efficiency of disaster relief,so it has been paid much attention of disaster relief agencies.In this thesis,the flood disaster as the research background,considering comprehensively many impact factors to the allocation of emergency relief supplies,methods of demand predicting,preparing and distribution of emergency relief supplies are established.The main works of the thesis is as follows:(1)Demands for emergency supplies are predicted.The relevant cases of flood disaster in recent years are collected.The data that have representative feature are chosen as the sample data.Radial basis neural network is used to train the feature data.The demand forecasting model is validated by using the test sample data,and the results of radial basis neural network and BP neural network are compared and analyzed.The validity of the proposed method is verified.The formula for supply demand estimation is established,and the demand of different types of emergency supplies in the short time after the occurrence of the flood disaster can be indirectly predicted.(2)Based on the result of supplies demand forecasting,a method of emergency supply preparing under different demand conditions is established.When the disaster happens,demand for emergency supplies in sudden situation is difficult to be satisfied by the means of utilizing emergency reserve supplies and the commandeering supplies to enterprises.In the thesis,production level of each enterprise,the cost of supply production and the time of each enterprise arriving at the disaster areas are considered mainly,and a model of emergency supply preparing is established and solved with the aim of minimizing the time and cost.(3)The traditional efficiency-fair model has been improved.In view of the different factors such as the time of emergency supply transportation,the affection degree of disaster and the different demand levels of different supplies in different areas,the model of utility loss for distribution time is established,the efficiency model for the affection degree of disaster areas and the demand level of supplies,and fairness model for the rationality of supply distribution,are established separately.According to the prior order for delivery time,supply satisfaction rate and distribution fairness,the problem of multi-objective optimization for emergency material distribution is solved by using the algorithm of variable neighborhood ant colony,with the hierarchical sequence method applied.The feasibility and rationality of models and methods proposed in this thesis are verified by case analysis.