Study On Network Planning Of Emergency Logistics

By the viewpoint of systems analysis, the new thought that the progress of emergency logistics relies on its special network is developed. In this study, the network is partitioned to static and dynamic part definitely according to the logistics characters and the planning task in emergency situation for the first time. Thus the commodities dispatching and the cost depressing would be actualized from peacetime to disaster time.Because of the strict request of time limit and covering space, the hub-and-spoke static network was set up newly. First, the area for emergency logistics planning is divided into several zones. On the basis of this, two different models for selecting terminal node are given. Then the hub location model which is similar to p-median and facility location model is described here. As the framework of this network is fitted for the government logistics mode which emphasizes on "unified leading and graded to implement", the great and continual supply due to paroxysmal occurrence could be satisfied.The "instant demand estimate model" based upon Case Based Reasoning is introduced in emergency logistics context, assistant for decision-making. The dynamic network is clearly compartmentalized and constructed by the idea of grading, step by step, with hub and with ascription. Furthermore, the tripartite start-up mode is described according as the develop tendency of paroxysmal occurrence. In emergency situation, aid-delivery has speedy requirement, while a great deal of material for salvation should be widely raised, it becomes in a dilemma. But the new manoeuvre method well solves this problem. At last, multi-commodity dispatch model for vehicle routing problem with time deadline is formulated. Genetic Algorithms is chose here to be the solution approach. All of this simulates the actual condition of emergency logistics.