Research On Location Selection Of Emergency Material Depositories In Unexpected Situation

In recent years,with the development and reform of society,the accelerating of global industrialization's pace,and many sudden disasters have followed,causing major casualties,property losses,serious damage to infrastructure and so on.These disasters are generally characterized by a wide variety,wide spread,and strong destructiveness.China is one of the countries with the most frequent disasters in the world.In order to save more lives before the disasters,China has built several national emergency material depositories in many cities,and has also established corresponding emergency material depositories in some provinces,cities and counties.But from the overall,due to the late start of the relevant research on the construction of the emergency material depositories in China,the existing depositories have several problems,such as too small capacity,low building standards,small quantity,unreasonable layout,and so on.This is not conducive to the unified distribution of materials,is not conducive to timely delivery of materials to the disaster sites,is not conducive to saving more lives.In order to perfect the layout theory of the emergency material depositories and provide technical support for the reconstruction of the existing depositories and the location of the new depositories,based on the comprehensive consideration of natural factors,time elements,traffic elements,cost elements,social factors and so on,this paper combines the research results of predecessors and constructs a location model for emergency material depositories under uncertainty of demand.It is expected that the layout of the emergency material depositories will be more rationally planned,thereby further improving rescue efficiency and reducing unnecessary expenses.This paper studies the location of emergency material depositories before a disaster.The research is mainly carried out from the following three aspects.First,determine processing method of emergency material demand at each demand point.Due to the influence of uncertain factors,the demand for material at each demand point is disturbed within a certain range,so the layout of the depositories is affected.In order to improve the use value of a depository,the emergency material depository should always cover as many of the areas as possible in the disturbance of the demand for emergency material at the demand points.By looking up the research results of the predecessors,we find that there are four methods that are commonly used to process uncertain variables.They are robust optimization method,triangular fuzzy number method,scenario analysis method,interval number optimization method.Because before a disaster,the material demand for future disaster sites can only be predicted based on historical data.However,the probability of a large-scale disaster is extremely small,and the acquisition of a disaster data is quite difficult,making it difficult to make reliable predictions about the future.However,in order to reduce the losses caused by disasters,it is necessary to build corresponding emergency material depositories in advance in all possible demand areas.Because interval number optimization method does not require a lot of accurate data,the upper and lower bounds of the demand for emergency material at each demand point that is determined by a small amount of information can meet the requirement.So this paper chooses the interval number optimization method to get a more reasonable layout of emergency material depositories under the condition of insufficient rescue data.Second,determine the algorithm for solving the problem of the location selection of emergency material depositories under uncertain demand.Summarize several popular heuristic algorithms,compare them and select simple genetic algorithm and NSGA-? algorithm,then improve them separately.That is,EGA-PSO is proposed based on simple genetic algorithm,and improved DE-NSGA? is proposed based on NSGA-?.Finally,combine the algorithms to solve the problem of the location selection of emergency material depositories under uncertain conditions.Third,build a multi-objective location model of depositories under uncertain demand.Through the reference to the predecessors' models,the goal-planning method is used to construct a two-objective uncertain optimization model with minimum scheduling time and lowest total emergency system cost.Since the demand for emergency material at demand points is denoted by the number of intervals,the model is essentially an interval number optimization model.That is,for certain depository address in a subregion,the total scheduling time and total cost corresponding to the subregion are not determined values,but the interval numbers.In this paper,the model is firstly transformed by the interval number optimization method,and then the appropriate algorithm combination can be selected to solve the transformed model.Through the analysis of the example,we can find that the combination of improved DE-NSGA? and EGA-PSO provides better address of depositories than the combination of NSGA-? and EGA-PSO.