Study Of Refuge Selection And The Application Of Evacuation Strategy

With the rapid economic development of China,all kinds of natural and man-made disasters have occurred frequently,and the impact of the outbreak is also expanding.In this case,it is necessary to do disaster prevention work and disaster preparedness.In view of the present situation of our society,the thesis studies the effective deployment of urban shelter site selection and efficient emergency evacuation strategy,by considering them together,which is a useful way to prevent natural disasters bringing us against.On the one hand,this thesis discusses state of the art on shelter site selection and emergency evacuation strategy,and gives a literature review on the two topics.Then,in order to study the location problem of the evacuation site,the thesis selects and analyzes the evaluation criteria of the suitability of the evacuation sites according to the characteristics of disasters.A set of factors that need to be considered are determined when selecting sites.The evaluation strategy of site adaptability is discussed.In order to improve the quality of shelter site selection,this thesis proposes an integrated evaluation model based on the traditional evaluation models,which combines the statistics and social science to construct a new evaluation model.Our experiments are conducted on the data from the surrounding environment of a university.Detailed results are reported in this thesis.Using the site selection model,we can select the appropriate shelters from the candidate sites,which is a necessary preparation for disasters.On the other hand,after a disaster happens,evacuation strategy is important.This thesis studies the evacuation strategy according to the candidate shelter sites.When planning a evacuation strategy,three algorithms are considered.They are genetic algorithm,particle swarm algorithm and simulated annealing algorithm.The total time required to evacuate the affected population when changing the number of selected evacuation paths is computed according to this three algorithms.Finally,the most efficient scheme is the fast one.It can be seen from the experimental results that the genetic algorithm is better than the particle swarm algorithm and the simulated annealing algorithm.In addition,this thesis designs and implies an evacuation system based on Baidu Map API,including the shelter site selection part and the emergency evacuation strategy part.