Study On Site Selection And Route Optimization Of Earthquake Emergency Relief Materials

In recent years,the frequent occurrence of earthquake disasters has not only brought huge economic losses to the country and society,but also caused immeasurable impact on people’s life and property safety.According to the nature of earthquake emergency rescue,a site-route optimization model of emergency materials was constructed in stages to provide reference for the distribution of materials and rescue work after the earthquake.Aiming at the site-path optimization of emergency materials in the early post-earthquake period,a site-path optimization model was constructed considering the urgency of the needs of disaster points.The model considered the influence of road conditions on the site-path decision-making of emergency materials,and took the total time of emergency rescue and the urgency ranking index as the model optimization objectives.Based on the site-route optimization of emergency materials during the post-earthquake emergency rescue period,a site-route optimization model considering rescue cost was constructed,with the optimization objective of minimizing the total rescue time and total rescue cost.Considering the maximum load limit of emergency rescue vehicles,the disaster sites where the material demand exceeds the maximum load of emergency vehicles are divided into virtual large demand points and virtual small demand points.The virtual large demand points are distributed in the way of "full load direct distribution",and the virtual small demand points are distributed in the way of "circular distribution".Combined with the characteristics of the site-routing optimization model of emergency materials in the early post-earthquake and post-earthquake emergency rescue period,the corresponding genetic algorithm is designed.For the related operation of genetic algorithm,the roulette is used to select the population,and the crossover operation is carried out in the way of partial matching and crossover.For the mutation operator,the combination of various mutation modes is adopted,including: flip mutation,exchange mutation and insert mutation,which effectively ensures the diversity of the population and achieves the purpose of optimization.A numerical example is generated by means of random data generation.The Python programming language is used to randomly generate index data such as the number of victims,the number of injured and the area of road damage at the disaster point,as well as relevant data such as coordinate locations of the disaster point and candidate emergency materials distribution center and material demand at the disaster point.On this basis,The genetic algorithm used to solve the site-route optimization model of earthquake emergency materials was used to locate the candidate emergency materials distribution center and plan the vehicle route.The method of random data generation is used to generate a calculation example.The Python programming language is used to randomly generate index data such as the number of victims,the number of injured people,and the area of road damage at the disaster point,as well as the coordinate location of the disaster point,the candidate emergency materials distribution center,and the material demand at the disaster point.The method of selecting the best of the best is adopted as the principle of optimal solution selection,and the solution results are analyzed on this basis to verify the effectiveness of the model and algorithm.The convergence of the algorithm is verified according to the iteration curve of the average fitness of the population,and the effectiveness of the designed genetic algorithm is proved.In addition,the algorithm has a certain stability.In the optimal results of multiple running programs,the difference between the optimal value and the worst value of the total rescue time is 7.9%,and the difference between the optimal value and the worst value of the urgency ranking index is4.4%,which is in a relatively small range.Finally,the results of different parameter Settings are analyzed to verify the validity and rationality of the model and algorithm.