Research On A Location Model Of Emergency Distribution Center Based On Improved NSGA-? Algorithm

Due to the enormous losses caused by the increasing public emergency events,the facility location has become a research focus in emergency management field.However,with lacking of scientific planning,an imbalance of the facility location still exists in regional distribution.Therefore,it is urgent to give an in-depth analysis on the location to large-scale emergency distribution center.Firstly,the thesis studys the dynamic multi-objective location model.A dynamic multi-objective location model is constructed by considering the fixed emergency centers which is no longer sufficient to deal with the needs of the affected areas in a serve crisis.Three objectives are considered in the model,the first objective maximizes the total utility of relief supplies,the second objective minimizes the number of temporary facilities needed to operate,and the third objective maximizes the satisfaction for all demanding points.Secondly,the thesis studys the algorithm for solving the multi-objective optimization problem.The NSGA-? still has some disadvantages.On the one hand,it is easy to produce duplicate individuals in the iterative process.On the other hand,the convergence of the algorithm is greatly reduced with the increase of variable's dimension.In this paper,we propose an improved NSGA-? to solve the optimization problem.In the first stage,a new mutation operator is designed which the variant individuals are generated from their neighborhood instead of randomly generating individuals.Through this mutation operator,the number of duplicate individuals is reduced.In the second stage,the idea of tabu search on the elitism strategy of NSGA-? is introduced,which improves the diversity of the offspring individuals and enhances local search capabilities of the algorithm.Finally,the thesis designs an empirical analysis to verify the superiority of the NSGA-?-TS algorithm.The computational experiments are divided into two sections.In the first procedure the numerical experiment is constructed by the classical functions ZDT1,ZDT2,ZDT3 and DTLZ2.The results show that the proposed algorithm generates more exact Pareto Front,the convergence and uniformity of the proposed algorithm are better than the NSGA-? and MOEA/D.In the second procedure,the simulation experiment is constructed in emergency management.The results show that the proposed algorithm is more reasonable than the traditional algorithm NSGA-? and MOEA/D in terms of the three objectives.The improved NSGA-? algorithm is proved that it has excellent feasibility of applying to the multi-objective location problem in the sudden disaster crisis and emergency management.