Abstract: In order to ensure the emergency supplies meet the needs of disaster-hit areas to the maximum extent and establish a more reasonable reserve strategy, factors such as the difference in emergency supplies demand of different disaster categories and levels,supplies modularization and the level of repository participating in the rescue were comprehensively considered and a multi-objective emergency supplies reserve model was built, with maximizing the degree of emergency supplies satisfaction and minimizing the costs of reserve and transportation as the objective functions. The model was solved by the non-dominated sorting genetic algorithm II and the dynamic neighbor table deletion hybrid optimization algorithm and analyzed by data simulation,so the types and optimum quantities of emergency supplies reserved in the repositories of each level were calculated. The results show that the hybrid optimization algorithm can effectively solve the multi-objective emergency supplies reserve model,and the emergency supplies reserve scheme can guarantee the rationality of the reserve.

Key words: supplies modularity, different level of repositories, emergency supplies reserve, non-dominated sorting genetic algorithm II, dynamic neighbor table deletion hybrid optimization algorithm