In order to improve the efficiency of emergency rescue, considering the characteristics of the two-level deployment of emergency materials and the demand of victims, this paper proposes an optimization model for emergency materials allocation scheme considering multi-material and multi-level distribution of emergency vehicles in the emergency distribution center. Combining the characteristics of the research model, the genetic algorithm based on real coding is developed to solve the model. A specific example is provided to analyze whether the emergency vehicle that has reached the distribution center needs to wait until the next batch of emergency materials arrives or not. This new decision-making method is compared with two traditional methods to verify the validity and feasibility of the model. The results show that the proposed optimization model combines the advantages of the two traditional methods, and reduces the transportation costs while improving the overall satisfaction of the victims on the amount and time of delivery of emergency materials.