Abstract: It is of great significance to study the location theory for emergency supply stockpile in response to frequently occurring large-scale emergency events. To overcome the limitation of the existing location theory, a location model is put forward by combining fuzzy TOPSIS method and multi-level coverage for emergency supply stockpile. By analyzing the factors that affect the emergency supply stockpile location, an indicator system of facility location is established. The fuzzy TOPSIS method is used to select the primary candidate facility set. Then, based on the idea of “backup coverage” and “partial coverage”, a multi-level coverage location model is constructed for determining the best solution. A genetic algorithm procedure based on Matlab 70 is developed to solve the model. Numerical experiment is conducted to test the proposed method and result shows that it is effective.