In order to accurately predict the material demand in the transitional resettlement stage of earthquakes and improve the efficiency and accuracy of emergency material mobilization，the factors that have a great impact on the number of resettled population were determined based on the historical seismic data in China. A prediction model of the resettled population based on GWO-BP was established，which combined with the quantitative relationship between the population and emergency supplies，to predict the material demand in the transitional resettlement stage after the earthquake. The experimental results show that the GWO-BP neural network model exhibits high accuracy and stability in predicting the number of relocated populations，and can effectively predict the number of relocated populations in disaster areas，thereby calculating the corresponding material demand. GWO-BP neural network model has a certain application value in predicting material demand in post-earthquake transitional resettlement stage，and can provide a reference for the decision-making of emergency material procurement after the earthquake.