Abstract: In response to the rational allocation of materials for urban flood and waterlogging, a loss-driven demand urgency calculation method was proposed, and a refined optimization configuration framework of materials for urban flood and waterlogging control was established. Taking Zhengzhou City as a study area, the multi-objective genetic algorithm was used to calculate the configuration results under the 'once in one hundred years' rainfall and flood return period, and compared the differences in configuration performance between the proposed loss-driven method and equal priority, TOPIS, and clustering classification methods. The results show that the proposed loss-driven material allocation framework achieves refined configuration of sandbags and drainage equipment, with a configuration result loss of 34.97%~56.58% lower than other methods, improving the effectiveness of limited material allocation. These results indicate that the loss-driven method approach can better reflect the actual needs of hazard bearing bodies. The research results can provide reference for the refined prevention and control of urban flood and waterlogging.