An optimized GM (1,1) model is designed for the bridge static load test under the condition of extremely incomplete data. Three pre neural network modules are used to replace the differential fitting process in the traditional model, and the regression ability of neural network is used to optimize it. One post neural network module is used to replace the data restoration process of traditional model with one data solving fuzzy module. Other data processing methods are based on the traditional model. Through simulation, it is found that the standard deviation of the optimization model is significantly lower than that of the traditional model under different bridge design scale, which proves that the optimization model has a significant calculation force improvement.