With the estimation of the crane residual life, we could make an effective and identification to the disaster in the working process to make sure it’s safety. This method is mainly used to study the uncertainty of the crane in the working process. The research object is bridge crane, use the cycles of different crane clusters of hoisting load when different rated hoisting crane is working with different load, through the estimation pattern of the residual life of crane with the neural network technology which based on two main theoretical foundation which are Miner fatigue cumulative damage theory and the theory of linear elastic fracture mechanic theory. Then finally estimate the residual life of the crane through the derivation which based on the two main theory and the formula. A series examples show the the laborer and resource input could be saved through the former two theories. The way of calculating crane’s residual life is more practical,and the way of estimating is scientific and effective.