Abstract: The main factors influencing subsidence coefficient were comprehensively analyzed. Then the model to calculate subsidence coefficient was established by applying the theory of artificial neural network (ANN). A large amount of data of observation stations was used as learning and training samples to train and test the artificial neural network model. The calculated results of the ANN model and the observed values were compared and analyzed. The results show that it is comparatively precise to calculate the subsidence coefficient of ground surface by ANN technology.