In this paper, the major influence factors on the production capability of fully mechanized sublevel caving face are analyzed theoretically, and the artificial neural network model is established for predicting the production capability of fully mechanized sublevel caving face. The established prediction model is used to predict the production capabilities of several fully mechanized sublevel caving faces. The predicted results almost are conformable with the actual production capabilities of the caving faces, and the maximum error value of the prediction is 9.1%, proves that this model is of good reliability and adaptability in predicting the production capability of fully mechanized sublevel caving face