Abstract: Based on the basic theory of ideal point method, considering the mechanism of rock burst, a model for predicting and classifying rockbursts is established. Five factors are selected as the evaluation indexes, including the ratio of the maximum main stress of the surrounding cave wall to uniaxial compressive strength of rock, the ratio of the maximum tangential stress of the surrounding cave wall to uniaxial compressive strength of rock, the ratio of compressive to tensile strength of rock, the elastic energy index of rock and the intactness index of rock mass. The weight coefficients of these evaluation indexes are determined by the AHP and information entropy theory. The statistics-related projects and the engineering case analyses show the feasibility and applicability of the proposed model. A computer evaluation interface for rockburst classification is developed based on the proposed model. This model and computer interface can be referred to for other similar engineering practice.