The evaluation index of coal mine rock burst has some uncertainties such as diversity and randomness, the scientific and accurate index weight can not be determined, and the evaluation result always deviates from the reality. In order to solve the above problems, an evaluation index system of coal mine rock burst was constructed, which takes geological conditions, safety management, mining technology, and other factors as criterion layer, and 24 factors such as mining depth and mechanical properties of coal and rock as index layer. The weights calculated by “AHP and Entropy weight method” are combined, and the Lagrange function is introduced to construct the optimal decision-making model. Through the Euclidean distance function, the preference coefficient is obtained, and the index combination weight is obtained. The evaluation model of rock burst in CW-TOPSIS coal mine is constructed, and the grade of rock burst is determined. The model was used to judge the rockburst grade of a coal mine in Long Coal Mining Group, and the grade is “weak impact”, and the calculated index combination weight was sorted, from large to small: mechanical properties of coal and rock > mining depth > roof strata structure characteristics > mining face advancing to goaf, fault time > roof management method. The evaluation result is consistent with the actual situation of the coal mine. Compared with the traditional evaluation method including, the comprehensive index method, the CW-TOPSIS evaluation model is more accurate, more precise, and more practical.