Abstract: Evaluation of seismic stability of rock slopes is the important foundation for prevention of seismic unstability of rock and soil slopes. In view of the multi-source and fuzzy characteristics in evaluation process, by using engineering variable fuzzy set theory for reference, fuzzy variable evaluation model is applied to evaluate the seismic stability of rock slopes. First, by comprehensive consideration of the rock slopes internal structure and external natural conditions, six parameters including the characteristics of the geotechnical body, the characteristics of the neotectonics, slope height, slope angle, the annual average precipitation, site seismic intensity are selected as the indices for evaluation index system and established rating standard, then the weights of evaluation factors in the model are obtained from the index practical data discreteness by using improved entropy weight method; finally, by changing the fuzzy variable evaluation model parameters to evaluated seismic stability of rock slopes by linear and nonlinear, and obtained stability level by mean value. The method has been applied to natural slope and cutting slope instances. The results show that the fuzzy variable evaluation model results are reasonable and objective; and they have better reliability and stability so as to provide a new research method and idea for seismic stability of rock slopes.

Key words: rock slopes, earthquake, slope stability, fuzzy set, relative membership degree, evaluation, information entropy