Abstract: In order to evaluate the stability grade of rock slope accurately,the multidimensional cloud model theory is proposed to evaluate the stability grade of rock slope according to the characteristics of randomness and fuzziness,which are influenced by many evaluation factors.From the five indexes of topography,rock mass environment,artificial excavation,earthquake,meteorology and hydrology,14 factors are selected as the evaluation indexes of rock slope stability.Based on the cloud theory,the digital characteristics of the cloud model are obtained by using the index approximation method,and the weight of each evaluation index is determined by combining with the improved critical method to generate a multidimensional cloud model for the evaluation of rock slope stability grade.By using this model,the measured values of 8 groups of rock slopes in xichangtaihe iron mine of Chongqing Iron and Steel Group are input to the comprehensive certainty degree of each standard slope stability grade,and the slope stability grade is determined according to the principle of maximum certainty degree.The evaluation results are compared with onedimensional cloud model,fuzzy evaluation method,CSMR method and the actual state of slope.The results show that the multidimensional cloud model is more efficient and accurate in the evaluation of rock slope stability.

Key words: stability assessment, multidimensional cloud model, improved CRITIC method, certainty