Abstract: The patrol inspection of oil and gas pipelines has the characteristics of randomness and fuzziness. In order to effectively assess the risk level of patrol inspection of buried long-distance oil and gas pipelines in tunnels, a safety risk assessment model for patrol inspection of pipelines was constructed by introducing the cloud theory and combination weighting method. Herein, study was performed on Ili?Guozigou Pipeline of PipeChina in a tunnel group. Specifically, 29 risk factors in 6 aspects, i.e., tunnel, road network, environment, human behavior, state of object and management level, were used as the assessment indicators based on the hazard theory. Meanwhile, the weight coefficients of each indicator were determined using the improved G2 method and CRITIC (Criteria Importance Though Intercrieria Correlation) method based on the combination of Euclidean distance. The characteristic parameters of the cloud model were calculated and the cloud image was generated with Matlab software according to the classification standards of each indicator. Besides, the indicator membership degree and the comprehensive membership degree under different risk levels were determined by the forward cloud generator. The results show that the combination weighting technique based on Euclidean distance can not only reduce the subjective uncertainty of decision-makers, but also eliminate the objective errors between data. In addition, the assessment results are in good agreement with the field engineering evaluation results. The assessment model constructed has good scientificity and applicability, and is capable of providing new ideas for the risk assessment of patrol inspection of the buried long-distance oil and gas pipelines in tunnels.