Abstract: An artificial intelligence reasoning and similar case retrieval methodology for tunnel projects characteristics, such as geographic region, monitor position, depth-width ratio, levels of surrounding rock mass and embedded depth of tunnel, and time series of arch-top deformation, is presented aimed at universality of prediction model and extrapolative accuracy. The similar case is searched and analyzed; and a forecast model is further built based on prior resource with LPG new kernel of support vector machine. The efficiencies of method are tested by predicting arch-top subsidence of a tunnel project example. Experimental results show that the approached method is suitable for the study of different tunnels or different sections of same tunnel. The average relative error of regression is 1.36% and the accuracy of extrapolation prediction is up to 97.28% within 8 steps proved case-based reasoning could achieve good prior resource and greatly help to enhance the ability of generalization of support vector machine. The better accuracy and reliability for extrapolative assessment of arch-top deformation can be utilized to server tunnel monitoring measurement

Key words: case-based reasoning, tunnel, arch-top deformation, support vector machine (SVM), extrapolative prediction