The unknown subsurface conditions in tunnelling projects have led to their management with many uncertainties. From these uncertainties, we can mention the geological condition of the tunnel route and the time and costs required for construction. In order to significantly reduce these uncertainties, techniques that have a high predictive power must be used. For this purpose, in this study, an autoregressive model was used to reduce the uncertainties related to geology and construction time and cost in tunnelling projects. A comparison between the predicted results and the actual values through several statistical indices showed the high-performance prediction of the autoregressive model in the prediction of tunnel resources. Also, three input parameters affecting tunnel construction time and costs, such as RQD, groundwater, and RMR, were considered. The sensitivity analysis of these parameters on the time and cost of tunnelling projects was investigated through mutual information test (MIT). The groundwater was the most effective parameter on the tunnel's time and cost.