A marine riser, one of the most important components of offshore oil/gas transportation, needs to be designed to eliminate the risks caused by complex ocean environments, platform displacement and internal corrosion, etc. In this study, a new analytical-numerical assessment approach is proposed in order to quantitatively investigate the reliability of internally corroded risers under combined loads including axial tension and internal pressure. First, an analytical solution of the limit state function of intact risers under combined loads is obtained, which is further modified by the non-dimensional corrosion depth(d/ t) for the risers with a narrow and long corrosion defect. The relationship between d/t and limited internal pressure is obtained by finite element analysis and nonlinear regression. Through an advanced first-order reliability method(HL-RF) algorithm, reliability analysis is performed to obtain the failure probability, the reliability index and the sensitivity. These results are further verified by Monte-Carlo importance sampling. The proposed approach of reliability analysis provides an accurate and effective way to estimate the reliability of marine risers with narrow and long corrosion defects under combined loads. A marine riser, one of the most important components of offshore oil / gas transportation, needs to be designed to eliminate the risks caused by complex ocean environments, platform displacement and internal corrosion, etc. In this study, a new analytical-numerical assessment approach is proposed in order to quantitatively investigate the reliability of internally corroded risers under combined stresses including axial tension and internal pressure. First, an analytical solution of the limit state function of intact risers under combined loads is obtained, which is further modified by the non- The relationship between d / t and limited internal pressure is obtained by finite element analysis and nonlinear regression. Through an advanced first-order reliability method (HL- RF) algorithm, reliability analysis is performed to obtain the failure probability, the reliability index and the sensitivity lts are further verified by Monte-Carlo importance sampling. The proposed approach of reliability analysis provides an accurate and effective way to estimate the reliability of marine risers with narrow and long corrosion defects under combined loads.