The tube hydroforming process has undergone extremely rapid development, especially in applications for the automobile industry. In-process variation of forming parameters such as material properties, friction conditions and part geometries will directly affect the quality of forming response by causing variation in the process output. To ensure a reliable hydroforming process at the design stage, applying robust design methodologies becomes crucial to the success of the resulting process. The reliability of the tube hydroforming process based on the tube wall thickness thinning ratio is studied in this paper. In order to improve the reliability of the process, the Taguchi method, which is capable of evaluating the effects of process variables on both the mean and variance of process output, is used to determine the optimal forming parameters for minimizing the variation and average value of the thinning ratio. The Taguchi method is applied to design experimental arrays which incorporate design (i.e., controllable) parameters and noise (i.e., non-controllable) parameters. Finite element simulation is used to analyze the virtual experiments according to the experimental arrays. Through statistical analysis, the influence of each design parameter on both the mean and variance of the thinning ratio is obtained, and is used to find the optimal combination of design parameters for minimum thinning ratio, minimum variance of thinning ratio, and maximum expected process reliability. A cross-extrusion hydroformed tube is employed as an example to illustrate the effectiveness of this approach.