A reliability-based structural design method for infrared cryostat is put forward to obtain a design result with a quantitative reliability index. In this method, the reliability analysis is performed by integrating the finite element software ANSYS (functioning as the deterministic analyzer) with the probabilistic engineering analysis software NESSUS (functioning as the probabilistic analyzer), in which design parameters are treated as random variables. The probability of failure and probabilistic sensitivity level of design parameters are calculated, which would provide a quantitative judgment about whether there should be a redesign and which parameters should be modified in the redesign. As an example to illustrate this method, the IR focal plane displacement induced by random vibration has been analyzed in this paper. The probability of the focal plane displacement value exceeding a critical value is calculated and the focal plane stability reliability level has been increased from 82% to 99.9999%. The method can be widely applicable in the fields where uncertainty is assumed to have a significant impact on the structural response.