The effects of varying distribution type of random variables and environmental, operational, and design random variables influenced by a shock wave caused from various origins on the failure probability are systematically investigated using the first order reliability method (FORM) for buried pipeline. It is found that the failure probability of the buried pipeline increases with faster P-wave velocity and slower S-wave velocity. The failure probability is estimated to be the largest for the Weibull distribution and the smallest for the lognormal distribution. A set of similar values of the failure probability for the normal distribution and lognormal distribution are noted. The normalized margin is defined and estimated. Furthermore, the normalized margin is used to predict the failure probability simply by using the fitting lines between the failure probability and the normalized margin.