To analyze the reliability of the machining center, the failure data of 50 foreign highend machining centers were recorded by time truncated testing. Through analyzing the failure part, failure mode and their criticality, the weak parts of the machining center were found out. To analyze the failure time and establish the distribution of the time between failures, the maximum likelihood method was utilized to estimate the parameters of several distributions. Then, the Weibull distribution was selected, and genetic algorithm was utilized to estimate the location parameter. And the distribution was inspected by KolmogorovSmirnov test. Eventually, the mean time between failures of the whole machining center was estimated. The result shows that the correlation coefficient method based on genetic algorithm can estimate the three parameter Weibull distribution model effectively, and this method can provide a reference for other scholars.