Abstract: Aiming at the phenomenon that conventional CNC machine tools only consider the failure time interval but ignore the nature of faults， a mixed Weibull distribution model combining mechanical failures with electrical failures is proposed to further improve the accuracy of the model. The maximum likelihood method is used to determine the mixed Weibull distribution model parameters， and the Pearson correlation coefficient method is applied to determine the relationship of fault times for the CNC machine tool. The K-S method is used to verify the model and obtain the mixed Weibull distribution model at last， and the mean time between failures(MTBF)of the CNC machine tool is obtained by using the reliability evaluation method.

Key words: fault analysis of machine tools, mixed Weibull distribution, maximum likelihood method, Pearson correlation coefficient, reliability evaluation