In order to solve the problem that the standard Rayleigh distribution is difficult to describe skewness, fat tail and non-monotonic hazard rate in complex life data, a Kumaraswamy Marshall-Olkin Rayleigh (KwMO-R) distribution was constructed and used for system reliability analysis. The distribution function, density function, quantile, moment, hazard rate function and stress-strength reliability expression are derived by extending the Rayleigh base model with Kumaraswamy morphological parameter and Marshall-Olkin tail adjustment parameter. Maximum likelihood estimation and Monte Carlo simulation are used to test the performance of parameter estimation. The results show that when the sample size increases from 50 to 500, the MSE of parameter b decreases from 2.2464 to 0.4229, and the deviation of scale parameter σ decreases to -0.0052. In the real data fitting, the AIC of failure data and survival time data of electronic equipment are 236.42 and 185.64, respectively, and the K-S statistics are 0.0821 and 0.0765, respectively, which are better than the competitive model, indicating that KwMO-R distribution has good applicability in the life modeling and reliability evaluation of complex systems.
