Direct Assessment Method for Series System Reliability Based on Small Sample Data
A direct assessment method for series system reliability using small-sample data is proposed in this paper,which can assess the reli-ability confidence limit of a series system based on the reliability confidence limits of its subsystems.This approach overcomes the limitations of traditional methods such as the LM and MML,which require the conversion of continuous subsystem data into less informative binomial data.This conversion often results in low accuracy or a dangerous estimate in reliability assessment.On this basis,a reliability assessment method for a series system with normal-distributed subsystems is established.This method is capable of assessing the series system reliability based on the small-sample metrological data from the subsystems.Extensive Monte Carlo simulations and examples demonstrate that the proposed method not only meets the confidence requirement but also offers higher assessment accuracy compared to traditional methods.Furthermore,the pro-posed methodology applies to common continuous distributions,including exponential,Weibull,and normal distributions.
Series systemSmall sampleReliability assessmentConfidence limitSuccess-failure testMetrological test
NETL
NSTL
万方数据
