摘要
复杂系统可靠度置信下限的寻求，一般说来，由于系统的构成非常复杂，现有的许多统计方法大多只能给出近似解．对于由两个寿命分别服从指数分布的部件构成的并联系统，本文利用样本空间排序法，给出了系统可靠度置信下限的精确解，并提出了行之有效的算法。
The problem to get confidence lower limit for the reliability of a complex system is much difficult but very important.In pratical applications,we can test the different components that make up the system respectively when it is being developed or berore it is put into use.With the experimental data,one can get a rudimental estimator also supply technical support to improve the system's performance.Thus,it is of great importance.However,in most cases,current statistical methods can only give an approximate situation to the confidence lower limit because of the complex structure of the system.To the system that consists of two parallel components ,both of which follow exponential distribution whose mean parameter may be different,the situation is much like this.But to this article,a exact solution with sample space ordering is given.so is a very feasible all orithm.Moreover,some results can be extented to systems that are composed of more than two paralled components.
关键词
并联系统
可靠度
置信下限
样本空间排序法
parallel system
reliability
confidence lower limit
sample space ordering method
