摘要
在实践中评定武器装备时,试验花费常常很大,本文基于Bayes方法设计了一个小子样试验方法以减少开支,提高项目的经济性。该方法可以计算出成败型产品试验的容量,同时可以分析增加试验次数对评定的影响;再进一步估计出参数区间和双方风险,为此类试验提出了一个参考流程,对评估成败型产品有一定的帮助。
Usually the test of armament evaluation costs a large sum of money. Based on Bayesian theory, a small sample test to reduce expenses and improve economical use of the project is designed. Test load can be calculated by introducing binomial distribution in Bayesian formula. The influence of the number of additional test on evaluation can be analyzed at the same time. Furthermore, the normal route of variables and the risk of both sides can be further estimated. Therefore, a reference process helpful to evaluating pass-fail products is proposed.
出处
《空军工程大学学报(自然科学版)》
CSCD
2003年第6期30-33,共4页
Journal of Air Force Engineering University(Natural Science Edition)
基金
军队科研基金资助项目
关键词
BAYES估计
试验设计
二项分布
Bayesian assessment
test design
binomial distribution