节点文献
非共轭先验分布下的贝叶斯统计质量控制研究
Bayesian Statistics Quality Control Based on Unconjugate Prior Distributions
【摘要】 贝叶斯质量控制是企业小批量、多品种生产中质量控制的有效手段。在非共轭先验分布的条件下,利用基于马尔科夫链—蒙特卡罗(MCMC)的方法可以解决贝叶斯质量控制中的后验分布确定问题,并用WinBUGS软件执行。在方差未知的正态分布多元质量特性参数控制的模型中,对生产过程中控制图的控制界限进行了预测,在有限的信息条件下,实现对质量特性参数的控制。
【Abstract】 Baysian statistics quality control is an effective tool for small-volume and multi-variety in manufacturing production.In the situation of unconjugate prior distributions for Baysian statistics quality control model,it determines the posterior distribution based on the Markov chain-Monte Carlo by using the WinBUGS software.It taking the model of multivariate normal distribution with unknown variance as an example,the posterior quality parameters were estimated and the value limits of control charts were also determined.Finally,it achieved the aim of quality control for small-volume production.
【Key words】 Bayesian statistics quality control; unconjugate prior distribution; small-volume;
- 【文献出处】 工业工程与管理 ,Industrial Engineering and Management , 编辑部邮箱 ,2012年05期
- 【分类号】O212.8;O213.1
- 【被引频次】8
- 【下载频次】271