中文摘要：

在制造过程中,对产品的不合格品数进行监控时,通常选用计数性控制图-np图,它是基于过程服从二项分布建立的,一般对于过程中出现的较大波动效果明显。为了提高控制图对不合格品数较小波动的监控效果,本文设计了产品不合格品数服从二项分布的EWMA控制图。提出可变抽样区间的二项EWMA控制图,并采用马可夫链法计算其平均报警时间。对固定抽样区间以及可变抽样区间二项EWMA控制图对比研究,表明当过程失控时,可变抽样区间二项EWMA控制图具有较小的失控平均报警时间,能够迅速监测出过程中的异常波动,明显优于固定抽样区间的二项EWMA控制图。

英文摘要：

When the quality characteristic is under binomial distribution, the np control chart can established to mainly monitoring the products unqualified number of production process. When the process has the small shift, the exponential weighted moving average （EWMA） control charts have the better monitoring efficiency than traditional shewhart control chart. EWMA control chart under binomial distribution for monitoring the products unqualified number is proposed in this paper. And EWMA control chart using variable sampling intervals policy （VSI） under binomial distributions is constructed also. The average time to signal is calculated using the Markov chain method. The computing results show that the VSI binomial variable EWMA control chart is the more efficient in detecting shifts than the fixed sampling interval （FSI） binomial variable EWMA control chart.