随机飘移下变抽样区间图的设计

利用质量控制图监督生产过程时,通常每隔固定时间从过程中抽取固定容量的样本,具有动态特性的控制图比传统静态图能更快地发现过程变化.假定过程均值的飘移服从某种连续概率分布下设计了变抽样区间图,计算了其统计性质和质量损失,并与传统休哈特图作了比较.结果显示,从整体表现上,变抽样区间图显著地优于休哈特图.本文的目标是从统计性质和经济成本2个角度显示变抽样区间图相对于传统休哈特图的优势所在,帮助决策者选择表现更好的控制图.

Design of Variable Sampling Intervals (x) Control Charts Based on Random Process Shifts

Using control chart to monitor a production process, usual practice is to take the samples at fixed time. Recent research shows that adaptive control charts is more efficient than the static charts. Based on the continuous probability distribution of the random process mean shifts, this paper designed variable sampling intervals(VSI)(x) control charts and calculated its statistical and economic properties. The results of the comparison with the fixed sampling intervals (FSI)(x) charts show that these charts are superior to the Shewhart control charts in view of overall performance. The purpose of this paper is to show the advantages of VSI charts relative to FSI charts from the two points of statistical properties and economic cost, and helps the decision-makers choose better control charts.