期刊
QUALITY AND RELIABILITY ENGINEERING INTERNATIONAL
卷 37, 期 2, 页码 502-526出版社
WILEY
DOI: 10.1002/qre.2746
关键词
average run length; censored CUSUM; censored EWMA; CEV CUSUM; CM CUSUM; CM EWMA; conditional expected values; conditional median; maximum likelihood estimate; type-I censoring
This article discusses the monitoring of censored data using CUSUM control charts, developing an efficient CUSUM structure through CEV and CM methods, detecting shifts in the mean of Weibull lifetimes, assessing the impact of parameter estimation on control chart performance, evaluating the proposed charts through ARL, and comparing the ARL performance of CUSUM charts with other control charts.
This article deals with the monitoring of censored data using the cumulative sum (CUSUM) control charts for Weibull lifetimes under type-I censoring. To develop an efficient CUSUM structure for censored data, we use the conditional expected value (CEV) and conditional median (CM) approaches. In particular, we focus on the detection of shifts in the mean of Weibull lifetimes assuming censored data. In addition to fixed/known parameter values, the effect of estimation is assessed on the detection power of control charts. The performance of the proposed charts is evaluated by the average run length (ARL). Furthermore, the ARL performance of CUSUM charts is compared with CEV- and CM-based exponentially weighted moving average (EWMA) control charts. Besides an extensive simulation study, the significance of the current work is illustrated by a data set on the response time of a thermostat experiment.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据