|
|
Journal of Korean Society for Quality Management 2000;28(2): 39-. |
| 분계점 붓스트랩 방법을 이용한 자기상관을 갖는 공정의 $ar{X}$ 관리도 |
| 김윤배1, 박대수2 |
1성균과대학교 시스템경영공학부
2한국통신 경영연구소 |
| $ar{X}$ control charts of automcorrelated process using threshold bootstrap method |
| Yun-Bae Kim1, Dae-Su Park2 |
1School of Systems Management Engineering, Sungkyunkwan University
2Management Research Lab. Korea Telecom |
|
|
|
|
|
| ABSTRACT |
|
${overline{X}}$ control chart has proven to be an effective tool to improve the product quality. Shewhart charts assume that the observations are independent and normally distributed. Under the presence of positive autocorrelation and severe skewness, the control limits are not accurate because assumptions are violated- Autocorrelation in process measurements results in frequent false alarms when standard control chats are applied in process monitoring. In this paper, Threshold Bootstrap and Moving Block Bootstrap are used for constructing a confidence interval of correlated observations. Monte Carlo simulation studies are conducted to compare the performance of the bootstrap methods and that of standard method for constructing control charts under several conditions. |
|
|
|
PDF Links
Full text via DOI 
Download Citation 
