標題: | Analyzing type II censored data obtained from repetitious experiments |
作者: | Tong, LI Yang, CH 工業工程與管理學系 Department of Industrial Engineering and Management |
關鍵字: | type II censored data;least square estimation;Torres's method;experimental design;Taguchi's parameter design |
公開日期: | 1-Jan-2006 |
摘要: | Experimental design and Taguchi's parameter design are widely employed by industry to optimize the process/product. However, censored data are often observed in product lifetime testing during the experiments. After implementing a repetitious experiment with type II censored data, the censored data are usually estimated by establishing a complex statistical model. However, using the incomplete data to fit a model may not accurately estimates the censored data. Moreover, the model fitting process is complicated for a practitioner who has only limited statistical training. This study proposes a less complex approach to analyze censored data, using the least square estimation method and Torres's analysis of unreplicated factorials with possible abnormalities. This study also presents an effective method to analyze the censored data from Taguchi's parameter design using least square estimation method. Finally, examples are given to illustrate the effectiveness of the proposed methods. |
URI: | http://dx.doi.org/10.2528/PIER06021901 http://hdl.handle.net/11536/12920 |
ISSN: | 0266-4763 |
DOI: | 10.2528/PIER06021901 |
期刊: | JOURNAL OF APPLIED STATISTICS |
Volume: | 33 |
Issue: | 1 |
起始頁: | 49 |
結束頁: | 63 |
Appears in Collections: | Articles |
Files in This Item:
If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.