尚未登入 請由此登入
 
 
 










知識分享平台Eshare檢視資訊View
返回前一頁Back
      [檢舉]

標題Title: Gauge Repeatability and Reproducibility with Interaction between Operators and Products
作者Authors: 方正中,王鵬森..等
上傳單位Department: 管理與資訊系
上傳時間Date: 2009-12-4
上傳者Author: 方正中
審核單位Department: 管理與資訊系
審核老師Teacher: 方正中
檔案類型Categories: 研討會報告Seminar Report
關鍵詞Keyword: Gauge analysis, Repeatability, Reproducibility, Interaction
摘要Abstract: Due to the rapid development of the high-tech industry and the requirement of the high quality from the customers, the need of accuracy and precision on the measurement system is highly demanding. The maintenance of measurement quality is thus important. In the QS 9000 quality assurance system, the analysis of repeatability and reproducibility of the measurement variation is to apply the verified measurement system to the measurement process. Most of the industries apply the QS 9000 criterion to judge the acceptance of their measurement systems.

The purpose of the research is to improve the analysis of the measurement system nowadays. The current measurement system usually utilizes the method of the analysis of variance (ANOVA) [1] to analyze the measurement system while there is interaction between the products and operators. There are other methods can be used to analyze the measurement systems such as Classical GR&R and Long Form methods. Although the two methods are easy in calculation, they are not suitable to analyze the measurement system while there is interaction between the products and operators. Based on the disadvantage of Classical GR&R [2] and Long Form [3] methods, the research modified the two methods to make them suitable to the cases no matter there is interaction between products and operators or not.

In the mean time, the research also compared the gauge variance estimators of the five different measurement methods (ANOVA, Classical GR&R, Long Form, Modified Classical GR&R, and Modified Long Form) by simulating a measurement system based on the criterion of the biasness, variance, and mean squares error (MSE) of the estimators. The research verified that the Modified Long Form method developed in the research gives smaller MSE and much smaller gauge variance estimator both in the cases of interaction and without interaction even though it gives biased estimator of the gauge variance. The Modified Long Form method gives 15% less width of confidence interval of the gauge variance than that of ANOVA method. In addition, the research also evaluated the accuracy and precision of the gauge variance estimators of the five methods by changing the parameters (n, p or k) of the measurement system subject to the total number of the combination being fixed.

檔案名稱
FileName
檔案大小
Size
檔案格式
Format
瀏覽次數
Browses
下載次數
Downloads
2009_12_4aa17606.doc 255Kb doc 419 66
文件中檔案:
 

開啟檔案Download
開啟Ps檔
 
 
返回前一頁