透過您的圖書館登入
IP:3.15.6.77
  • 期刊

Applying Principal Component Analysis to a GR&R Study

應用主成份分析法於量測重複性與再現性之研究

摘要


量測重複性與再現性研究一般均著重於單一品質特性,但在製造測試方面量測重複性與再現性研究實際上是具有多變量常態分配與多重品質特性之因子。本研究利用主成份分析法(Principal Components Analysis; PCA)將有相關之多個品質特性轉換成一個或少數幾個不相關且可提供足夠資訊之變數,並針對這些新的變數再利用變異數分析法(ANOVA),來探討多重品質特性量測能力的問題。而在探討量測能力指標方面有:精密度/規格公差比(P/T比),及量測變異對量測系統總變異貢獻度(σ(上標 2 下標 gauge)/σ(上標 2 下標 Total))等二個綜合指標,以作為評估量測過程的適當性,同時對所提研究方法之應用,列舉一實例作為論證。

並列摘要


The gauge repeatability and reproducibility (GR&R) study is typically conducted on a single quality characteristic. However, manufacturing tests in a GR&R study usually have multiple characteristics with a multivariate normal distribution. Principal component analysis (PCA) method can transform the multiple characteristics into one or a few irrelevant variables and provide sufficient information. Then, these irrelevant variables were analyzed using analysis of variance. Two composite indices such as precision to tolerance (P/T) ratio and measurement variation to total variation of measurement system ratio (σ(superscript 2 subscript gauge)/σ(superscript 2 subscript Total)) combining from all variables were used to evaluate the adequacy for the measurement process. A real example was used to demonstrate the application of the proposed methodology.

參考文獻


AIAG Editing Group(1998).Measurement Systems Analysis.
Anderson, T. W.(1963).Asymptotic theory for principal component analysis.Annals of Mathematical Statistics.34,122-148.
Burdick, R. K.,G. A. Larsen(1997).Confidence intervals on measures of variability in R&R Studies.Journal of Quality Technology.29,261-273.
Cadima, J.,I. T. Jolliffe(1995).Loading and correlation in the interpretation of principal components.Journal of Applied Statistics.22,203-214.
Jackson, J. E.(1980).Principal component and factor analysis: part I-principal components.Journal of Quality Technology.12,201-213.

延伸閱讀