题名

建置統計製程管制系統提升發光二極體出貨品質水準

并列篇名

Establishing a Statistical Process Control System to Improve the Outgoing quality of LEDs

DOI

10.6840/cycu201700366

作者

葉士豪

关键词

LED ; 預測區間 ; 管制圖 ; 迴歸 ; Gauge R&R ; 單因子變異數分析 ; LED ; Pediction Interval ; Control Chart ; Regression ; Gauge R&R ; ANOVA

期刊名称

中原大學工業與系統工程學系學位論文

卷期/出版年月

2017年

学位类别

碩士

导师

江瑞清

内容语文

繁體中文

中文摘要

自1996 年,白光發光二極體(LED)的技術突破,正式宣告第四世代光源–LED省電、低功率的照明時代來臨了。而對於未來的幾年,在面板市場應用上,背光源仍以LED發光二極體提供光源的技術背景下進行研發,故LED必須漸朝向高精度的趨勢發展。 本研究的第一個目的在於利用統計探討LED測試值、標準之間的相互關係,並且以迴歸分析方法,迅速找出變數之間的相互關係,提升封裝校驗效率及預測能力。本研究第二個目的為分析機台與材料之變異、機差的差異,以巨觀的測試數據透過Gauge R&R、管制圖、預測區間、單因子變異數分析等工具,降低影響因子的變異。最後,根據本研究結果顯示,預期案例可縮短封裝測試校驗證前置時間,改善校驗與測誤差,並於量產規模情況下,縮減色度(CIE)公差、亮度(IV)公差。此項結果顯示藉由管制圖、預測區間的監控,可即時找出測試生產過程中的變異作為調整方向,期以更好的良率,使LED光學特性達到更高精度的測量評價。

英文摘要

Since 1996, the technology of white light has an important breakthrough in the emitting diode (LED), it would be announce the light generation of fourth stage, it’s also can call the low-power lighting generation. In the future, backlight technology is still developing by the technical background of LED in the application of the panel market, so the LED must be developed toward the higher accuracy. The first objective of this research about the relationship between the LED test values and standards, then we can use the regression analysis to find out the relationship between the variables quickly that can improve the ability of efficiency and prediction. The second objective of this research about the machine and material difference analysis of variation, it will use Gauge R&R, control chart, prediction interval, ANOVA methods to decrease the process factor variation by mass database. Finally, according to the results of this study case, the lead time of calibration can be shortened, and improving calibration and measurement errors, then the tolerance of CIE and the brightness can be decreased in mass production stage. This result shows it can use control chart, prediction interval method to find out variation of process immediately, and to adjust them. We need to use those methods to promote the optical accuracy and process.

主题分类 電機資訊學院 > 工業與系統工程學系
工程學 > 工程學總論
参考文献
  1. 6. Ferguson, G.A., 1976, “Statistical analysis in psychology”, Education. 4th ed, New York.
    連結:
  2. 7. AIAG, “Measure System Analysis”, Automotive Industry Action Group, Detroit, Nework.
    連結:
  3. 8. Hoaglin, D. C. and Welsch, R. E., 1978, “The hat matrix in regression and ANOVA”, The American Statistician 32, pp.17-22.
    連結:
  4. 9. Chatterjee, S. and Hadi, A. S.,1986, “Influential observations, high leverage points And outliers in linear regression”, Statistical Science 1, pp.379-393.
    連結:
  5. 10. Kv˚alseth, T., 1985, “Cautionary note about R2”, The American Statistician, pp.279-285.
    連結:
  6. 3. 潘南飛,2005,“迴歸與灰色模式在預測上之比較”,工程科技與教育學刊 2卷4期。
    連結:
  7. 4. 黃郁芬,2013,“線性迴歸模型的影響測度比較”,中國統計學報 51卷2期,225-245。
    連結:
  8. 5. 黃郁芬,2016,“淺談迴歸分析教育”,Journal of the Chinese Statistical Association Vol. 54,62-78
    連結:
  9. 英文文獻
  10. 1. Grum, F., 1980, “Optical radiation measurements”, Vol. 2.
  11. 2. Bartelson, C.J., 1980, “Color measurements Academic press”, New York.
  12. 3. Wyszecki, G. and Stiles, WS., 1982, “Color Science”, 2nd ed. John Wiley and Sins, New York.
  13. 4. CIE, 1987a, 1987, “International Lighting Vocabulary”, CIE 17.4.
  14. 5. CIE, 127, 2007, “Standardization of LED measurements”, New Publication of CIE Technical Report.
  15. 11. Douglas, C. Montgomery and George C. Runger, 2011, “Applied Statistics and Probability for Engineers Fifth Edition”, pp.406.
  16. 中文文獻
  17. 1. LED照明標準及研發聯盟,2008,“LED光學與電性量測標準草案”。
  18. 2. 黃素貞,1975,“迴歸分析自變數離群值的一種評估方法”,中國統計學報第35卷第1期,27-39。
  19. 6. 吳宗正,1993,“迴歸分析-結束與應用”,複文書局。
  20. 7. 王鵬森、方正中、李妤莉,2006,“量測重複性與再現性之研究”,2006第10屆科際整合研討會,577-594。