(3.235.139.152) 您好!臺灣時間:2021/05/11 12:19
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果

詳目顯示:::

: 
twitterline
研究生:朱彩碧
研究生(外文):Cai-Bi Zhu
論文名稱:基於多重樣本製程期望損失統計分配和推論性質
論文名稱(外文):Distributional and Inferential Properties of the Process Expected Loss Based on Multiple Samples
指導教授:林鴻欽林鴻欽引用關係
指導教授(外文):Hung-Chin Lin
學位類別:碩士
校院名稱:萬能科技大學
系所名稱:經營管理研究所
學門:商業及管理學門
學類:企業管理學類
論文種類:學術論文
論文出版年:2008
畢業學年度:96
語文別:中文
論文頁數:54
中文關鍵詞:製程期望損失多重樣組漸近分配信賴區間上限
外文關鍵詞:process expected lossmultiple samplesasymptotic distribution100(1-α)% upper confidence bound
相關次數:
  • 被引用被引用:0
  • 點閱點閱:104
  • 評分評分:系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔系統版面圖檔
  • 下載下載:7
  • 收藏至我的研究室書目清單書目收藏:0
製程損失指Le是以二次損失函數觀點所發展的指標;事實上,Le= Lpe+Lot,其中,Lpe表示潛在相對期望損失,Lot表示偏離目標值相對損失。又Le指標的估計式比Cpm指標的估計式有較優勢的統計推論性質,以及Le指標定義出兩個分離的項式,分別評估製程品質的準確性和精確性。過去學者對於Le指標的研究成果,大都針對單一樣組製程數據進行分析討論;不過,有些時候製程品質數據會以多重樣組的型態收集來呈現。本文就製程品質數據為多重樣組型態所組成時,推導Lpe、Lot和Le指標估計式的動差、期望值、變異數和均方差等分配性質,以及推導Le指標估計式抽樣分配,並且探討與比較各不同指標估計式的推論性質,包含不偏性、一致性和有效性等。
本文亦討論Lpe、Lot和Le指標估計式的漸近分配等相關性質。接著在不同條件下對於Le指標估計式,利用估算信賴水準100(1-α)%的信賴區間上限的方式,提出評估製程品質績效的可靠程序,以利產業實務之應用。並進一步根據信賴區間上限和Le指標估計值的比率,在配合抽樣計畫與檢測成本之情況下,提供實務應用時樣本組合數量決定之參考。
Process expected loss indices, Lpe, Lot, and Le provide measures to determine the quality performance of a process. In fact, Le= Lpe+Lot, Lpe denotes the potential relative expected loss, Lot is the relative off-target squared. In real situations where the actual values of Lpe, Lot, and Le are unknown one may estimate them by their corresponding sample counterparts. Most of the results obtained regarding the distributional and inferential properties of estimated expected loss indices were based on one single sample. In practice, however, process information is often derived from multiple samples rather than from one single sample. In this paper, we first investigate the distributional and inferential properties of the estimators of process expected loss indices based on multiple samples. We then investigate the performance evaluation of the process based on the determination of 100(1-α)% upper confidence bound of process loss index, Le, for various combinations of sample size, and implement the evaluating procedure. Moreover, this paper provides equation to estimate the approximate the combinations of sample size necessary to achieve a desired upper conference bound with specified confidence level. The technique provided in this paper will be applicable when the process measurements are taken from control chart samples.
中文摘要...................................................i
英文摘要..................................................ii
誌謝.....................................................iii
目錄......................................................iv
表目錄.....................................................v
圖目錄....................................................vi
第一章 導論................................................1
1.1 研究背景............................................1
1.2 研究動機與目的......................................5
1.3 研究架構............................................7
第二章 估計Lpe 和Lot 指標..................................8
2.1 Lpe指標的估計式.....................................8
2.2 Lpe指標估計式的統計性質............................11
2.3 Lot指標的估計式....................................13
2.4 Lot指標估計式的統計性質............................18
第三章 估計製程損失指標...................................21
3.1 Le指標的估計式.....................................21
3.2 Le指標估計式的統計性質.............................23
第四章 製程損失指標的信賴上限.............................27
4.1 Le指標的信賴上限...................................27
4.2 適當樣本組合數量的決定.............................36
第五章 結論與展望.........................................41
5.1 結論...............................................41
5.2 未來展望...........................................42
參考文獻..................................................44

表目錄
表1.1 製程A、B、C、D和E品質績效的比較......................4
表1.2 不同的Lpe指標值和Lot指標值組合下的製程良率...........4
表4.1 90%信賴水準的信賴區間上限LU,m=10(5)30,n=2(1)10,以及^Le =0.01(0.01)0.10.......................................32
表4.2 95%信賴水準的信賴區間上限LU,m=10(5)30,n=2(1)10,以及^Le =0.01(0.01)0.10.......................................33
表4.3 97.5%信賴水準的信賴區間上限LU,m=10(5)30,n=2(1)10,以及^Le =0.01(0.01)0.10.....................................34
表 4.4 99%信賴水準的信賴區間上限LU,m=10(5)30,n=2(1)10,以及^Le =0.01(0.01)0.10.....................................35
表 4.5 90%信賴水準之^Le /LU值,m=5(5)60,n=2(1)15.........38
表 4.6 95%信賴水準之^Le /LU值,m=5(5)60,n=2(1)15.........39
表 4.7 97.5%信賴水準之^Le /LU值,m=5(5)60,n=2(1)15.......39
表 4.8 99%信賴水準之^Le /LU值,m=5(5)60,n=2(1)15.........40

圖目錄
圖2.1 不同Lpe值的B(~Lpe)曲線,N=60,(m,n)=(a)(10,6)、(b)(12,5)、(c)(15,4)、(d)(20,3)和(e)(30,2)...................10
圖2.2 不同Lpe值的MSE(^Lpe)曲線,N=60,(m,n)=(a)(10,6)、(b)(12,5)、(c)(15,4)、(d)(20,3)和(e)(30,2)...................10
圖2.3 不同Lpe值的MSE(~Lpe)曲線,N=60,(m,n)=(a)(10,6)、(b)(12,5)、(c)(15,4)、(d)(20,3)和(e)(30,2)...................11
圖2.4 不同Lpe值的B(^Lot)曲線,m=10,n=(a)4、(b)5、(c)6、(d)8和(e)10...................................................16
圖2.5 不同Lpe值的B(^Lot)曲線,n=5,m=(a)5、(b)10、(c)15、(d)20和(e)30.................................................16
圖2.6 不同Lpe值的MSE(^Lot)曲線,Lot=0.06,m=10,n=(a)4、(b)5、(c)6、(d)8和(e)10......................................16
圖2.7 不同Lpe值的MSE(^Lot)曲線,Lot=0.06,n=5,m=(a)5、(b)10、(c)15、(d)20和(e)30...................................17
圖2.8 不同Lpe值的MSE(~Lot)曲線,Lot=0.06,m=10,n=(a)4、(b)5、(c)6、(d)8和(e)10......................................17
圖2.9 不同Lpe值的MSE(~Lot)曲線,Lot=0.06,n=5,m= (a)5、(b)10、(c)15、(d)20和(e)30...................................17
圖2.10 不同Lpe值的MSE(^Lot)曲線,(m,n)=(10,6),Lot =(a)0.56、(b)0.25、(c)0.06、(d)0.03和(e)0.00..................18
圖2.11 不同Lpe值的MSE(~Lot)曲線,(m,n)= (10,6),Lot =(a)0.56、(b)0.25、(c) 0.06、(d)0.03和(e)0.00.................18
圖3.1 不同Lpe值的B(^Le)曲線,N=60,(m,n)=(a)(10,6)、(b)(12, 5)、(c)(15,4)、(d)(20,3)和(e)(30,2).......................26
圖3.2 不同Lpe值的MSE(^Le)曲線,Le =0.06,N=60,(m,n)=(a)(10,6)、(b)(12,5)、(c)(15,4)、(d)(20,3)和(e)(30,2)........26
圖3.3 不同Lpe值的MSE(^Le)曲線,(m,n)=(10,6),Le =(a)0.11、(b)0.06、(c)0.05、(d)0.04和(e)0.03........................26
圖 4.1 不同Lot/Lpe值,95%信賴水準的信賴區間上限LU曲線,^Le=0.11,(m,n)=(10,6)、(12,5)、(15,4)、(20,3)和(30,2) (曲線由上至下依序顯示).........................................30
圖 4.2 不同Lot/Lpe值,95%信賴水準的信賴區間上限LU曲線,^Le=0.06,(m,n)=(10,6)、(12,5)、(15,4)、(20,3)和(30,2) (曲線由上至下依序顯示).........................................30
圖 4.3 不同Lot/Lpe值,95%信賴水準的信賴區間上限LU曲線,^Le=0.05,(m,n)=(10,6)、(12,5)、(15,4)、(20,3)和(30,2) (曲線由上至下依序顯示).........................................30
圖 4.4 不同Lot/Lpe值,95%信賴水準的信賴區間上限LU曲線,^Le=0.04,(m,n)=(10,6)、(12,5)、(15,4)、(20,3)和(30,2) (曲線由上至下依序顯示).........................................31
圖 4.5 不同Lot/Lpe值,95%信賴水準的信賴區間上限LU曲線,^Le=0.03,(m,n)=(10,6)、(12,5)、(15,4)、(20,3)和(30,2) (曲線由上至下依序顯示).........................................31
圖4.6 不同Lot/Lpe值,90%信賴水準的^Le /LU曲線,(m,n)=(a)(10,6)、(b)(12,5)、(c)(15,4)、(d)(20,3)和(e)(30,2) (曲線由上至下依序顯示).............................................37
圖4.7 不同Lot/Lpe值,95%信賴水準的^Le /LU曲線,(m,n)=(a)(10,6)、(b)(12,5)、(c)(15,4)、(d)(20,3)和(e)(30,2) (曲線由上至下依序顯示).............................................37
圖4.8 不同Lot/Lpe值,97.5%信賴水準的^Le /LU曲線,(m,n)=(a)(10,6)、(b)(12,5)、(c)(15,4)、(d)(20,3)和(e)(30,2) (曲線由上至下依序顯示).............................................38
圖4.9 不同Lot/Lpe值,99%信賴水準的^Le /LU曲線,(m,n)=(a)(10,6)、(b)(12,5)、(c)(15,4)、(d)(20,3)和(e)(30,2) (曲線由上至下依序顯)..............................................38
1.Arnold, S. F., Mathematical Statistics, Prentice-Hall,
1990.
2.Anderson, T. W., An Introduction to Multivariate
Statistical Analysis, John Wiley and Sons, 1984.
3.Boyles, A. B., “The Taguchi Capability Index,” Journal
of Quality Technology, Vol. 23, 1991, pp. 17–26.
4.Chan, L. K., Cheng, S. W. and Spiring, F. A., “Anew measure of process capability: Cpm,” Journal of Quality
Technology, Vol. 20, 1988, pp. 162–173.
5.Franklin, L. A., “Sample size determination for
lowerconference limits for estimating process capability indices,” Computers & Industrial Engineering, Vol. 36, pp. 603-614, 1999.
6.Hubele, N. F. and Vännman, K., “The effect of pooled and un-pooled variance estimators on Cpm when using subsamples,” Journal of Quality Technology, Vol. 18, 2004, pp. 41-52.
7.Hsiang, T. C. and Taguchi, G.., A tutorial on quality control and assurance – the Taguchi methods, ASA Annual Meeting, Las Vegas, Nevada, 1985.
8.Kirmani, S. N. U. A., Kocherlakota, K. and Kocherlakota, S., “Estimation of σ and the process capability index based on subsamples,” Communications in Statistics – Theory and Methods, Vol. 20, 1991, pp. 275-291.
9.Kane, V. E., “Process capability indices,” Journal of Quality Technology, Vol.18, 1986, pp.41–52 (Corrigenda 260).
10.Montgomery, D. C., Introduction to Statistical Quality Control, John Wiley & Sons, New York, 2001.
11.Patnaik, P. B., “The non-central χ2 and F-distributions and their approximations,” Biometrika, Vol. 36, pp. 202-232, 1949.
12.Pearn, W. L., Lin, G. H. and Chen, K. S., “Distributional and inferential properties of the process accuracy and process precision indices,” Communications in Statistics–Theory and Methods, Vol. 27, 1998, pp. 985–1000.
13.Pearn, W. L. and Lin, G. H., “Estimated Incapability Index: Reliability and decision making with sample information,” Quality and Reliability Engineering International, Vol. 18, 2002, pp. 141-147.
14.Pearn, W. L., Chang, Y. C. and Wu, C. W., “Distributional and inferential properties of the estimated precision Cp based on multiple samples,” Quality and Quantity, Vol. 37, 2003, pp. 443-453.
15.Pearn, W. L., Chang, Y. C. and Wu, C. W., “Distributional and Inferential Properties of the Process Loss Indices,” Journal of Applied Statistics, Vol. 31, No 9, 2004, pp. 1115-1135.
16.Pearn, W. L., Chang, Y. C. and Wu, C. W., “Bootstrap approach for estimating process quality yield with application to light emitting diodes,” International Journal of Advanced Manufacturing Technology, Vol. 25, 2005, pp. 560-570.
17.Pearn, W. L., Chang, Y. C. and Wu, C. W., “Multiple-process performance analysis chart based on process loss indices,” International Journal of Systems Science, Vol. 37, 2006, pp. 429-435.
18.Serfling, R. J., “Approximation Theorems of Mathematical Statistics,” John Wiley and Sons, 1980.
19.Tsui, K. L., Interpretation of process capability indices and some alternatives, Quality Engineering, Vol. 9, pp. 587-596, 1997.
20.Zhang, N. F., “Combining process capability indices from a sequence of independent samples,” International Journal of Production Research, Vol. 39, 2001, pp. 2769-2781.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
系統版面圖檔 系統版面圖檔