臺灣博碩士論文加值系統

在生產過程中生產環境或原物料的改變均可能導致製程狀態偏離目標值，這些使產品品質產生變異的原因包含了製程平均值與變異數的改變，因此為了能即時監控製程狀態，在本研究中將分別探討管制圖的偵測能力並提出改善方法，以期進一步提升管制圖的整體偵測成效。
在計量值管制圖方面，當母體分配為常態時我們將提出多重累積和管制法及合併多重累積和管制法與輔助法則，這兩種方法以改善累積和管制法之偵測效益，並探討輔助法則的建立與相關管制法的參數設定。當製程之母體分配為指數分配時，我們將以類神經網路管制法與合併類神經網路-累積和管制法。來改善累積和管制法之偵測能力，並探討合併類神經網路與累積和管制法的管制效益。
在計數值管制圖方面，本文將探討不合格點數管制圖，在不合格點數產生群聚現象時 管制圖的應用與缺點，並提出以尼曼分配為不合格點之機率模型時，累積和管制法之改善效益。再者，當產品品級為多項等級分類時，本研究亦將提出以多項分類為基礎的累積和管制法。以及使用機率密度函數估計方法來建立逐次機率比檢定管制法以監視製程平均值的改變。
這些改善方案的共同特徵為，同時使用歷史數據為依據，並藉由異常數據之出現順序，以獲得更多製程資訊及增進管制方法之偵測能力。整個研究的重點為使用類神經網路及統計方法為工具，並以實例說明這些改善方案的整體偵測成效。在這些改善方案的研究結果中，吾人發現多重累積和管制法之偵測成效將優於單一累積和管制法。而類神經網路管制法的偵測成效則稍優於指數累積和管制法，但與指數累積和管制法合併後將可更進一步提升偵測能力。在計數值管制圖改善方法中，本研究也發現，當不合格點產生群聚現象時，可經由適當的參數設定來使用累積和管制法以改善製程。另外在一些以類別資料為基礎的管制法中，本文則發現，經由適當的機率密度函數估計再套用逐次機率比管制法將可大幅提升偵測製程微量變動的能力。

Control charts are widely used for both manufacturing and service industries. Cumulative sum (CUSUM) charts are known to be very sensitive in detecting small shifts in the process mean. In this research, first, we proposed a multiple CUSUM which used several simultaneous conventional CUSUM statistics with different resetting boundaries. Secondary, we proposed a neural network as an alternative approach to CUSUM charts when monitoring exponential mean. Finally, we also explore the attribute control charts for a clustering phenomenon of defective items and multiple classifications for quality level described by linguistic variables.
The performance of these procedures was evaluated by estimating the average run lengths (ARL’s) using theoretical computing or simulation. The results obtained with simulated and real-world data suggest that both multiple CUSUM and neural network are significantly more sensitive to process shifts than CUSUM charts. As for attribute control charts, the presented formula of reference value can form a CUSUM control chart for monitoring the clustering phenomenon of defective items. The results suggest that proposed method are sensitive than exists control chart, especially in small mean shifts. By way of estimated probability density function of based variable of categorical data, the charting procedure based on sequential probability ratio testing also work better than fuzzy control charts under multiple classification.

目 錄
頁次
中文摘要 i
英文摘要 ii
誌謝 iii
目錄 iv
表目錄 vii
圖目錄 viii
一、 緒論 1
1.1 研究背景與動機 1
1.1.1 計量值管制圖：常態分配 2
1.1.2 計量值管制圖：指數分配 3
1.1.3 計數值管制圖：尼曼分配 4
1.1.4 計數值管制圖：多項分配 5
1.2 研究目的 7
1.3 研究範圍 7
1.4 研究方法與架構 8
二、 文獻探討 10
2.1 計量值管制圖：常態分配 10
2.2 計量值管制圖：指數分配 14
2.2.1 統計方法應用文獻探討 14
2.2.2 類神經網路應用文獻探討 16
2.3 計數值管制圖：尼曼分配 19
2.4 計數值管制圖：多項分配 21
2.4.1 多項分配法(Multinomial methods) 22
2.4.2 群組資料法(Group data methods) 23
2.4.3 模糊理論法(Fuzzy methods) 23
三、 計量值管制法之改善研究 26
3.1 常態分配 26
3.1.1 多重累積和管制法 26
3.1.2 合併多重累積和管制法與輔助法則 27
3.2 指數分配 28
3.2.1 類神經網路法 28
3.2.2 合併類神經網路-累積和管制法 36
四、 計數值管制法之改善研究 38
4.1 尼曼分配 38
4.2 多項分配 39
4.2.1 邊界精確量測法 40
4.2.2 主觀量測法 43
五、 研究結果評估與比較 47
5.1 常態分配 47
5.1.1 管制效益評估 47
5.1.2 管制效益分析 47
5.2 指數分配 49
5.2.1 視窗大小之影響 49
5.2.2 類神經網路訓練樣本數的決定 50
5.2.3 類神經網路管制法的適用範圍 51
5.2.4 合併類神經網路-累積和管制法 52
5.3 尼曼分配 53
5.3.1 累積和管制平均連串法長度計算 53
5.3.2 值大小對管制成效的影響 55
5.3.3 效益評估比較 55
5.3.4 研究結果 59
5.4 多項分配 59
5.4.1 邊界精確量測法 59
5.4.2 主觀量測法 60
六、 應用實例 64
6.1 多重累積和管制法 64
6.2 合併類神經網路與指數累積和管制法 67
6.3 具群聚現象之累積和管制法 69
6.4 模糊分類之SPRT管制法 73
七、 結論與建議 82
7.1 總結 82
7.2 建議與展望 83
參考文獻 86
附錄 90
表目錄
頁次
表1.1 計量方法與計數方法的特性比較 6
表2.1 累加數據管制法在製程數據分析上之應用 14
表2.2 應用類神經網路之相關研究比較 19
表3.1 訓練樣本結構 31
表5.1 多重管制法與單一累積和管制法ARL之比較 48
表5.2 合併多重累積和管制法與輔助法則之ARL 49
表5.3 類神經網路和指數累積和管制法ARL之比較 50
表5.4 類神經網路與多重累積和管制法之ARL比較 50
表5.5 類神經網路和指數累積和管制法ARL之比較 53
表5.6 值大小對管制成效之影響 55
表5.7 各種管制法之平均連串長度比較 56
表5.8 參數組合比 時之ARL 57
表5.9 參數組合比 時之ARL 57
表5.10 參數組合比 時之ARL 58
表5.11 雙邊累積和管制法之ARL 59
表5.12 ARL比較 60
表5.13 製程正常時之樣本資料 62
表5.14 管制效益之比較 62
表6.1 多重累積和管制法運算範例 66
表6.2 數據模擬與結果分析 68
表6.3 晶圓表面缺陷數資料 71
表6.4 鋁料之最大彎距與分類結果(管制內) 76
表6.5 鋁料之最大彎距與分類結果(管制外) 78
表6.6 各項分類之機率密度函數比 80
表6.7 管制外之偵測效益比較 81
圖目錄
頁次
圖1.1 研究架構 9
圖2.1 指數分配( )機率密度函數圖 15
圖3.1 類神經網路之分析視窗 30
圖3.2 訓練樣本範例 31
圖3.3 轉換函數 32
圖4.1 語意資料的產生方法 43
圖5.1 訓練樣本數對管制成效的影響 51
圖5.2 不同管制內ARL下類神經網路之偵測效益 52
圖5.3 轉移機率矩陣 54
圖5.4 各個語意值之歸屬函數圖 61
圖6.1 積體電路之製造流程 69
圖6.2 晶圓檢測儀KLA2139 70
圖6.3 鋁擠型工廠之製造流程圖 74
圖6.4 分類之歸屬函數 75
圖6.5 最大彎距量測資料與分類結果(管制內) 77
圖6.6 最大彎距量測資料與分類結果(管制內) 78
圖6.7 最大彎距之分配型態 79

1. Albin, S., and D. J. Friedman, "Clustered defects in IC fabrication: Impact on process control charts," IEEE transactions on semiconductor manufacturing, 4, 36-42 (1991).
2. Bissell, A. F., "CUSUM techniques for quality control," (with discussion), Applies Statistics, 18, 1-30 (1969).
3. Brook, D., and D. A. Evans, "An approach to the probability distribution of CUSUM run length," Biometrika, 59, 539-549 (1972).
4. Champ, C. A., and W. H. Woodall, "Exact results for Shewhart control charts with supplementary runs rules," Technometrics, 29, 393-399 (1987).
5. Chang, S. I., and C. A. Aw, "A neural fuzzy control chart for detecting and classifying process means shifts," International Journal of Production Research, 34, 2265-2278 (1996).
6. Chang, T. C., and F. F. Gan, "A cumulative sum control chart for monitoring process variance," Journal of Quality Technology, 27, 109-119 (1995).
7. Charles, W. C., and S. E. Rigdon, "A comparison of the Markov chain and the integral equation approaches for evaluating the run length distribution of quality control charts," Communication Statistics-Simulation and Computation, 20, 191-204 (1991).
8. Cheng, C. S., "A cumulative score control scheme for detecting process shifts," Communications in Statistics-Theory and Methods, 24, 755-744, (1995).
9. Cheng, C. S., "A multi-layer neural network model for detecting changes in the process mean," Computers & Industrial Engineering, 28, 51-61, (1995).
10. Cheng, C. S., "Detecting changes in the process mean using artificial neural networks approach," Journal of Chinese Institute of Industrial Engineers, 11, 47-54 (1994).
11. Cheng, C. S., and C. A. Tzeng, "A neural network approach for detecting shifts in the process mean and variability," Journal of the Chinese Institute of Industrial Engineers, 11, 67-75 (1994).
12. Crosier, R. B., "A new two-sided cumulative sum quality control scheme," Technometrics, 28, 187-194 (1986).
13. Crowder, S. V., and M. D. Hamilton, "An EWMA for monitoring a process standard deviation," Journal of Quality Technology, 24, 12-21 (1992).
14. Davis, R. B., and A. Homer, "Performance of the zone control chart," Communications in Statistics-Theory and Methods, 19, 1581-1587 (1990).
15. Gan, F. F., "Design of one- and two-sided exponential EWMA charts," Journal of Quality Technology, 30, 55-69 (1998).
16. Gan, F. F., "Design of optimal exponential CUSUM control charts," Journal of Quality Technology, 26, 109-123 (1994).
17. Gan, F. F., and K. P. Choi, "Computing average run lengths for Exponential CUSUM schemes," Journal of Quality Technology, 26, 134-143 (1994).
18. Guo, Y., and K. J. Dooley, "Identification of change structure in statistical process control," International Journal of Production Research, 30, 1655-1669 (1992).
19. Guo, Y., and K. J. Dooley, "The application of neural networks to a diagnostic problem in quality control," Proceedings of ASME-WAM: Monitoring and Control for Manufacturing Processes, 111-122 (1990).
20. Hamilton, M. D., and S. V. Crowder, "Average run lengths of EWMA control charts for monitoring a processing standard deviation," Journal of Quality Technology, 24, 44-50 (1992).
21. Hwarng, H. B., and N. F. Hubele, " control chart pattern identification through efficient off-line neural network training," IEE Transactions, 25, 27-40 (1993).
22. Hwarng, H. B., and N. F. Hubele, "Back-propagation pattern recognizers for control chart: methodology and performance," Computers and Industrial Engineering, 24, 219-235 (1993).
23. Jaehn, A. H., "Zone control chart: A new tool for quality control," Tappi journal 70, 159-161 (1987).
24. Kanagawa, A., F., Tamaki, and H. Ohta, "Control charts for process average and variability based on linguistic data," International Journal of Production Research, 31, 4, 913-922 (1993).
25. Kenneth, G. M., M. A. Woodbury, and H. D. Tolley, Statistical Application Using Fuzzy Sets, Wiley, NY (1994).
26. Laviolette, M., J. W. Seaman, J. D. Barrett, and W. H. Woodall, "A probabilistic and statistical view of fuzzy methods," Technometrics, 37, 249-262 (1995).
27. Lucas, J. M., "Combined Shewhart-CUSUM quality control schemes," Journal of Quality Technology, 14, 51-59 (1982).
28. Lucas, J. M., "Counted data CUSUM’s," Technometrics, 27, 129-144 (1985).
29. Lucas, J. M., and R. B. Crosier, "Fast initial response for CUSUM quality control schemes: Give your CUSUM a head start," Technometrics, 24, 199-205 (1982).
30. Marcucci, M., "Monitoring multinomial process," Journal of Quality Technology, 17, 86-91 (1985).
31. Nelson, L. S., "Interpreting Shewhart control charts," Journal of Quality Technology, 17, 114-116 (1985).
32. Nelson, L. S., "The Shewhart control chart-tests for special cause," Journal of Quality Technology, 16, 237-239 (1984).
33. Page, E. S., "Continuous inspection schemes," Biometrika, 41, 100-115 (1954).
34. Pham, D. T., and E. Oztemel, "Control chart pattern recognition using neural networks," Journal of System Engineering, 2, 256-262 (1992).
35. Pugh, G. A., "A comparison of neural networks to SPC charts," Computers and Industrial Engineering, 21, 253-255 (1991).
36. Pugh, G. A., "Synthetic neural networks for process control," Computers and Industrial Engineering, 17, 24-26 (1989).
37. Raz, T., and J. H. Wang, "Probabilistic and membership approaches in the construction of control charts for linguistic data," Production Planning & Control, 1, 147-157 (1990).
38. Reynolds, J. H., "The run sum control chart procedure," Journal of Quality Technology, 3, 23-27 (1971).
39. Reynolds, M. R. Jr., R. W. Amin, and J. C. Arnold, "CUSUM charts with variable sampling intervals, " Technometrics, 32, 371-384 (1990).
40. Roberts, S. W., "Control chart tests based on geometric moving averages," Technometrics, 1, 239-250 (1959).
41. Shenton, L. R., and K. O. Bowman, "Remarks on large sample estimators for some discrete distributions," Technometrics, 9, 587-598 (1967).
42. Shewhart, W. A., Economic control of quality of manufactured product, Van Nostrand, NY (1931).
43. Smith, A. E., "X-bar and R control chart interpretation using neural computing," International Journal of Production Research, 32, 309-320 (1994).
44. Stiner, S. H., P. L. Geyer, and G. O. Wesolowsky, "Control charts based on grouped observations, "International Journal of Production Research, 32, 75-91 (1994).
45. Vardeman, S., and D. Ray, "Average run lengths for CUSUM schemes when observations are exponentially distributed," Technometrics, 27, 145-150 (1985).
46. Wald, A., Sequential Analysis, Wiley, NY (1947).
47. Waldmann, K. H., "Bounds for the distribution of the run length of one-sided and two-sided CUSUM-quality control schemes," Technometrics, 28, 61-67 (1986).
48. Wang, J. H., and T. Raz, "On the construction of control charts using linguistic variables," International Journal of Production Research, 28, 477-487 (1990).
49. Western Electric Company, Statistical Quality Control Handbook, Western Electric Company Incorporated, Indiana (1958).
50. White, C. H., J. B. Keats and J. Stanley, "Poisson cusum versus c chart for defect data," Quality Engineering, 9, 673-679 (1997).
51. Xiao, H., "A cumulative score control scheme," Applied Statistics, 41, 47-54 (1992).
52. Yashchin, E., "One the analysis and design of CUSUM-Shewhart control schemes," IBM Journal of Research and Development, 29, 377-384 (1985).
53. Yashchin, E., "Some aspects of the theory of statistical control schemes," IBM Journal of Research and Development, 31, 199-205 (1987).
54. Zadeh, L. A., "Fuzzy set," Information and Control, 8, 338-353 (1965).