動態管制圖之研究

行政院國家科學委員會專題研究計畫 成果報告

動態管制圖之研究

中 華 民 國 94 年 10 月 17 日

NSC93

M

004

006

93

8

1

94

7

31

In the former project (92.8~93.7), we will propose the designs of various adaptive control charts for two dependent process steps, and the measurement approach for the control charts performance are discussed In this project (93.8~94.7 and 94.8~95.7)), these provided adaptive control charts for multiple dependent processes will be designed from the economic statistical viewpoint. Their performances are compared with traditional fixed control charts. Data analyses show that the cost and performance of the economic statistical adaptive control charts outperform the fixed control charts. These topics have not been addresses so far, so the results will be useful and powerful in practice and academic research.

Keywords: Adaptive control charts, Markov

chain, fixed control policy, performance.

2.RESEARCH MOTIVATION AND

OBJECTIVE

To detect an out-of-control process efficient. So far, many papers consider the process mean control for a single process with adaptive sample size, sampling interval or sample size and sampling interval. However, many products are produced from dependent process steps. Hence, some

design of the adaptive control policy for the dependent process steps has not been addressed. In the project, we propose the approaches to solve the problem from statistical and economic viewpoints.

3. RESULTS AND DISCUSSION

Results:

Two topics are finished, and the two papers will be submitted to international journal. The two topics are as follows. (1) An economic statistical model of adaptive sampling interval control chart for dependent process steps is proposed, whose quality can be affected by the occurrence of two special causes, which result in a shift in the process means. The proposed model improves the performance and cost of the fixed sampling interval control charts on dependent process steps. Using the proposed adaptive process control policy for dependent process steps, the process may be monitored with better performance and lower cost.

A Markov chain approach is extended to derive the economic statistical model of the dependent process steps used to determine the design parameters of the adaptive X and e control charts under two special causes. The expression for the economic model is easier to obtain through the proposed approach rather than by adopting others. The developed model can

interval and sample size of the dependent process steps which together minimize the long term cost. Using the proposed design, the dependent process steps may be monitored adaptively with minimum cost. A Markov chain approach is extended to derive the adaptive economic model of the dependent process steps used to determine the design parameters of the adaptive sampling interval and sample size control chart, which together minimize the long term cost. It is demonstrated that the expression for the ASSI economic model is easier to obtain through the proposed approach rather than by others, and the performance and cost of the optimal adaptive controlling policy outperforms the optimal fixed controlling policy. The proposed approach can be applied to solve the real problem of the dependent process steps. The proposed processes control approach may use not only on manufacturing industry but also on service industry.

Suggestion:

All the two topics consider the adaptive process control for dependent process steps. The approaches may be extended to derive the adaptive control charts from economic viewpoints for over-adjusted process steps, or for other control charts, like CUSUM, EWMA and attribute control charts etc.

4. EVALUATION

Two topics are finished and will be submitted to international journals. We expect that they will be published on the journals.

REFERENCES

1 Amin, R. and Miller, R. (1993). A robustness study of X charts with variable sampling intervals. Journal of Quality Technology 25, 36-44.

2. Baxley, R. and JR (1995). An application

of variable sampling interval control charts. Journal of Quality Technology,27,275-282.

3. Costa , A.(1994). X charts with variable sample size. Journal of

Quality Technology, 26, 155-163.

4.Costa (1997). X charts with variable sample size and sampling interval. Journal of Quality Technology, 29, 197-204.

5.Costa (1999a). Joint and R charts with variable sample size and sampling interval. Journal of Quality Technology, 31, 387-397.

6.Costa (1999b). X charts with variable parameters. Journal of Quality Technology, 31, 408-416.

7. Daudin (1992). Double sampling X

charts. Journal of Quality Technology, 24, 78-87.

8. Prabhu S., Runger, G. and Keats, J. (1993). An adaptive sample size X chart. International Journal of Production Research 31, 2895-2909.

9. Prabhu S. ,Montgomery, D. and Runger, G.(1994). A combined adaptive sample size and sampling interval X control scheme. Journal of Quality Technology, 26, 7164-176.

10. Prabhu, S. ,Montgomery, D. and Runger, G.(1995). A design tool to evaluate average time to signal of properties of adaptive X charts. Journal of Quality Technology,27, 74-83.

11. Prabhu S. ,Montgomery, D. and Runger, G. (1997). Economic-statistical design of adaptiveX chart. International journal of Production Economics. 49,1-15. 12. Reynolds M, Amin,R.,Arnold, J. and

Nachlas, J. (1988), Charts with variable sampling intervals. Technometrics 30, 181-192.

13. Reynolds, M. (1996a). Shewhart and EWMA variable sampling interval control charts with sampling at fixed times. Journal of Quality

Technology, 28, 199-212.

14. Reynolds, M. (1996b). Variable-sampling-interval control charts with sampling at fixed times. IIE transactions, 28, 497-510.

15. Reynolds M,Arnold, J. and Balk, J. (1996), Variable sampling intervals charts in the presence of correlation. Journal of Quality Technology, 28, 12-30.

16. Reynolds, JR, and Arnold, J. (2001). EWMA control charts with variable sample sizes and sampling intervals. IIE transactions 33, 511-530.

17. Saccucci, M.Amin,R. and Lucas, J.(1992). EWMA control schemes with variable sampling interval. Communications in Statistics. 21,627-656.

18. Annadi , H.Keats,J., Runger,G.and Montgomery, D .(1995).An adaptive sample size CUSUMcontrol chart. International journal of Production Economics 33, 1605-1616.

19. Magal haes, M. , Eppr echt , E. and Costa, A. (2001). Economic desi gn of a VP X chart. I nt er nat i onal j our nal of Pr oduct i on Economi cs, 74, 191- 200.

20. Park, C. and Reynolds, M. and JR (1994). Economic design of a variable sample size X chart. Communi cat i ons i n statistics- Si mul at i on and computation. 23, 467- 483.

21. Park, Reynolds and JR (1999). Economic design of a variable sample rate X chart. Journal of Quality Technology, 31, 427-443. 22. Zhang, G.(1984), A new type of