在現今供應商選擇的問題中，無論是傳統製造業或是科技產業等，在整個採購作業中，最重要的考量因子即是品質。選擇一個好的供應商，不但能降低生產成本，並可提高企業品牌形象。因此供應商評選對製造商提升其競爭力扮演著極為重要的角色。製程能力指標(Process Capability Indices, PCIs)能有效評估製程產出之品質水準並且提供一個量化的數值，製程能力指標Cpk至今已廣泛運用於業界衡量製程能力，但是Cpk並無有效衡量非對稱型製程規格之能力，而基於非對稱型製程能力指標Cpk"無論對於對稱型製程規格或是非對稱型製程規格皆有良好的估計能力，本研究以製程指標Cpk"為基礎，比較供應商產出績效方法之研究，但Cpk"指標估計量之抽樣分配相當複雜，導致以兩供應商製程能力比較問題的精確方法難以建構。因此本研究採用了七種近似的評估方法分別為Bonferroni法、Difference法、Generalized Confidence Intervals (GCI)法以及四種利用Bootstrap抽樣技巧的Standard Bootstrap (SB)法、Percentile Bootstrap (PB) 法、Biased-Corrected Percentile Bootstrap (BCPB) 法與Bootstrap-t(BT)法。建構出兩指標差異檢定與比值檢定之信賴下界,為了驗證各評估方法的績效及表現，透過模擬方式分析型I誤差和檢定力分析各方法之間的績效，此外為使實務方便參考，本研究亦針對在特定的檢定力需求下提供所需樣本數建議，最後透過實例分析說明以提供決策者在評估供應商或製程之產出績效的參考依據。

Currently, when facing the problem of supplier selection, quality is the main consideration in the whole procurement process to both traditional manufactory and technology industries. With a good supplier, production costs would be reduced and the brand image may be enhanced. Therefore, supplier selection plays an important role in promoting competition.
Process Capability Indices (PCIs) can effectively evaluate the quality of output and provide quantitative values on process performance. However, it has been shown that the Cpk index is not appropriate for measuring process performance if the tolerance is asymmetric even the Cpk index is widely used index in practice. To remedy for this,the Cpk" index has been developed for measuring process performance with asymmetric tolerance.
In this thesis, the performance of the supplier is evaluated and compared based on the Cpk" index. Seven approximations include Bonferroni method, Difference method, Generalized Confidence Intervals (GCI) method and four resampling techniques using the Bootstrap Standard Bootstrap (SB) method, Percentile Bootstrap (PB) method, Biased-Corrected Percentile Bootstrap (BCPB) method and the Bootstrap-t (BT) method, have been studied. There are two test statistics, difference statistics and the ratio statistics . In order to examine the performance of this method, type I error and the power of the test are analyzed via simulations. Moreover, minimal sample size required for achieving a specified power is provided for practical applications. Finally, an application example is illustrated for demonstration.

1. Bissell, F. A.(1990). How reliable is your capability. Appl. Stat., 331–340.
2. Boyles, R.A.(1994) Process capability with asymmetric tolerances.Commun Statist-Simul 23:615-643.
3. Chan, L.K. Cheng, S.W. and Spiring, F.A. (1999) A new measure of process ca-pabil¬ity: . Journal of Quality Technology 20:162-175.
4. Dickson, G. W. (1966). An analysis of vendor selection systems and decisions. Journal of Purchasing, 2, 5-17.
5. Pearn, W.L., Lin, P.C. and Chen, K.S.,(2004).The index for asymmetric tolerances: implication and inference tolerances:implications and inference. Metrika 60 (2),119-136.
6. Eforn, B. (1981). Nonparametric standard errors and confidence intervals. Can. Journal of Statist., 139–172.
7. Efron, B. (1979). Bootstrap methods: another look at Jackknife. Annals of Statis-tics, 7, 1-26.
8. Efron, B. (1982). The Jackknife, the Bootstrap and Other Resample Plans. Soci-ety of Industrial and Applied Mathematics (SIAM), Philadelphia, PA.
9. Efron, B. and Tibshirani, R.J. (1986). Bootstrap methods for standard errors, confidence interval,. Statist. Sci., 54–77.
10. Franklin, L.A. and Wasserman, G.S.(1991).Bootstrap confidence interval estimates of : An introduction. Communications in Statistics: Simulation and Computation. Journal of Quality Technology, 1992b, 24(4), 231-242.
11. Franklin, L. A. and Wasserman, G. S. (1992a). A note on the conservative nature of the tables of lower confidence limits for with a suggested correction. Comm. in Statistics: Simul. &Computa., 21(4), 1165–1169.
12. Franklin, L. A. and Wasserman, G. S. (1992b). Bootstrap lower confidence limits for capability indices. Journal of Quality Technology, 24(4), 196-210.
13. Iyer,H.K. and Patterson,P. (2002). A recipe for constructing generalized pivotal quantities and generalized confidence intervals,. Tech. Rep. 2002-10, Department of Statistics, Colorado State University, 2002. Available at http://www.stat.colostate.edu/research/2002_10.pdf
14. Juran, M.J. (1974). Quality Control Handbook, 3rd ed. USA: McGraw-Hill.
15. Kane,V.E.(1986).Process capability indices. Journal of Quality Technology,41–52.
16. Kushler,R.H. and Hurley,P. (1992). Confidence bounds for capability indices. Journal of Quality Technology,24:188–195.
17. Mathew, T., Sebastian, G. and Kurian, K. M. (2006). Generalized confidence in-tervals for process capability indices. Quality and Reliability Engineering Inter-national, 23(4), 471-481.
18. Olhager, J. and Selldin, E. (2004). Supply chain management survey of Swedish manufacturing firms. International Journal of Production Economics, 89, 216-233.
19. Pearson, J. N. and Ellram, L. M. (1995). Supplier selection and evaluation in small versus large electronics firms. Journal of Small Business Management, 33(4), 53-65.
20. Pearn,W.L. and Chen,K.S. (1998) New generalization of process capability index . Journal of Applied Statistics ,25:801–810.
21. Pearn, W. L.,Lin, C. and Chen,K.S. (2004) The index for asymmetric tolerances:Implications and inference.Metrika 60(2) ,119–136.
22. The Outsourcing Institute (2003). Survey of current and potential outsourcing end-users. Available online at: http://www.outsourcing.com
23. Weerahandi, S. (1993). Generalized confidence intervals. Journal of the American Statistical Association, 88, 899-905.
24. Weber, C. A. (1991). Vendor selection criteria and methods. European Journal of Operational Research, 50, 2-18.
25. Wu, C. W., Shu, M. H., Pearn, W. L. and Liu, K. H. (2008). Bootstrap approach for supplier selection based on production yield. International Journal of Pro-duction Research, 46(18), 5211-5230.
26. Wu, C. W. and Huang, P. H.,(2008). Generalized Confidence Intervals for Com-paring the Capability of Two Processes. Communications in Statistics - Theory and Methods, 39: 13, 2351 -2364.