驗收抽樣系統（Acceptance Sampling System）是由兩個以上之驗收抽樣計畫（Acceptance Sampling Plan）及轉換規則所組成，提供更有彈性之品質驗收程序，亦為品質管制中一項重要之工具，用來決定允收或拒絕該受檢貨批，以保證允收貨批之品質。傳統的品質管制觀念中，只要產品之品質特徵落於上下規格界限內即可稱為「良品」。田口博士( Dr. Genichi Taguchi )於1950年代所提出品質特徵值偏離目標值即會產生損失，並建議以二次損失函數來衡量品質水準。因此，本研究擬利用損失函數 進行建構驗收抽樣系統，此驗收抽樣系統中含有兩個驗收抽樣計畫，又稱為Tightened-Normal-Tightened Sampling System（TNT驗收抽樣計畫），系統中包含：加嚴檢驗（Tightened Inspection）與正常檢驗（Normal Inspection），首次檢驗均由加嚴檢驗開始，當在加嚴檢驗中連續 次允收即可轉換至正常檢驗，若正常檢驗之下一次被拒絕後 次內又被拒絕，則需轉換回加嚴檢驗，利用此規則讓不同品質條件之貨批在加嚴檢驗與正常檢驗中互相轉換，更能保障買賣雙方之風險與有效地減少驗收抽樣之成本。
針對系統參數部分，本研究探討 TNT-(nT=mnN,nN;k)、TNT-(nT,nN;k) 與 TNT-(n;kT,kN)等三種不同型態。除針對各型態之抽樣系統分析其操作特性曲線（Operating Characteristic Curve, OC Curve）及平均抽樣樣本數（Average Sample Number, ASN）外，並與傳統單次抽樣計畫進行比較。最後，透過實務案例之分析與說明，以提供使用者在應用驗收抽樣系統之操作程序及準則。

Acceptance sampling system plays an important role in quality control to make sure consumers to accept products with good quality, it contains two or more acceptance sampling plans and provides more flexible acceptance sampling procedure than existing acceptance sampling plans. Most existing acceptance sampling plans were developed mainly based on the product defective rate or process yield. However, the yield management does not distinguish among the product that fall inside of the specification limits. The process loss index was proposed for measuring process performance by considering the concept of quality loss function introduced by Dr. Genichi Taguchi. This thesis based on process loss index developed an acceptance sampling system contains two acceptance sampling plans: tightened inspection and normal inspection. The first sampling procedure needs to begin from tightened inspection, and switch to normal inspection if lots in a raw are all accepted. If there is an additional lot is rejected in the next lots after a rejection under normal inspection, it needs to switch to tightened inspection. The acceptance sampling system observes the switching condition from tightened inspection to normal inspection and switch back to tightened inspection, so it is also called “Tightened-Normal-Tightened Sampling System （TNT Sampling System）”.
The behavior of the proposed three types sampling system: TNT-(nT=mnN,nN;k), TNT-(nT,nN;k) and TNT-(n;kT,kN) are investigated and discussed in this paper. The results for sampling systems indicate that the proposed sampling system gives the desired protection with smaller sample size than traditional variables single sample plan. Lastly, we analyzed a practical case to assist users to learn more from the proposed sampling system.

1. Aslam, M., Yen, C. H., Chang, C. H., and Jun, C. H. (2013). Multiple states repetitive group sampling plans with process loss consideration. Applied Mathematical Modelling, 37(20), 9063-9075.
2. Aslam, M., Yen, C. H., Chang, C. H., and Jun, C. H. (2014). Multiple dependent state variable sampling plans with process loss consideration. The International Journal of Advanced Manufacturing Technology, 71(5-8), 1337-1343.
3. Aslam, M., Yen, C. H., Chang, C. H., Jun, C. H., Ahmad, M., and Rasool, M. (2012). Two-stage variables acceptance sampling plans using process loss functions. Communications in Statistics-Theory and Methods, 41(20), 3633-3647.
4. Aslam, M., Yen, C. H., and Jun, C. H. (2011). Variable repetitive group sampling plans with process loss consideration. Journal of Statistical Computation and Simulation, 81(11), 1417-1432.
5. Balamurali, S., and Jun, C. H. (2009). Designing of a variables two-plan system by minimizing the average sample number. Journal of Applied Statistics, 36(10), 1159-1172.
6. Calvin, T. (1977). TNT zero acceptance number sampling. Paper presented at the American Society for Quality Control Annual Technical Conference Transactions.
7. Chan, L. K., Cheng, S. W., and Spiring, F. A. (1988). A new measure of process capability: Cpm. Journal of Quality Technology, 20(3), 162-175.
8. Dodge, H. F., and Romig, H. G. (1959). Sampling inspection tables. New York: John Wiley & Sons. 2nd.
9. Dumicic, K., Bahovec, V., and Zivadinovic, N. K. (2006). Studying an OC curve of an acceptance sampling plan: a statistical quality control tool. Paper presented at the Proceedings of the 7th WSEAS International Conference on Mathematics and Computer in Business and Economics.
10. Johnson, T. (1992). The relationship of Cpm to squared error loss. Journal of Quality Technology, 24(4), 211-215.
11. Kane, V. E. (1986). Process capability indices. Journal of Quality Technology, 18(1), 41-52.
12. Montgomery, D. C. (2009). Introduction to Statistical Quality Control: John Wiley and Sons. 6th.
13. Muthuraj, D., and Senthilkumar, D. (2006). Designing and construction of tightened-normal-tightened variables sampling scheme. Journal of Applied Statistics, 33(1), 101-111.
14. Pearn, W. L., Chang, Y., and Wu, C. W. (2004). Distributional and inferential properties of the process loss indices. Journal of Applied Statistics, 31(9), 1115-1135.
15. Pearn, W. L., Chang, Y., and Wu, C. W. (2006). Measuring process performance based on expected loss with asymmetric tolerances. Journal of Applied Statistics, 33(10), 1105-1120.
16. Pearn, W. L., and Chen, K. S. (1997). A practical implementation of the process capability index Cpk. Quality engineering, 9(4), 721-737.
17. Pearn, W. L., Chang, Y. C., and Wu, C. W. (2006). Multiple-process performance analysis chart based on process loss indices. International journal of systems science, 37(7), 429-435.
18. Radhakrishnan, R., and Sivakumaran, P. (2010). Construction and Selection of Tightened-Normal-Tightened Schemes of Type TNT-(n1, n2; c) indexed through Six Sigma Quality Levels. Paper presented at the Proceedings of the 2010 International Conference on Industrial Engineering and Operations Management (IEOM 2010), Dhaka, Bangladesh.
19. Schilling, E. G. (1982). Acceptance Sampling in Quality Control: CRC Press.
20. Senthilkumar, D., & Muthuraj, D. (2010). Construction and selection of tightened–normal–tightened variables sampling scheme of type TNTVSS (n1, n2; k). Journal of Applied Statistics, 37(3), 375-390.
21. Soundararajan, V., and Vijayaraghavan, R. (1992). Construction and selection of tightened-normal-tightened sampling inspection scheme of type TNT-(n1, n2; c). Journal of Applied Statistics, 19(3), 339-349.
22. Taguchi, G. (1986). Introduction to Quality Engineering: Designing Quality into Products and Processes. Asian Productivity organization, Tokyo.
23. Tsui, K. L. (1997). Interpretation of process capability indices and some alternatives. Quality engineering, 9(4), 587-596.
24. Vijayaraghavan, R., and Soundararajan, V. (1996). Procedures and tables for the selection of tightened-normal-tightened (TNT-(n; c1, c2)) sampling schemes. Journal of Applied Statistics, 23(1), 69-80.
25. Yen, C. H., and Chang, C. H. (2009). Designing variables sampling plans with process loss consideration. Communications in Statistics-Simulation and Computation, 38(8), 1579-1591.
26. 莊寶鵰, 與黃信穎. (2008). 考慮品質損失與量測系統之預防型抽樣計畫.
27. 鄭春生, (2008). 品質管理: 修訂版: 全華圖書.