半導體封裝測試產業中，成型切割站點為一個對品質極為敏感的製程環節。切割刀片隨著使用時間或切割距離的增加，會逐漸產生磨耗，使產品樣本寬度出現平均值上升與變異擴大的現象。這類變異具有明顯的趨勢性特徵，其影響將隨時間累積，最終導致產品品質劣化與超規風險升高。當製程輸出隨時間發生系統性改變時，傳統以常態穩定假設為前提的製程能力指標將無法反映製程的實際能力，進而產生誤判風險。傳統的刀片更換策略通常依據固定使用時間或累積切割距離，未能考量實際產品輸出的變異趨勢，無法即時因應當下製程能力的下降。容易導致過早更換而提高成本，或延遲更換而引發品質問題。本研究針對上述問題，提出一套結合趨勢變異修正製程能力指標與總擁有成本模型的刀片更換決策架構。將樣本寬度平均值與標準差的時間變動納入製程總變異的計算中，進一步修正製程能力指標，使其能反映趨勢性劣化對製程能力的影響。為量化換刀時機，本研究建立動態成本模型，整合刀片成本、維護與停機成本及品質損失成本，模型透過每期樣本的平均值與標準差進行線性迴歸，估算趨勢斜率與總變異，進而修正製程能力指標並比較不同換刀時機下的平均成本，以判斷是否需立即換刀或延後一週期。此方法為企業提供一套可即時反映樣本結果的決策工具，能更精確評估趨勢變異對品質的潛在威脅，並兼顧經濟成本與風險控制。相較於傳統以固定週期或經驗法則為主的換刀方式，本研究模型可改善決策依據不足、製程能力高估與品質風險未預警等問題，協助工程師依據實際樣本資料做出最佳化換刀決策，兼顧品質穩定與成本效益。

In the semiconductor packaging and testing industry, the molding and singulation station is highly sensitive to product quality. As blade usage time or cutting distance accumulates, wear occurs, increasing both the mean and variance of sample width. This time-dependent systematic change, defined by Xie et al., (2002) as trend variance, leads to progressive quality degradation and greater risk of out-of-spec products. However, traditional process capability indices assume process stability and normality, failing to capture such trend-related variations and thus resulting in overly optimistic assessments. Conventional blade replacement strategies based on fixed cutting distances overlook these variation trends, potentially causing premature replacement or delayed quality responses. To address this, this study proposes a blade replacement decision model integrating a trend-adjusted process capability index with a total cost of ownership (TCO) framework. Time-based changes in mean and variance are incorporated into total process variation to correct the index. A dynamic cost model combining blade, maintenance, and quality loss costs estimates optimal replacement timing. Linear regression predicts future variation and compares average costs across replacement cycles. The proposed approach offers a more accurate and cost-effective method for managing blade wear and maintaining process quality.

蘇紹榕，刀具剩餘壽命估測技術之研發，國立中正大學機械工程系研究所學位論文，2019
葉伯霆，切削加工中考量刀具成本與換刀損失之研究，逢甲大學工業工程與系統管理學系學位論文，2023
Adler, Y., Shper, V., & Maksimova, O. (2011). Assignable causes of variation and statistical models: Another approach to an old topic. Quality and Reliability Engineering International, 27(5), 623-628.
Antony, J. (2004). Six sigma in the UK service organizations: Results from a pilot survey. Managerial Auditing Journal, 19(8), 1006-1013.
Bebr, L., Bícová, K., & Zídková, H. (2017). Use of the ppm and its function in the production process. Procedia Manufacturing, 13, 608–615.
Busch, J. (1994). Cost modeling as a technical management tool. IEEE Transactions on Components, Packaging, and Manufacturing Technology: Part C, 17(1), 6–12.
Chen, C. C., & Lyu, J. (2009). A lean Six-Sigma approach to touch panel quality improvement. Production Planning & Control, 20(5), 445-454.
Conrad, C. J., & McClamroch, N. H. (1987). The drilling problem: A stochastic modeling and control example in manufacturing. IEEE Transactions on Automatic Control, 32(11), 947–956.
Degraeve, Z., Roodhooft, F., & van Doveren, B. (2005). The use of total cost of ownership for strategic procurement: A company-wide management information system. Journal of the Operational Research Society, 56(1), 51–59.
Hao, L., Bian, L., Gebraeel, N., & Shi, J. (2017). Residual life prediction of multistage manufacturing processes with interaction between tool wear and product quality degradation. IEEE Transactions on Automation Science and Engineering, 14(2), 1211-1223.
Hendricks, K. B., & Singhal, V. R. (1997). Does implementing an effective TQM program actually improve operating performance? Empirical evidence from firms that have won quality awards. Management Science, 43(9), 1258-1274.
Jarrell, G., & Peltzman, S. (1985). The impact of product recalls on the wealth of sellers. Journal of Political Economy, 93(3), 512–536.
Kane, V. E. (1986). Process capability indices. Journal of Quality Technology, 18(1), 41–52.
Lee, S. L., Liew, C. Y., Chen, C. K., & Voon, L. L. (2022). Geometric Brownian motion-based time series modeling methodology for statistical autocorrelated process control: Logarithmic return model. International Journal of Mathematics and Mathematical Sciences, 2022, Article 4783090
Meeker, W. Q., Escobar, L. A., & Pascual, F. G. (2021). Statistical methods for reliability data (2nd ed.). Wiley.
Montgomery, D. C. (2019). Introduction to statistical quality control (8th ed.). Wiley.
Øksendal, B. (2003). Stochastic differential equations: An introduction with applications (6th ed.). Springer.
Rolluqui, L. L., Cabading Jr, P. A., & Domingo, J. P. (2021). Increasing Singulation Machine UPH through Characterization and Standardization of Saw and Handler Parameters through DMAIC Methodology. Journal of Engineering Research and Reports, 20(3), 123–140.
Vijchulata, P. (2003). Unique fiducial designs for CSP singulation process. IEEE/CPMT International Electronics Manufacturing Technology Symposium, 7-12
Xie, M., Tsui, K. L., Goh, T. N., & Cai, D. Q. (2002). Process capability indices for a regularly adjusted process. International Journal of Production Research, 40(10), 2367-2377.
Ye, Z.-S., & Xie, M. (2014). Stochastic modelling and analysis of degradation for highly reliable products. Applied Stochastic Models in Business and Industry, 30(1), 16–32.