錫膏印刷為表面黏著技術(Surface Mount Technology, SMT)之第一個子製程。許多研究指出，落錫量不佳即佔了SMT的不良原因52%至71%。以往製程之參數設定與材料選擇大多以實驗設計來進行，不僅耗時且隱藏成本高，如何利用現存之製程資料並從中發掘出有意義之預測模型實為一重要的課題。本研究應用模式樹並以網板開孔形狀做為控制變數，將五處量測點、錫膏品牌(包含金屬成分、黏滯性與微粒尺寸)、間距(snapoff)、網板厚度、刮刀材質、刮刀角度、刮刀壓力、印刷速度、相對印刷方向以及相對焊墊大小等10種屬性做為輸入變數，並以品檢之錫膏落錫量檢測做為輸出變數，來進行SMT錫膏印刷製程品質管制系統預測模式之建構。最後以間距、網板厚度與刮刀材質做為屬性分割點，並發現錫膏品牌與相對焊墊大小等兩項並非顯著屬性。本研究最後亦與分類迴歸樹做預測能力之比較，結果證明模式樹的預測，確實有較佳的預測效果。

Solder paste printing is the first assembly process of surface mount technology (SMT). Researches had been shown that 52%-71% of the assembly defects were caused by the improper setup of the printing process. The former setup of process parameter and material choice are most proceeding by design of experiment (DOE). This will be not only waste time also high cost. It is an important issue for using current process data and how to deal with it that becomes the meaning predict model. This research applies model tree and use stencil open hole shape to be the control variable. The five measure points, solder paste brand (include metal ingredient, stickiness and micro-particle size), snapoff, stencil thickness, squeegee material, squeegee angle, squeegee pressure, print speed, relative print direction and relative pad size, amount to ten attributes, will be the input variables. The checkout solder paste quantity will be the output variable. We construct the quality control forecast system model for proceeding SMT solder paste printing process. Finally, we use the snapoff, stencil thickness and squeegee material for the attribute partition point, and find both solder paste brand and relative pad size are two statistical insignificance attributes. This research also makes the comparison with forecasting ability to use Classification and Regression Tree (CART). Finally, it indeed actually has better forecasting effect.

參考文獻

陳建智，應用模式樹建構印刷電路板裝配之表面黏著製程品質管制模式之研究，國立成功大學工業與資訊管理研究所碩士論文，民國96年6月。
葉俊吾，運用類神經網路建構SMT錫膏印刷製程品質管制系統，國立成功大學製造工程研究所碩士論文，民國91年5月。
Alexander, W. P. and Grimshaw, S. D. (1996). Treed regression. J. Comput. Graph. Statist. 5, 156-175.
Arulvanan, P., Zhong, Z., Shi, X. (2006).Effects of process conditions on reliability, microstructure evolution and failure modes of SnAgCu solder joints. Microelectronics and Reliability, Vol.46, Issues 2-4, 432-439.
Breiman, L., Friedman, J. H., Olshen, R. A., and Stone, C. J. (1984). Classification and Regression Trees . New York：Chapman and Hall.
Chaudhuri, P., Huang, M. C., Loh, W. Y.,and Rubin, R. (1994). Piecewise-polynomial regression tree, Statistica Sinica, Vol.4, 143-167.
Dobra, A.,and Gehrke, J. E. (2002). SECRET：A Scalable Linear Regression Tree Algorithm. Proceedings of the Eighth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 481-487.
Donato Malerba, A. Appice, M. Ceci, and M. Monopoli, (2002).“Trading-Off Local versus Global Effects of Regression Nodes in Model Trees,” Proc. Foundations of Intelligent Systems, 13th Int’l Symp., H.-S. Hacid, Z. Ras, D. Zighed, and Y. Kodratoff, eds., pp. 393-402.
Donato Malerba, Floriana Esposito, Michelangelo Ceci, and Annalisa Appice, (2004).“Top-Down Induction of Model Trees with Regression and Splitting Nodes ” IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 26, NO. 5.
Fujiuchi, S., and Toriyama, K. (1994). Collective screen printing for carrier bump and SMT pads. Proceedings of IEEE/CPMT International Electronic Manufacturing Technology Symposium, 109-112.
Han, J., and Kamber, M. (2001). Data Mining：Concepts and Techniques. San Francisco：Morgan Kaufmann.
Ho, S. L., Xie, M., Tang, L. C., Xu, K., and Goh, T. N. (2001). Neural network modeling with confidence bounds: a case study on the solder paste deposition process. IEEE Transactions on Electronics Packaging Manufacturing, Vol.24,Issue 4, 323-332.
Jung, S.W., Jung, J.P., Zhou, Y. (2006).Characteristics of Sn–Cu Solder Bump Formed by Electroplating for Flip Chip. IEEE Transactions on Electronics Packaging Manufacturing, Vol.29, Issue 1, 10-16 .
Karalic, A. (1992). Employing linear regression in regression tree leaves. Proceedings of the 10th European Conference on Artificial Intelligence, 440-441.
Krishnan, A. S., Srihari, K. (1995). A knowledge-based object oriented DFM advisor for surface mount PCB assembly. The International Journal of Advanced Manufacturing Technology, Vol.10, 317-329.
Kusiak, A. and Kurasek, C. (2001). Data mining of printed-circuit board defects. IEEE Transactions on Robotics and Automation, Vol.17, No.2, 191-196.
Lau, F. K. H. and Yeung, V. W. S. (1997). A hierarchical evaluation of the solder paste printing process. Journal of Materials Processing Technology, Vol.69, 79-89.
Lee, N. C. (1999). Optimizing the reflow profile via defect mechanism analysis. Soldering and Surface Mount Technology, Vol.11, 13-20.
Lewis, C. D. (1982). Industrial and Business Forecasting Method, London：Butterworth Scientific.
Lideen, J.D., Dahl, A.O.,(1994).Printing techniques for fine pitch screen printing, Proceedings of the Technical Program- NEPCON West, 1862-1877.
Liu, Y., Tian, G., Gale, S., Johnson, R.W., Crane, L.(2006).Lead-free chip scale packages: assembly and drop test reliability. IEEE Transactions on Electronics Packaging Manufacturing, Vol.29, Issue 1, 1-9.
Loh, W. Y. (2002). Regression trees with unbiased variable selection and interaction detection. Statistica Sinica, Vol.12, 361-386.
Mannan, S. H., Ekere, N. N., Ismail, I., and Ko, E. K. (1994). Squeegee Deformation Study in the Stencil Printing of Solder Pastes. IEEE Transactions on Components, Hybrids, and Manufacturing Technology- Part A, Vol.17, No.3, 470-475.
Markstein, H. W. (1997). Controlling the Variables in Stencil Printing. Electronic Packaging and Production, Vol.37, No.2, 48-56.
Montgomery, D. C., Keats, J. B., Perry, L. A., Thompson, J. R., and Messina, W. S. (2000). Using statistically designed experiments for process development and improvement: an application in electronics manufacturing. Robotics and Computer-Integrated Manufacturing, Vol.16, Issue 1, 55-63.
Pan, J., Tonkay, G. L., Storer, R. H., Sallade, R.M., and Leandri, D. J. (2004). Critical variables of solder paste stencil printing for micro-BGA and fine-pitch QFP. IEEE Transactions on Electronics Packaging Manufacturing, Vol.27,Issue 2, 125-132.
Poon, N. M., Wu, C. M. L., Lai, J. K. L., Chan, Y.C., (2000).Residual shear strength of Sn-Ag and Sn-Bi lead-free SMT joints after thermal shock, IEEE Transactions on Advanced Packaging, Vol.23, Issue 4, 708-714.
Quinlan, J. R. (1992). Learning with continuous classes. Proceedings of the Australian Joint Conference on Artificial Intelligence, 343-348.
Quinlan, J. R. (1993). Combining instance-based and model-based learning. Proceedings of the 10th International Conference on Machine learning, 236-243.
Ramkumar, S. M., Ghaffarian, R., and Varanasi, A. (2006). Lead-free 0201 manufacturing, assembly and reliability test results. Microelectronics and Reliability, Vol.46, Issues 2-4, 244-262.
Roiger, R. J., and Geatz, M. W. (2003). Data Mining：A Tutorial-Based Primer. New York：Addison Wesley.
Shapiro, A. A., Bonner, J. K., Ogunseitan, O. A., Saphores, J.-D. M., Schoenung, J. M.(2006).Implications of Pb-free microelectronics assembly in aerospace applications. IEEE Transactions on Components and Packaging Technologies, Vol.29, Issue 1, 60-70.
Soto, H. P. (1998). Study of the thermal behavior of a printed circuit board during the reflow process inside a surface mount technology oven. Unpublished master dissertation, American University of Puerto Rico, Bayamon.
Su, Y. Y.,and Srihari, K. (1996). Placement sequence identification using artificial neural networks in surface mount PCB assembly. The International Journal of Advanced Manufacturing Technology, Vol.11, 285-299.
Torgo, L. (1997). “Functional Models for Regression Tree Leaves,” Proceedings of International Machine Learning Conference, D. Fisher (Ed), Morgan Kaufmann, San Mateo, CA,
Tsai, T. N.,and Yang, T. (2005). A neuro-computing approach to the thermal profile control of the second-side reflow process in surface mount assembly. Journal of Manufacturing Technology Management, Vol.16, Issue.3 343-359.
Tsai, Y. S., King, P. H., Higgins, M. S., Pierce, D., and Patel, N. P. (1997). An expert-guided decision tree construction strategy：An application in knowledge discovery with medical database. AMIA Annual fall symposium, 280-12.
Whitmore, M., Mackay, C., and Hobby, A. (1997). Plastic Stencils for Bottom-Side Chip Attach. Electronic Packaging and Production, Vol.37, No.13, 68-72.
Wang, Y.,and Witten, I. H. (1997). Inducing model trees for continuous classes. Proceedings of poster papers of 9th European Conference on Machine Learning.
Witten, I. H., and Frank, E. (2005). Data Mining：Practical Machine Learning Tools and Techniques (2nd ed.). San Francisco：Morgan Kaufmann.
Wu, C. M. L., Huang, M. L., (2005). Microstructural evolution of lead-free Sn-Bi-Ag-Cu SMT joints during aging, IEEE Transactions on Advanced Packaging, Vol.28, Issue 1, 128-133.
Yang, T., Tsai, T. N., and Yeh, J. (2005). A neural network-based prediction model for fine pitch stencil-printing quality in surface mount assembly. Engineering Applications of Artificial Intelligence, Vol.18, Issue 3, 335-341.