近年來，觸控面板產業急速地興起，這個新興的輸入介面正漸漸影響人們對各種電子產品的使用習慣。台灣廠商亦積極希望進入此市場提升公司業績。但觸控面板因大多搭載平面顯示器使用，故在外觀上必須趨近透明不可見，故各種外觀視覺上的缺陷、異常均需要以嚴格的標準檢驗。目前國內各廠商均大量使用作業員以目視檢查方式進行檢驗。但因異常類別種類繁多，且人員對異常判定的標準不一，同樣的異常可能會有不同的判定結果而產生異常分析的困難，亦增加了檢驗員訓練的難度。
本研究著重改善現有的觸控面板目檢教育訓練與生產流程，以提高檢驗人員的教育訓練通過率並改善檢查所造成的產品報廢率。首先透過現場觀察、資訊蒐集的方法了解目前訓練狀況，運用強化記憶印象的通則對教材進行重新編輯，運用分類、顏色、圖形、差異化、多重感官記憶等方法確實有助改善人員學習效率，提升檢驗人員的鑑定成績及通過鑑定比例，學科筆試成績在新教材導入後，首次測驗平均成績由81.5進步至92.2分。在實務操作面，則利用人眼辨識視覺感知特性，設計對比背景屏幕提高人員檢出率，並用不同訓練片及教授經驗法則提高人員判等正確性，術科操作成績在新教材導入後，前三次平均成績由67.4分進步至76.5分。針對現場作業流程的手法，設計新的載具及包裝方式，有效改善因檢查產生之報廢，在現場改善治具、手法導入後報廢率由8.2%下降至4.9%均有顯著成效。
本研究提供一簡單的綜合型運用案例，綜整有關人眼視覺辨識及教育訓練方法的相關背景知識，加以應用在個案公司所面臨的實際問題上，此改善流程可供後續各種應用人眼目視檢查的單位作為績效提升的參考。

In recent years, the demand of touch panel increases rapidly. The new input interface is gradually affecting people's usage of various electronic products. Touch panel is often equipped with flat-panel displays, so it must be transparent well and invisible. The defect of touch panel appearance has to be inspected critically. At present, the appearance quality of touch panel has been determined subjectively by means of human visual inspection. However, large amount of various defects make opeartors hard to remenber, and different opeartors may judge the same defect with different defect code. As a result, it makes inspection outcomes faithless and increases the difficulty of inspector training.
This research focuses on the improvement of touch panel inspection training and inspected process to increase inspector’s certificated ratio and reduce products scrapped ratio. Through factory on-site obervation and consulting with experienced and new inspectors, the training situation was realized. The training materials were reedited using the rule of memory. Such as “Classified defects”, “different color impression”, “illustration”, “different layout”, and “mulit-perception”. The modified training materials have been verified to be truly helpful for inspector’s learning and increase certificated ratio. At practical inspection workstation characteristics of visual perception ( different luminance contrast ) was applied to increase hit rate. The written test average score from 81.5 to 92.2, the operation test average score from 67.4 to 76.5. Besides, a new carrier and packaging have been designed to reduce product damage, the scrap ratio from 8.2% to 4.9%.
This research provides a simple hybrid application case. Applying knowledge of human visual inspection on the modification of training program, a practical inspection problem has been solved. This improving process can be generalized to other inspection station.

Bushnell, D.S., “Input, Process, Output：A Model For Evaluating Training”, Training & Development Journal, March 1990 ,p41-43.(1990)
Chang, J. J. 2005 “The Development of a Training Expert System for TFT-LCD Defects Inspection” M. S. Thesis o f Department of Industrial Engineering and Engineering Management, National Tsing Hua University, Taiwan, ROC.
Chen, C.C. (2005). “A Study of Human Visual Threshold on Mura Defects in LCD”. M. S. Thesis of Department of Industrial Engineering and Engineering Management, National Tsing Hua University, Taiwan, ROC.
Drury, C.G. and Sinclair, M.A., “ Human and machine performance in an inspection task”. Human Factors, 25(4), p391-400 (1983).
Gramopadhye, A.K., Drury, C.G., & Sharit, J., “Feedback Strategies for visual search in airframe structural inspection”. Industrial Ergonomics, 19(5), p333-344 (1997).
Hata, S., Miyashita, Y. and Hanafusa, H., “ Human sensitivity of color defects inspection”, Proc Industrial Electronics, Control, and Instrumentation, 2, p713-718 (1996).
Hou, T., Lin, L., and Drury, C.G., “An empirical study of hybrid inspection systems and allocation of inspection functions”. Human Factors in Manufacturing, 3, p351-367 (1993).
Jiang, X.C., Gramopadhye, A.K., Melloy, B.J., and Grimes, L.W., “Evaluation of best system performance: Human, automated, and hybrid inspection systems”. Human Factors and Ergonomics in Manufacturing & Service Industries, 13(2), p137-152 (2003).
Krantz, J. H., Silverstein, L. D. and Yeh, Y.Y. “ Visibility of Transmissive Liquid Crystal Displays under Dynamic Lighting Conditions”, Human Factors, 34 (5), p615-632 (1992).
Lee, D.G., Kim, I.H., Jeong, M.C., Oh, B.K., and Kim, W.Y., “ Mura analysis method by using JND luminance and the SEMU definition”, Proc Int Disp Workshops, Japan. 10, p1467-1470 (2003).
Marcus, N., Cooper, M., & Sweller, J. “Understanding instruction”, J. of Educational Psychology, 88(1), p81-97. (1996)
Miller, G.A., “The magical number seven, plus or minus two”. Psychological Review, 63,p81-97. (1956)
Ojanpää, H. and Näsänen, R., “ Effects of luminance and colour contrast on the search of information on display devices”. Displays, 24, p167-178 (2003).
SEMI, “New Standard: Definition of Measurement and Index (SEMU) for Luminance Mura in FPD Image Quality Inspection”, SEMI Draft Document #3324 (2002).
Tamura, T., Tanaka, K., Baba, M., Suzuki, M. and Furuhata, T., “ Just Noticeable Difference(JND) Contrast of “Mura” in LCDs on the Five Background Luminance Levels”, Proc Int Disp Workshops, Japan. 11, p1623-1626 (2004).
Tu, C.W. 2005 “Analysis of Mura in TFT-LCD”. M. S. Thesis of Department of Electrical Engineering, National Central University, Taiwan, ROC.
Wang, M.J., Lin, S., And Drury, C.G., “Training for strategy in visual search”. Industrial Ergonomics, 20(2), p101-108 (1997).
ITBS觸控面板研究資料，2009
Micro chip Technology Incorporated研究資料，2008
玉兒光雄著 蘇楓雅譯，“用腦科學打造記憶力：這樣記不會忘”， 出版社：天下遠見 台北市(2009)
李再長、黃雪玲、李永輝、王明揚著，“人因工程” 二版，出版社：華泰文化 台北市(2011)
奇美電子內部教育訓練教材，2011
電子工程專輯 http://www.eettaiwan.com/，2011
戴維思著，“超強記憶訓練”，出版社：如何 台北市(2002)
顏世霖 2003 “訓練方式與訓練講師對訓練成效影響之研究-以製造業為例” 成功大學碩士論文
蘇琬婷 2008 “量測多倍視覺差異對比閾值建立液晶顯示器不均勻現象之分級標準” 清華大學碩士論文。