面板產業中玻璃基板切割製程具高度自動化，現場普遍仰賴 AOI（Automated Optical Inspection）與 EIM（Edge Inspection Machine）進行缺陷檢測，傳統品質監控多依賴靜態閾值判定與人工經驗判讀，面對製程波動、設備退化及跨製程交互影響時，難以即時掌握異常趨勢及潛在風險，尤其在長時間連續生產下，缺陷易呈現累積與擴散現象，進而影響良率與品質穩定性。為改善上述限制，本研究提出一套結合影像辨識、資料探勘與即時預警機制之智慧品質監控架構。
本研究整合 AOI、EIM 與 EDC（Equipment Data Collection），建構多源異質資料整合流程，並透過資料清理、時間對齊與特徵萃取建立分析模型。在影像辨識層面，採用 YOLOv8 建立多類缺陷辨識模型，將原始影像轉換為具一致性之結構化缺陷特徵，以作為後續分析之輸入。在資料探勘層面，本研究以 J48 決策樹建構 OK、CF 與 NG 三類品質狀態分類模型，並結合 Severity Score、類別不平衡處理與代價敏感學習，提升模型對高風險樣本辨識能力。透過 FP-Growth 探勘缺陷之間及缺陷與製程參數之共現關係，以揭示潛在缺陷組合、設備狀態與品質風險之關聯結構。透過影像辨識、品質分類與關聯規則之整合，本研究建立由缺陷偵測、品質預測至異常解釋之完整分析流程，使系統具備預測能力、可解釋性與現場決策支援能力。
研究結果顯示，本架構可有效整合多源資料並提升異常辨識能力。關聯規則分析指出缺陷之間具有顯著共現關係，並與設備補正與刀具磨耗狀態相關，顯示缺陷之間可能存在連鎖共現特徵，並與特定製程狀態具有關聯性。在應用層面，本研究建立以 30 分鐘為週期之即時監控與預警機制，當偵測到異常趨勢時，透過即時通訊平台發送警示，並結合視覺化儀表板呈現品質狀態，以協助工程師快速判斷與改善製程。整體而言，本研究建立多源資料整合與分析架構，並結合影像辨識與資料探勘方法，使系統同時具備品質預測與異常解釋能力，進一步發展具即時預警功能之品質監控系統，提升製程監控即時性與決策效率，並展現其於智慧製造中由事後檢測轉向事前預防之實務價值。

The glass substrate cutting process in the flat panel display industry relies heavily on Automated Optical Inspection (AOI) and Edge Inspection Machine (EIM) systems for defect detection. Traditional quality monitoring methods based on static thresholds and manual interpretation are often insufficient for identifying abnormal trends caused by process variation and equipment degradation. To address these limitations, this study proposes an intelligent quality monitoring framework integrating image recognition, data mining, and real-time early warning mechanisms. The data collected from AOI, EIM, and Equipment Data Collection (EDC) systems were processed by data cleaning, temporal alignment, and feature extraction. Decision tree learning combined with severity score, class imbalance handling, and cost-sensitive scheme was employed for classification when products are categorized as pass, conditional pass, and fail. FP-Growth algorithm was further applied to identify associations among defect types and process conditions. The experimental results demonstrate that the proposed framework effectively enhances anomaly detection and supports quality prediction and interpretation. A real-time monitoring mechanism with a 30-minute decision cycle was implemented to provide timely alerts and visualized decision support. The proposed framework not only facilitates proactive quality management, but also improves monitoring efficiency in smart manufacturing applications.

Agrawal, R., & Srikant, R. (1994). Fast algorithms for mining association rules in large databases. In Proceedings of the 20th International Conference on Very Large Data Bases (pp. 487–499).

Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.

Carion, N., Massa, F., Synnaeve, G., Usunier, N., Kirillov, A., & Zagoruyko, S. (2020). End-to-end object detection with transformers. In European Conference on Computer Vision (pp. 213–229).

Carvalho, T. P., Soares, F. A., Vita, R., Francisco, R. D. P., Basto, J. P., & Alcalá, S. G. (2019). A systematic literature review of machine learning methods applied to predictive maintenance. Computers & Industrial Engineering, 137, 106024.

Chang, J.-R., Chen, Y.-S., Lin, C.-K., & Cheng, M.-F. (2021). Advanced data mining of SSD quality based on FP-Growth data analysis. Applied Sciences, 11(4), 1715.

Chawla, N. V., Bowyer, K. W., Hall, L. O., & Kegelmeyer, W. P. (2002). SMOTE: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16, 321–357.

Girshick, R., Donahue, J., Darrell, T., & Malik, J. (2014). Rich feature hierarchies for accurate object detection and semantic segmentation. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 580–587).

Han, J., Pei, J., & Tong, H. (2022). Data mining: Concepts and techniques (4th ed.). Morgan Kaufmann.

Han, J., Pei, J., & Yin, Y. (2000). Mining frequent patterns without candidate generation. In Proceedings of the ACM SIGMOD International Conference on Management of Data (pp. 1–12).

Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017). MobileNets: Efficient convolutional neural networks for mobile vision applications. arXiv preprint arXiv:1704.04861.

Lin, C. S., Kuo, J., Lin, C. C., Lay, Y. L., & Shei, H. J. (2011). Automatic inspection and strategy for surface defects in the PI coating process of TFT-LCD panels. Assembly Automation, 31(3), 244–250.

Lin, T. Y., Dollár, P., Girshick, R., He, K., Hariharan, B., & Belongie, S. (2017). Feature pyramid networks for object detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 936–944).

Lin, T. Y., Goyal, P., Girshick, R., He, K., & Dollár, P. (2017). Focal loss for dense object detection. In Proceedings of the IEEE International Conference on Computer Vision (pp. 2980–2988).

Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., ... & Sánchez, C. I. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, 42, 60–88.

Liu, W., Anguelov, D., Erhan, D., Szegedy, C., Reed, S., Fu, C. Y., & Berg, A. C. (2016). SSD: Single shot MultiBox detector. In European Conference on Computer Vision (pp. 21–37).

Neo, D. W. K., Liu, K., Huang, R., & Wu, H. (2021). Edge grinding characteristics of display glass substrate. Journal of Manufacturing and Materials Processing, 5(1), 20.

Paiva, R. G., Cavalcante, C. A., & Do, P. (2024). Applying association rules in the maintenance and reliability of physical systems: A review. Computers & Industrial Engineering, 194, 110332.

Qi, S., Yang, J., & Zhong, Z. (2020). A review on industrial surface defect detection based on deep learning technology. In Proceedings of the 2020 3rd International Conference on Machine Learning and Machine Intelligence (pp. 24–30).

Qiu, R. Q., Tsai, M. L., Chen, Y. W., Singh, S. P., & Lo, C. Y. (2024). Integrated automatic optical inspection and image processing procedure for smart sensing in production lines. Sensors, 24(5), 1619.

Quinlan, J. R. (1993). C4.5: Programs for machine learning. Morgan Kaufmann Publishers.

Redmon, J., & Farhadi, A. (2018). YOLOv3: An incremental improvement. arXiv preprint arXiv:1804.02767.

Ren, S., He, K., Girshick, R., & Sun, J. (2015). Faster R-CNN: Towards real-time object detection with region proposal networks. Advances in Neural Information Processing Systems, 28.

Tabernik, D., Šela, S., Skvarč, J., & Skočaj, D. (2020). Segmentation-based deep-learning approach for surface-defect detection. Journal of Intelligent Manufacturing, 31(3), 759–776.

Tan, M., & Le, Q. (2020). EfficientDet: Scalable and efficient object detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 10781–10790).

Tercan, H., & Meisen, T. (2022). Machine learning and deep learning based predictive quality in manufacturing: A systematic review. Journal of Intelligent Manufacturing, 33(7), 1879–1905.

Wang, C. Y., Bochkovskiy, A., & Liao, H. Y. M. (2023). YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 7464–7475).

Yang, Y., Dong, Z., Meng, Y., & Shao, C. (2021). Data-driven intelligent 3D surface measurement in smart manufacturing: Review and outlook. Machines, 9(1), 13.

Yu, Z., Wang, D., & Wu, H. (2025). Defect detection method for large-curvature and highly reflective surfaces based on polarization imaging and improved YOLOv11. Photonics, 12(4).

Zeng, H., Zhang, G., Huang, G., & Wang, G. (2008). Process quality optimization model based on ARM and immune principle. In 2008 International Symposium on Computational Intelligence and Design (Vol. 1, pp. 389–393).

Zhou, L., Pan, S., Wang, J., & Vasilakos, A. V. (2017). Machine learning on big data. Neurocomputing, 237, 350–361.

Zhu, X., Su, W., Lu, L., Li, B., Wang, X., & Dai, J. (2020). Deformable DETR: Deformable transformers for end-to-end object detection. arXiv preprint arXiv:2010.04159.

Zou, Z., Chen, K., Shi, Z., Guo, Y., & Ye, J. (2023). Object detection in 20 years: A survey. Proceedings of the IEEE, 111(3), 257–276.

簡禎富、賀桂芬（2019）。《工業3.5：台灣企業邁向智慧製造與數位決策的戰略》。天下雜誌出版。

曾照銘（2009）。《應用關聯法則於半導體表面黏著技術異常診斷之研究》。元智大學工業工程與管理學系碩士論文。

黃思博（2013）。《應用自動化光學檢測技術於 TFT-LCD 模組瑕疵辨識系統之開發與研究》。國立臺灣科技大學自動化及控制研究所碩士論文。

謝宗達（2020）。《影像辨識於面板產業之產品缺陷檢查》。國立成功大學工程管理碩士在職專班碩士論文。

魏紹偉（2020）。《利用 Apriori 演算法探討異常通報事件關聯規則之初探》。國立彰化師範大學企業管理學系碩士論文。

林祐誠、黃俊翔、羅鈺淳、何昭慶（2023）。基於機器視覺與深度學習的面板切割缺陷檢測。《科儀新知》，235，67–75。

張俊傑、黃崇岳、鄭宏岑（2003）。TFT-LCD 製造中缺陷檢測技術之探討。《科儀新知》，25（2），23–30。