近幾年來，面板製造產業競爭逐漸激烈，製造工廠紛紛走向自動化生產作業，在一條龍生產模式下，製程站點前後彼此牽制，當其中製程站發生異常時，整條生產線也隨之停擺，進而影響整體生產績效。
因此，本研究藉由蒐集製造工廠中不同生產系統的資訊，如：生產線體、人力派工、產品種類、產品架構及生產狀況等資訊，經由資料整併與處理，使用M5P分類預測法建構模式樹，最後對整棵模式樹做解讀。本研究由分析結果中找到6個判斷生產績效的關鍵因子以及訓練出8條對於產出績效的預測迴歸式。結論出生產線體、生產班別、產品種類、派工人員以及生產機況等，都是影響產出績效的屬性類別，唯有設計架構不是主要影響屬性。進而從模式樹結果中解讀出生產建議，於生產班別條件下，不應將夜班生產作為特殊排程計畫排除條件；從派工人員中找出有兩類別人員須加強訓練；設備部分建議優先針對顯著負影響的四個製程單元做首要調整；物料部分建議增加批次上料單位以降低停線更換頻率為長期對策；在自動生產線體規劃上，建議將點檢工序步驟減少或優化。除了從模式樹找出影響產出績效的關鍵因素使產線管理者有明確方向著手改善，此外迴歸模型提供了現有資源的結果預測及計算最佳預測排線組合，進而達到成本回收及提升生產績效。

The competition in the panel manufacturing industry becomes more intense in recent years. Manufacturing factories thus gradually automate their production processes. When the stations in a production line are interrelated, the whole production line will be stopped when an anomaly occurs in one of the machines at a station. The performance of the production line is greatly influenced by the anomaly. In this study, data resulting from various production systems in manufacturing factories, such as production line configurations, workforce allocation, product types, and product design, was collected for integration and processing. Numeric prediction method M5P is then applied on the data to grow a model tree for interpretation. The model tree has eight linear regression models, and six influencing factors are identified from the tree structure and the regression models. The influencing factors include two types of operators and four manufacturing processes. Managers should pay their attention on those factors for production improvement. Furthermore, the regression models can provide predictions or calculate the optimal lineup combination, thereby achieving cost recovery and enhancing production performance.

英文文獻
Azzeh, M. (2011). Software effort estimation based on optimized model tree. In Promise '11: Proceedings of the 7th International Conference on Predictive Models in Software Engineering (pp. 1-8). New York, NY, United States.
Breiman, L. (1984). Classification and regression trees. (1st ed.). Routledge.
Bhattacharya, B., &Solomatine, D. P. (2005). Neural networks and M5 model trees in modeling water level-discharge relationship. Neurocomputing, 63, 381-396.
Cheah, C. K., Prakash, J., & Ong, K. S. (2020). An integrated OEE framework for structured productivity improvement in a semiconductor manufacturing facility. International Journal of Productivity and Performance Management, 69(5), 1081-1105.
Enzi, A. and Khan, S. A. (2022). Automated production line reliability analysis of the crankshaft manufacturing process. Advances in Science and Technology Research Journal, 16 (1), 15–27.
Frank, E., Wang, Y., Inglis, S., Holmes, G., & Witten, I. H. (1998). Using model trees for classification. Machine Learning, 32, 63-76.
Kutner, M., Nachtsheim, C., &Neter, J. (2004). Applied linear regression models. 4th Edition. McGraw-Hill/Irwin.
Malerba, D., Esposito, F., Ceci, M., &Appice, A. (2004). Top-Down Induction of Model Trees with Regression and Splitting Nodes. IEEE Transactions on Pattern Analysis and Machine Intelligence, 26(5), 612-625.
Roiger, R. J., &Geatz, M. W. (2003). Data Mining: A Tutorial-Based Primer. (2nd ed.). Addison Wesley, New York.
Rencher, A. C., & Christensen, W. F. (2012). Methods of multivariate analysis. (3rd ed.). Wiley Series in Probability and Statistics.
Render, B., Stair, R. J., & Hanna, M. E. (2012). Quantitative analysis for management. Pearson Education.
Solomatine, D. P., & Xue, Y. (2004). M5 model trees and neural networks: Application to flood forecasting in the upper reach of the Huai River in China. ASCE, 9(6), 491.
Soltanali, H., Khojastehpour, M. ,and Farinha, J. T. (2021). Measuring the production performance indicators for food processing industry.Measurement,173, 108394.
Wang, Y., & Witten, I. H. (1997). Inducing model trees for continuous classes. Department of Computer Science, University of Waikato, New Zealand.
Witten, I. H. and Frank, E. (2000). Data mining: practical machine learning tools and techniques with java implementations. San Francisco: Morgan Kaufmann.
中文文獻
Stevenson, W. J. (著)、何應欽(編譯) (2013)。作業管理-精簡版11版，華泰文化事業股份有限公司。
中國生產力中心 (2013)，現場管理實務手冊，初版，財團法人中國生產力中心。
何錦軒、黃榮華、楊長林(2008)，製造部門績效評估系統建構之研究，中國管理評論國際學報，第十一卷第二期，7-12。
李訓兆 (2022)，以線性迴歸與機器學習模型預估Airbnb房源出租價格，長庚大學工商管理學系碩士學位論文。
林君怡 (2015)，運用M5'模式樹分析放款發生逾期因子，國立成功大學高階管理碩士在職專班碩士論文。
林清河 (2020)，工業工程與管理，初版，華泰文化事業股份有限公司。
陳正昌、程炳林、陳新豐、劉子鍵 (2011)，多變量分析方法：統計軟體應用，第6版，五南圖書出版。
康鈞誌 (2009)，運用模式樹建構半導體測試時間之診斷模式，國立成功大學工業與資訊管理學系碩士在職專班碩士論文。
莫凡 (2022)，機器學習演算法的數學解析與Python實現，初版，機械工業出版社。
程一鵬 (2008)，使用模式樹建構SMT錫膏印刷製程品質管制模式之研究，國立成功大學工業與資訊管理學系專班碩士論文。
黃誌成 (2017)，應用模式樹與偏最小平方法預測氧化銦錫薄膜電阻值之研究，國立成功大學工業與資訊管理學系碩士在職專班碩士論文。
賴士葆 (2004)，生產與作業管理，初版，華泰文化事業股份有限公司。
謝孟書 (2014)，應用模式樹比較不動產之實價與銀行鑑價之研究，國立成功大學工業與資訊管理學系碩士在職專班碩士論文。
魏曉云(2007)，決策樹分類方法研究，計算機系統應用，第二卷第9期，42-45。