本研究結合流程分析與可能影響PCB鑽孔機效能的因素來提供條件式預防性保養建議以利於減少可能的耗材與工時的損失。本研究開發出PCB(印刷電路板)鑽孔機的條件式預防性保養系統(Condition-Based Preventive Maintenance System，下稱為 CBPMS)。作業時間、斷刀記錄、異常訊息及停頓時間等資訊傳輸到系統之後分析並確認是否已經到達需要進行預防性保養的條件。系統此時也提供建議的保養步驟來避免對機台本身進一步的損害及耗材的浪費。
當CBPMS 自機台收集異常訊息時，這些異常訊息被進行可能發生機台故障機率的分析。本研究以斷刀為主要研究方向並分析可能的斷刀異常訊息以及於四小時內的臨界期間所實際發生的斷刀。亦即是在發生可能的斷刀異常訊息之後，任何於四小時內所發生的斷刀則被視為有效的預測斷刀。如果接著是發生可能的斷刀異常訊息的話，則依舊算作是成功的斷刀預測，因為重覆的可能的斷刀異常訊息是為了提示可能發生的斷刀。此系統預計可達到70%以上的預測斷刀的精確度。並在提示可能發生斷刀時，操作員應該進行系統所建議的預防性保養步驟。

This research combines the process analysis with the factors that affect the performance of the PCB drilling machine to provide condition based preventive maintenance recommendations to reduce the possible material waste and man hour losses. The research developed a condition-based preventive maintenance system (CBPMS) for the PCB drilling machine. The records of the operation time, broken tools, error messages and pauses of each run, etc., are then sent to the system to be analyzed and checked if it meets the conditions for preventive maintenance. The system then provides the recommended actions that should be taken to prevent further damage to the machine or waste of materials.
When the CBPMS collects the error data from the machines, the errors are analyzed to determine the possibility of the machine to experience down time. This research focuses on the down time caused by the tool breakage and analyzes the occurrences of the possible tool breakage errors and the actual tool breakages within a four-hour threshold. That is, when a possible tool breakage error is flagged, any subsequent tool breakage within four hours is considered a successful prediction of the tool breakage. If any subsequent possible tool breakage error is flagged within the four hour threshold of the previous possible tool breakage error, then it is still considered a successful prediction of the tool breakage as the possible tool breakage errors are repeatedly flagged to notify the possible tool breakage. The system is expected to achieve more than 70% accuracy when predicting possible upcoming tool breakages. Upon receipt of the tool breakage notification, the operator should carry out the suggested actions for preventive maintenance.

[1] Tongtai Machine & Tool Co., Ltd. http://tongtai.com.tw

[2] Sieb & Meyer AG, http://www.sieb-meyer.de

[3] B.F. Arnold, ”Staticstical Tests Optimally Certain Fuzzy Requirements on the Power Function and on the Sample Size”, Fuzzy Sets and System, Vol. 75, Issue 3, pp 365-372,1996.

[4] W. Li, D. Li, J. Ni," Diagnosis of tapping process using spindle motor current”, International Journal of Machine Tools and Manufacture, Vol. 43, Issue 1, pp 73-79, Jan. 2003

[5] N. Casoetto, D. Djurdjanovic ,“Multisensor Process Performance Assessment through Use of Autoregressive Modeling and Feature Maps”, Journal of Manufacturing Systems, Vol. 22, No.1, pp 1-9, 2003

[6] S. Dey, J. A. Stori,” A Bayesian network approach to root cause diagnosis of process variations”, International Journal of Machine Tools and Manufacture, Vol. 45, Issue 1, pp 75-91, Jan. 2005

[7] J. Sun, G.S. Hong, Y.S. Wong, M. Rahman, Z.G. Wang,” Effective training data selection in tool condition monitoring system”, International Journal of Machine Tools and Manufacture, Vol. 46, Issue 2, pp 218-224, Feb. 2006

[8] S.H. Yang, Ki-Hoon Kim, Yong Kuk Park,“Measurement of spindle thermal errors in machine tool using hemispherical ball bar test”, International Journal of Machine Tools and Manufacture, Vol. 44, Issues 2-3, pp 333-340, Feb. 2004

[9] O. Rothe, “ISMI Next Generation Factory e-Manufacturing Workshop”, presentation document, International SEMATECH, 2008

[10] D. Stark, “Prognostics and Health Management (PHM)”, presentation document, International SEMATECH, 2010

[11] Q.H. Chiu, “Tool WPH Monitor and Management System and Its Empirical Study in Semi Conductor Manufacturing”, National Tsing Hua University, 2005

[12] S.C. Chuang, “The Wear Analysis of End Milling Tool”, Yuan Ze University, 2005

[13] T.T. Huang, “Investigation on the Predictive Model for Tool Wear and Surface Roughness in Milling SKD11 Steel”, National Pingtung University of Science and Technology, 2003

[14] S.P. Kuo, “Study on High Speed Milling with Vibration using Coated Tools”, National Yunlin University of Technology and Science, 2005

[15] J.B. Liang, “A Study of Optimization Parameters for Tool Life”, Chienkuo Technology University, 2007

[16] Z.W. Lin, “Machine Tool Performance Prediction and Diagnosis System Development”, National Cheng Kung University, Aug. 2010

[17] H.L. Peng, “A Machine Alarm Management System for Semi Conductor Manufacturing”, 2008

[18] C.S. Tsai, “The Investigation on the Predictive Model for the Wear of Cutting Tool and the Optimized Model for Parameters Based on Cutting 17-4PH Stainless Steel”, National Pingtung University of Science and Technology, 2002

[19] S.H. Yeh, “Development of a Machine Vision System for Cutting Tool Wear Detection”, I-SHOU University, 2008