臺灣博碩士論文加值系統

過去研究發現，建構即時排程系統知識庫如果能擁有較好的分類正確率，會擁有較佳的生產績效。在機器學習為基礎的排程知識庫有兩項議題對歸納能力有顯著影響：（1）屬性子集合的篩選（2）分類器的集成。
屬性子集合的篩選，可被視為最佳化問題；其求解過程涉及以測試資料集的分類正確率為目標值。以搜尋求解最佳屬性的子集合。分類的集成是將個別不同分類器的預測以相等或不相等權重的方式結合以進行分類預測。過去文獻中，在許多的案例證明使用集成模式，相對於各別分類器在分類正確率有顯著改善的成效。
本研究的目的是將三種不同的機器學習演算法（BPNN,DT,and SVM）以GA進行屬性篩選並同時進行參數最佳化建立出機器學習分類器，然後以多數決的投票機制將三種分類器進行集成並命名為GA+Voting分類器。最後將GA+Voting相對於機器學習為基所建構的分類器進行線上模擬實驗，驗證本實驗提出的方法在不同生產績效指標皆具有優越表現，長期而言對於生產績效提升具有顯著的成效。

On the basis of earlier study results have found that if the constrtlction of the machinelearning based RTS system knowledge base (KB) results in a scheduling KB with bettergeneralization ability (i.e., classification accuracy rate); the KB also leads to superiorproduction performance with respect to the RTS system KBs with bad generalization ability.Besides, there are two issues that significantly affect the generalization ability of machinelearning-based classifiers: the issue of feature subset selection (including parameteroptimization) and ensemble of the classifiers.
The fealure subset selection issue results fi'om the Ihct that a data set usually containshundreds of features, many of which are irrelevant and heavily correlated with others.Without feature selection, they tend to deteriorate perlbrmance of the model, as well asincrease the model training time. Feature subset selection can be t~3rmulated as anoptimization problem which involves searching the space of possible features to identify asubset that is optimum or near-optimal with respect to classification accuracy rate. Ensemblecombines the predictions of individual classifiers with the equal weight or weights based onestimated prediction accuracy; earlier study indicated that ensemble models have demonstrated consistent in some cases, remarkable improvements in prediction accuracy overindividual classifiers.
This researchwill develop GA-based wrapper feature selection andparameter optimization with three different machine learning algorithms (BPNN, DT, and SVM) as a base learner and then uses majority voting scheme that ensemble threeclassification algorithms denoted as GA+Voting classifiers.
The performance of this thesis proposed ensemble of the classifiers based on GA wrapper approach will be compared using datasets in FMS model (i.e., RTS system on line simulation experiment verification) to those of other machine learning-based classifiers (GA+BPNN, GA+DT, and GA+SVM) to demonstrate the proposed approaches superiority.

摘要 ............................................................ Ⅱ
ABSTRACT ....................................................... Ⅲ
目錄 ............................................................. Ⅳ
表錄 ............................................................. Ⅵ
圖錄 ........................................................... Ⅷ
一、緒論 .......................................................... 1
1.1 研究背景..................................................... 1
1.2 研究目的 .................................................... 3
1.3 研究流程 .................................................... 4
二、文獻探討 ...................................................... 5
2.1即時排程法則選擇機制　........................................ 5
2.2彈性製造系統 ................................................. 7
2.3 k交叉驗證 ................................................... 9
2.4屬性篩選 .................................................... 10
2.4.1 Wrapper Model ............................................. 11
2.4.2 Filter Model ............................................... 11
2.5基因演算法 .................................................. 12
2.6決策樹 ...................................................... 17
2.7 類神經網路 ................................................. 21
2.8支援向量機 .................................................. 23
2.8.1線性支援向量機 ............................................ 24
2.8.2非線性支援向量機 .......................................... 26
2.9集成 ........................................................ 27
2.9.1 多數決投票 ............................................... 28
三、研究方法 ..................................................... 29
3.1 模擬為基礎的訓練樣本產生機制 ............................... 30
3.2 資料正規化 ................................................. 35
3.3整合式GA+BPNN、GA+DT、GA+SVM ........................... 35
3.3.1 GA+BPNN染色體表示 ...................................... 36
3.3.2 GA+DT染色體表示 ......................................... 39
3.3.3 GA+SVM染色體表示 ....................................... 40
3.3.4最初母體和適合度函數的選擇 ................................ 42
3.3.5基因運算 .................................................. 43
3.3.6產生一個新的母體 .......................................... 43
3.3.7終止條件 .................................................. 44
3.4集成投票 .................................................... 44
四、實驗與結果分析 ............................................... 46
4.1 模擬實驗模式之建構與訓練樣本之產生 ......................... 46
4.2 三種不同的機器學習演算法屬性篩選結果與參數 ................. 48
4.3 分類正確率之比較 ........................................... 58
4.4線上實驗結果................................................. 61
五、結論與未來研究方向 ............................................ 63
5.1結論 . .............................................63
5.2未來研究方向 ...................................... 64
參考文獻 ......................................................... 65

[1] 王派洲，資料探勘-概念與方法，第二版，滄海書局，台中市，2008
[2] 林育生，「晶圓廠智慧型排程系統之發展」，臺灣大學 工業工程學研究所，2003
[3] 莊明傑，「利用RFID與支持向量機的生產排程於SFIS系統」，東海大學資訊工程與科學研究所，2008
[4] 薛友仁，「整合機器學習方法於決策樹為基智慧型排程系統之研究」，國立交通大學工業工程與管理研究所博士論文，2000
[5] Shiue, Y. R., and Guh, R. S., “Learning-based multi-pass adaptive scheduling for a dynamic manufacturing cell environment.”, Robotics and Computer-Integrated Manufacturing, 22, 2006, pp.203-216.
[6] Shiue, Y. R., ” Data-mining-based dynamic dispatching rule selection mechanism for shop floor control systems using a support vector machine approach”, International Journal of Production Research, Vol. 47, 3669–3690 ,2009,
[7] Shiue, Y. R., Guh, R. S.,and Tseng, T. Y” GA-based learning bias selection mechanism for real-time scheduling systems”, Expert Systems with Applications, 36, 11451–11460, 2009
[8] Quinlan, J. R., C4.5: Programs for Machine Learning （San Manteo: Morgan KaufmBPNN）, 1993
[9] Quinlan, J. R., “Induction of decision trees,” Machine Learning, 1, 81-106, 1986.
[10] Quinlan, J. R., “Simplifying Decision Trees,” International Journal of Man-Machine Studies, 27, 221-234, 1987.
[11] CM Kuan , H White , " Artificial Neural Networks: An Econometric Perspective", Econometric Reviews, 1994
[12] Kohavi, R., and John, G. H., “Wrappers for Feature Subset Selection,” Artificial Intelligence, , pp. 273-324, 1997
[13] Liu, H. and Motoda, H., “Feature Selection for Knowledge Discovery and Data Mining”, Kluwer Academic Publishers, Boston, 1998.
[14] Werbos P. J. “Beyond Regression: New Tools for Prediction and Analysis in the Behavioral Sciences”. Masters thesis, Harvard University.1974
[15] DE Rumelhart, GE Hinton, RJ Williams, “Learning internal representations by error propagation”, Parallel distributed processing, 1986
[16] John Holland, “Adaptation in Natural an Artificial System,” University of Michigan Press, 1975
[17] N. J. Nilsson, “Learning machines: Foundations of trainable pattern classifying systems”. McGraw-Hill, 1965
[18] DW Opitz, JW Shavlik,”Advances in neural information processing systems”, citeseerx.ist.psu.edu ,1996-
[19] David E. Goldberg, “Genetic Algorithms in Search”, Optimization and Machine Learning, 1st edition, 1989
[20] Priore, P., De La Fuente, D., Gomez, A. and Puente, J., “A review of machine learning in dynamic scheduling of flexible manufacturing systems.”, Artificial Intelligence for Engineering Design, Analysis & Manufacturing, 15, 2001, pp.251-263.
[21] Rumelhart, D. E., Hinton, G. E., and Williams, R. J., “Learning internal representations by error propagation.”, Parallel Distributed Processing （MIT press: Cambridge） , 1986.
[22] Cho, H., and Wysk, R. A., “A robust adaptive scheduler for an intelligent workstation controller.”, International Journal of Production Research, 31, 1993, pp.771-789.
[23] Arzi, Y., and Iaroslavitz, L., “Neural network-based adaptive production control system for flexible manufacturing cell under a random environment”, IIE Transactions, 31, 1999, pp.217-230.
[24] Sabuncuoglu, I., “A study of scheduling rules of flexible manufacturing systems: a simulation approach”, International Journal of Production Research, 36, 527-546, 1998.
[25] Park, S. C., Raman, N., and Shaw, M. J., “Adaptive scheduling in dynamic flexible manufacturing systems: a dynamic rule selection approach.”, IEEE Transactions on Robotics and Automation, 13, 1997, pp.486-502.
[26] Su, C. T., and Shiue, Y. R., “Inteligent scheduling controller for shop floor control systems: a hybrid genetic algorithm/decision tree learning approach.”, International Journal of Production Research, 41, 2003, pp.2619-2641.
[27] Arzi, Y., and Iaroslavitz, L., “Operating an FMC by a decision-tree-based adaptive production control system.”, International Journal of Production Research, 38, 2000, pp.675-697.
[28] Kim, J., H. R. Weistroffer, and R.T .Redmond, “Expert System for Bond Rating：A Comparative Analysis of Statistical, Rule-Based and Neural Network Systems,” Expert Systems Vol.10, pp.167-172,1993
[29] Beaver, William H., “Financial ratios as predictors of failure,” Journal of Accounting Research, Vol.4, pp.71-111, 1966.


[30] Lee, C. Y., Piramuthu, S., and Tsai, Y. K., “Job shop scheduling with a genetic algorithm and machine learning.”, International Journal of Production Research, 35, 1997, pp.1171-1191.
[31] Kim, C. O., Min, H. S., and Yih, Y., “Integration of inductive learning and neural networks for multi-objective FMS scheduling.”, International Journal of Production Research, 36, 1998, pp.2497-2509.
[32]eM-Plant, 2003, Objects Manual Version and Reference Manual 7.0 （Tecnomatix Technologies: Stuttgart）.
[33] L. Breiman, J.H. Friedman, R.A.Olsen, and C.J. Stone.”Classification and RegressionTrees” , Wadsworth Statistic/Probability Series, 1984.
[34] Montazeri, M. and Van Wassenhove, L. N., “Analysis of scheduling rules foran FMS,” International Journal Production Research, 28, 785-802, 1990.
[35] Blackstone, J. H., Philips, D. T. Jr., and Hogg, G. L., “A state-of-the-art survey of dispatching rules for manufacturing job shop operations”, International Journal of Production Research, 20, 27-45, 1982.
[36] Maher, J. J. and T. K. Sen , “Predicting Bond Ratings using Neural Networks：A Comparison with Logistic Regression,”Intelligent System in Accounting、Finance and Management ,Vol.6, pp.59-72,1997.
[37] Kim, J., H. R. Weistroffer, and R.T .Redmond,“Expert System for Bond Rating：A Comparative Analysis of Statistical, Rule-Based and Neural Network Systems,” Expert Systems Vol.10, pp.167-172,1993.
[38] Cho, H., and Wysk, R. A., “A robust adaptive scheduler for an intelligent workstation controller,” International Journal of Production Research, 31, 771-789, 1993.
[39] Arzi, Y., and Iaroslavitz, L., “Operating an FMC by a decision-tree-based adaptive production control system,” International Journal of Production Research , 38, 675-697, 2000.
[40] Park, S. C., Raman, N., and Shaw, M. J., “Adaptive scheduling in dynamic flexible manufacturing systems: a dynamic rule selection approach,” IEEE Transactions on Robotics and Automation, Vol. 13, pp 486-502, 1997.
[41] Esposito, F.& Malerba, D.& Semeraro, G.,” A Further Study of Pruning Methods in Decision Tree Induction,” Proceedings of the Fifth International Workshop on Artificial Intelligence and Statistics, pp.211-218, 1995.
[42] Wu, S. D., and Wysk, R. A., “An application of discrete-event simulation to on-line control and scheduling in flexible manufacturing”, International Journal of Production Research, 27, 1603-1623, 1989.
[43] Wein, L. M., “Scheduling semiconductor wafer fabrication.”, IEEE Transactions on semiconductor manufacturing, 1, 1988, pp.115-130.
[44] Kumar, P. R., “Scheduling semiconductor manufacturing plants.”, IEEE Control Systems, December, 1994, pp.33–40.
[45] Li, S., Tang, T. and Collins, D. W., “Minimum inventory variability schedule with application in semiconductor fabrication.” IEEE Transactions on Semiconductor Manufacturing, 9, 1996, pp.145–149.
[46] Baker, K. R., “Sequencing rules and due-date assignments in a job shop,” Management Science, 30, 1093-1104, 1984.
[47] Blackstone, J. H., Philips, D. T. Jr., and Hogg, G. L., “A state-of-the-art survey of dispatching rules for manufacturing job shop operations”, International Journal of Production Research, 20, 27-45, 1982.
[48] Montazeri, M. and Van Wassenhove, L. N., “Analysis of scheduling rules foran FMS,” International Journal Production Research, 28, 785-802, 1990.
[49] Salzberg, S. L., “On Comparing Classifiers: Pitfalls to Avoid and A Recommended Approach,” Data Mining and Knowledge Discovery, Vol. 1, No. 3, pp. 317-328, 1997.