現在的產業多以數據導向做決策，透過數據分析和應用幫助企業提升營運效率成為很重要的議題，大數據分析與應用的呼聲高漲，過去大多會使用統計方法作為分析的工具，現今機器學習逐漸成熟與普及，有更多有效及快速的方法可以做出良好的預測。
然而實務上並非皆能蒐集到海量的資料，本研究以優化半導體熔噴製程參數為例，個案公司因製造成本的考量，無法在生產線中取得大量的製程參數，本研究所使用的資料筆數並不如以往建構類神經網路的多，但在選擇適合的網路結構和調整超參數之後，仍然找到了合適的類神經網路架構，再透過基因演算法成功協助個案公司的新產線找到最佳的製程參數組合。
本研究首度引入類神經網路結合基因演算法於優化熔噴製程的製程參數，經過和個案公司取得的統計實驗比較後，證明本研究應用類神經網路結合基因演算法的模式，確實取得更好的表現且具有相當的可行性與實用性。

With the rapid progress of data science, data and analytics are applied in various field to drive efficiency, accuracy and ultimately generate more profits. People used statistic method in most cases before, while there are amounts of more and more effective and useful methods could be chosen in recent years.
However, it is not always available to obtain amounts of data in practice. This study takes the melt-blown process in semiconductor industry as an example. For case company, the cost of products is too time-consuming and expensive to collect a great quantity of data. Thus, the amount of data used in this study is not as large as in the most cases for constructing neural networks. But after adjusting the number of hidden layers and the hyperparameters, the suitable model is built successfully. And the optimal process parameters are searched through genetic algorithm. The case company successfully found the best combination of process parameters.
This study is the first study to apply the integrated method of neural network and genetic algorithm on the melt-blown process. The result has better performance compared with the statistic method on addressing the process parameter design problems. It proves the proposed method has more considerable usability and practicality

摘要 II
Abstract III
List of Figures V
List of Tables VI
Chapter 1. Introduction 1
Chapter 2. Related Work 3
2.1 Neural Networks 3
2.2 Genetic Algorithms 5
2.3 Integration of Neural Networks and Genetic Algorithms 7
Chapter 3. Proposed Method 9
3.1 Proposed Procedure 9
3.2 Methods 10
Chapter 4. Case Study 19
4.1 The Case 19
4.2 Statistical Analysis 24
4.3 NN+GA 29
4.4 Final result 34
Chapter 5. Conclusion 35
5.1 Summary 35
5.2 Future research 36
References 37

1. Aggarwal, C. C. (2018). Neural Networks And Deep Learning.
2. Brynjolfsson, E., & McElheran, K. (2016). The Rapid Adoption of Data-Driven Decision-Making. American Economic Review, 106(5), 133-139.
3. Chen, T., Li, L., & Huang, X. (2005). Predicting the fibre diameter of melt blown nonwovens: comparison of physical, statistical and artificial neural network models. Modelling and Simulation in Materials Science and Engineering, 13(4), 575-584.
4. Chen, T., Wang, J., & Huang, X. B. (2006). Artificial neural network modeling for predicting melt-blowing processing. Journal of Applied Polymer Science, 101(6), 4275-4280.
5. Chicano, F., Sutton, A. M., Whitley, L. D., & Alba, E. (2015). Fitness Probability Distribution of Bit-Flip Mutation. Evolutionary Computation, 23(2), 217-248.
6. Diederik, & Ba, J. (2017). Adam: A Method for Stochastic Optimization. arXiv pre-print server.
7. Hagewood, J. (2014). Technologies for the manufacture of synthetic polymer fibers. In Advances in Filament Yarn Spinning of Textiles and Polymers (pp. 48-71).
8. Lin, H.-C., Su, C.-T., Wang, C.-C., Chang, B.-H., & Juang, R.-C. (2012). Parameter optimization of continuous sputtering process based on Taguchi methods, neural networks, desirability function, and genetic algorithms. Expert Systems with Applications, 39(17), 12918-12925.
9. Su, C.-T., Hsiao, Y.-H., & Chang, C.-C. (2012). Parameter Optimization Design for Touch Panel Laser Cutting Process. IEEE Transactions on Automation Science and Engineering, 9(2), 320-329.
10. Su, C. T., Yeh, C. J., Chou, C. J., & Chang, C. A. (2014). An Experimental Approach to Enhance Cu Wire Bonding Yield Through Parameter Optimization. Experimental Techniques, 38(3), 29-36.
11. Sun, Q., Zhang, D., Chen, B. Z., & Wadsworth, L. C. (1996). Application of neural networks to meltblown process control. Journal of Applied Polymer Science, 62(10), 1605-1611. 4628(19961205)62:10<1605::Aid-App12>3.0.Co;2-0
12. Sun, Y., & Wang, X. (2010). Optimization of air flow field of the melt blowing slot die via numerical simulation and genetic algorithm. Journal of Applied Polymer Science, 115(3), 1540-1545.
13. Sun, Y. F., Liu, B. W., Wang, X. H., & Zeng, Y. C. (2011). Air-flow field of the melt-blowing slot die via numerical simulation and multiobjective genetic algorithms. Journal of Applied Polymer Science, 122(6), 3520-3527.
14. Yılmaz KAYA, M. U., Ramazan TEKĐN. (2011). .