科技發展日新月異，伴隨著技術突破，獲取製程的詳細資料已不再是難事；與此同時，低廉的記憶體價格讓工廠能輕鬆儲存大量資料。然而，未經分析的資料並不能帶來效益，如何妥善處理及運用這些資料是企業的重要課題。
在工廠中，製程優化是十分重要的任務。由於顧客不只會要求產品品質，同時也期待訂單交期縮短；對於現場人員而言，使產線穩定地輸出產品是他們的目標。
近年來，在資料科學領域蓬勃發展的情況下，許多工廠在進行製程改善方案時，轉為使用資料科學相關技術，而非傳統統計方法。本研究使用類神經網路建立模型，作為目標函數以探討製程控制因子與輸出反應值之間的關係，並在後續使用啟發式演算法找出最佳製程參數配置。
本研究以台灣封裝測試大廠之點膠製程為例，使用製程歷史資料建立類神經網路模型，後續使用三種啟發式演算法執行製程參數最佳化，並執行確認實驗以驗證最佳化結果。研究結果顯示本研究提出的方法能有效提升製程品質，證實了方法之有效性。

With the rapid development of technology, companies can easily get details regarding manufacturing process. Also, they are able to store large amount of data because of the low price of memory. However, the unprocessed data is worthless; it is important to process and use the data properly.
For factories, process optimization is one of the important issues. Their customers not only request high-quality products, but also expect shorten delivery time. That is the reason why making the manufacturing stable is the goal of enterprise.
In recent years, the field of data science is booming. Instead of using traditional statistical methods, many factories have used methods related to data science to conduct projects about manufacturing process optimization.
This study proposes a method consist of neural network and heuristic algorithm. Neural network is used to explore the relationship of process input and output, and the model becomes an objective function for heuristic algorithm. Heuristic algorithm is used to optimize the process factors.
The proposed method was applied to the dispensing process of the semiconductor fabrication plant in Taiwan. A neural network model was built through the historical data from the dispensing process, and three heuristic algorithms was adopted for the process factors optimization. The confirmation test was conducted for verification. The result of this study shows that the proposed method is able to enhance the dispensing process and also proves the validity of the method.

Alvarado Iniesta, A., García Alcaraz, J. L., & Rodríguez Borbón, M. (2013). Optimization of injection molding process parameters by a hybrid of artificial neural network and artificial bee colony algorithm. Revista Facultad de Ingeniería Universidad de Antioquia(67), 43-51.
Chakraborty, S., & Diyaley, S. (2018). Multi-objective optimization of yarn characteristics using evolutionary algorithms: a comparative study. Journal of The Institution of Engineers (India): Series E, 99(2), 129-140.
Fausett, L. V. (1994). Fundamentals of Neural Networks : Architectures, Algorithms, and Applications. Englewood Cliffs, NJ: Prentice-Hall.
Holland, J. H. (1992). Adaptation in Natural and Artificial Systems : An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence (1st MIT Press ed. ed.). Cambridge, Mass: MIT Press.
Javed, M. S., Ma, T., Jurasz, J., Ahmed, S., & Mikulik, J. (2020). Performance comparison of heuristic algorithms for optimization of hybrid off-grid renewable energy systems. Energy, 210, 118599.
Karaboga, D. (2005). An idea based on honey bee swarm for numerical optimization (Vol. 300, pp. 1-10): Technical report-tr06, Erciyes university, engineering faculty, computer engineering department.
Karaboga, D., & Basturk, B. (2008). On the performance of artificial bee colony (ABC) algorithm. Applied Soft Computing, 8(1), 687-697.

Kennedy, J., & Eberhart, R. (1995). Particle swarm optimization. Paper presented at the Proceedings of ICNN'95-international Conference on Neural Networks.
Kulkarni, V. R., & Desai, V. (2016). ABC and PSO: A comparative analysis. Paper presented at the 2016 IEEE International Conference on Computational Intelligence and Computing Research (ICCIC)
Liao, Y.-X., She, J.-H., & Wu, M. (2009). Integrated hybrid-PSO and fuzzy-NN decoupling control for temperature of reheating furnace. IEEE Transactions on Industrial Electronics, 56(7), 2704-2714.
Muthiah, A., & Rajkumar, R. (2014). A comparison of artificial bee colony algorithm and genetic algorithm to minimize the makespan for job shop scheduling. Procedia Engineering, 97, 1745-1754.
Su, C.-T., Lin, C.-M., & Chang, C. A. (2012). Optimization of the bistability property for flexible display by an integrated approach using Taguchi methods, neural networks and genetic algorithms. Microelectronics Reliability, 52(7), 1492-1500.
Su, C.-T., Hsiao, Y.-H., & Chang, C.-C. (2011). Parameter optimization design for touch panel laser cutting process. IEEE Transactions on Automation Science and Engineering, 9(2), 320-329.
Vagheesan, S., & Govindarajalu, J. (2019). Hybrid neural network–particle swarm optimization algorithm and neural network–genetic algorithm for the optimization of quality characteristics during CO2 laser cutting of aluminium alloy. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 41(8), 1-15.
Whitley, D. (1994). A genetic algorithm tutorial. Statistics and computing, 4(2), 65-85.
Zhang, H.-C., & Huang, S. (1995). Applications of neural networks in manufacturing: a state-of-the-art survey. International Journal of Production Research, 33(3), 705-728.
徐瑞昇 (2022)。應用田口法改善封裝點膠製程穩定性之研究-以A封裝廠為例。(碩士)，國立清華大學，新竹市。
蘇朝墩（2013）。品質工程 : 線外方法與應用：前程文化。