機器學習是人工智能的一個分支，它讓電腦能夠在無需編寫大量程序的情況下進行學習，即電腦可以累積「經驗」並自動搜索信息以捕捉規律和趨勢。在現代製造業中，確保產品品質的穩定性，對於企業的競爭力至關重要，而機器學習在製造業品質預測上的應用更為企業提供了更可靠和準確地品質控制手段，在本研究中，我們將機器學習方法應用於預測樹脂砂生產過程的品質，數據來自台灣的一家樹脂砂生產廠。利用生產設備的過程參數來預測未來生產的品質。選擇矽砂類型、樹脂比例、混合溫度和批量加熱作為輸入，以預測輸出（顆粒大小、抗彎強度、均勻點和落試驗）並使用最佳結果優化配方。本研究採用了多種機器學習方法來預測品質。此外，還考慮了不同特徵選擇方法的優缺點。結果顯示，根據不同的特點，其最佳的模型結果不太相同，若是以可解釋性來說，基於K-Means Clustering 的Recurrent Neural Network 具有更優越的模型解釋性。但若以準確度來比較，Filter Support Vector Regression 有較精準的準確度，通過比較經驗配方和人工智能推薦配方，我們還發現AI 推薦的配方實現了更高且更穩定的品質。

Machine learning, a subfield of artificial intelligence, enables computers to learn
without the need for extensive programming, allowing them to accumulate
"experience" and automatically search for information to capture patterns and trends.
In the realm of modern manufacturing, the assurance of product quality stability is
crucial to a company's competitiveness. The application of machine learning in
predicting manufacturing quality equips enterprises with more reliable and precise
means of quality control. In this study, we apply machine learning methods to predict
the quality of the resin sand production process, with data collected from a resin sand
manufacturer in Taiwan. Process parameters of the production equipment are utilized
to predict the quality of future production. The type of silica sand, resin ratio, mixing temperature, and batch heating are selected as inputs to predict the outputs (particle size, flexural strength, uniformity point, and drop test) and optimize the formulation using the best results. This study employs various machine learning methods to predict quality, including Support Vector Regression, Random Forest, eXtreme Gradient Boosting, Recurrent Neural Network, Long Short-Term Memory, and Gated Recurrent Unit.
Additionally, different feature selection methods' advantages and disadvantages are
considered. The results show that the optimal model outcome varies depending on the
specific features; if interpretability is the main criterion, the K-Means Clustering based RNN demonstrates superior model interpretability. However, if accuracy is the main concern, Filter Support Vector Regression exhibits higher precision. By comparing
experience-based formulations and artificial intelligence-recommended formulations,
we also find that AI-recommended formulations achieve higher and more stable quality.

Bai, Y., Sun, Z., Deng, J., Li, L., Long, J., & Li, C. (2017). Manufacturing quality prediction
using intelligent learning approaches: A comparative study. Sustainability,
10(1), 85.
Breiman, L. (2001). Random forests. Machine learning, 45, 5-32.
Cha, Y. J., Choi, W., & Büyüköztürk, O. (2017). Deep learning-based crack damage
detection using convolutional neural networks. Computer-Aided Civil and
Infrastructure Engineering, 32(5), 361-378.
Chandrashekar, G., & Sahin, F. (2014). A survey on feature selection methods.
Computers & Electrical Engineering, 40(1), 16-28.
Chate, G. R., Patel, G. M., Kulkarni, R. M., Vernekar, P., Deshpande, A. S., &
Parappagoudar, M. B. (2018). Study of the effect of nano-silica particles on
resin-bonded moulding sand properties and quality of casting. Silicon, 10,
1921-1936.
Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. Proceedings
of the 22nd acm sigkdd international conference on knowledge discovery and
data mining,
Chen, X.-w., & Jeong, J. C. (2007). Enhanced recursive feature elimination. Sixth
international conference on machine learning and applications (ICMLA 2007),
Chumachenko, D., Turiy, A., & Chukhray, A. (2019). Application of statistical simulation
for measles epidemic process forecasting. 2019 IEEE 2nd Ukraine Conference
on Electrical and Computer Engineering (UKRCON),
Chung, J., Gulcehre, C., Cho, K., & Bengio, Y. (2014). Empirical evaluation of gated
recurrent neural networks on sequence modeling. arXiv preprint
arXiv:1412.3555.
El Naqa, I., & Murphy, M. J. (2015). What is machine learning? Springer.
Fu, R., Zhang, Z., & Li, L. (2016). Using LSTM and GRU neural network methods for
traffic flow prediction. 2016 31st Youth Academic Annual Conference of
Chinese Association of Automation (YAC),
Guo, R., Zhao, Z., Wang, T., Liu, G., Zhao, J., & Gao, D. (2020). Degradation state
recognition of piston pump based on ICEEMDAN and XGBoost. Applied Sciences,
10(18), 6593.
Haghiabi, A. H., Nasrolahi, A. H., & Parsaie, A. (2018). Water quality prediction using
machine learning methods. Water Quality Research Journal, 53(1), 3-13.
Hota, H., Handa, R., & Shrivas, A. K. (2017). Time series data prediction using sliding
window based RBF neural network. International Journal of Computational
Intelligence Research, 13(5), 1145-1156.
Li, X., Peng, L., Hu, Y., Shao, J., & Chi, T. (2016). Deep learning architecture for air quality
predictions. Environmental Science and Pollution Research, 23, 22408-22417.
Lieber, D., Stolpe, M., Konrad, B., Deuse, J., & Morik, K. (2013). Quality prediction in interlinked manufacturing processes based on supervised & unsupervised
machine learning. Procedia Cirp, 7, 193-198.
Mahalingam, U., Elangovan, K., Dobhal, H., Valliappa, C., Shrestha, S., & Kedam, G.
(2019). A machine learning model for air quality prediction for smart cities.
2019 International conference on wireless communications signal processing
and networking (WiSPNET),
McCulloch, W. S., & Pitts, W. (1943). A logical calculus of the ideas immanent in
nervous activity. The bulletin of mathematical biophysics, 5, 115-133.
Nada, O. A., ElMaraghy, H. A., & ElMaraghy, W. H. (2006). Quality prediction in
manufacturing system design. Journal of Manufacturing systems, 25(3), 153-
171.
Olah, C. (2015). Understanding lstm networks.
Rosenblatt, F. (1958). The perceptron: a probabilistic model for information storage
and organization in the brain. Psychological review, 65(6), 386.
Shewalkar, A., Nyavanandi, D., & Ludwig, S. A. (2019). Performance evaluation of deep
neural networks applied to speech recognition: RNN, LSTM and GRU. Journal
of Artificial Intelligence and Soft Computing Research, 9(4), 235-245.
Unler, A., Murat, A., & Chinnam, R. B. (2011). mr2PSO: A maximum relevance minimum
redundancy feature selection method based on swarm intelligence for support
vector machine classification. Information Sciences, 181(20), 4625-4641.
Wang, C., Huang, Y., Shao, M., Hu, Q., & Chen, D. (2019). Feature selection based on
neighborhood self-information. IEEE Transactions on Cybernetics, 50(9), 4031-
4042.
Wang, K., Gopaluni, R. B., Chen, J., & Song, Z. (2018). Deep learning of complex batch
process data and its application on quality prediction. IEEE Transactions on
Industrial Informatics, 16(12), 7233-7242.
Werbos, P. J. (1994). The roots of backpropagation: from ordered derivatives to neural
networks and political forecasting (Vol. 1). John Wiley & Sons.
Yahmed, Y. B., Bakar, A. A., RazakHamdan, A., Ahmed, A., & Syed Abdullah, S. M. (2015).
ADAPTIVE SLIDING WINDOW ALGORITHM FOR WEATHER DATA
SEGMENTATION. Journal of Theoretical & Applied Information Technology,
80(2).
Yamak, P. T., Yujian, L., & Gadosey, P. K. (2019). A comparison between arima, lstm, and gru for time series forecasting. Proceedings of the 2019 2nd International
Conference on Algorithms, Computing and Artificial Intelligence,