在這個大數據的時代，我們再也不需要只靠自己的經驗來決定機台或生產線的設定。我們可以在有興趣參數的相對應位置上設置感測器，並記錄其數值，再藉由資料分析的方法，把這些數值轉化成對我們有用的資訊，從而達到改善機台效率或生產線效率的目標。
空壓機在半導體的製造產業中扮演了舉足輕重的角色，它可以把機械能轉換成氣體壓力能。確切來說，空壓機用來壓縮空氣藉以提高氣體壓力並為生產線提供能源。然而，空壓機是相當耗能的，它的性能表現會被其參數甚至是環境因素所影響。正所謂牽一髮動全身，就算是參數設置上的些微改變，也會對空壓機的產出造成影響。這篇論文以台灣的一家半導體公司為例，進行產學合作。在這一例子中，他們擁有三種不同類型的空壓機，而每種空壓機皆有不同類型或不同數量的參數，且皆會影響其性能表現與產出。此外，空壓機也有兩種狀態：低載與滿載。
本研究首先利用逐步迴歸來建立每台空壓機的產出與其參數之間的關係式，之後，建立一個全系統的模型，以個案公司為例，其總共包含了九台的空壓機。藉由模型的目標式與限制式，能找到滿足公司需求且最節能的機台配置，本研究也發展了結合機台表現的動態規劃來更有效率的解決問題。此篇論文可以幫助半導體公司節省能源並降低成本。

Having fully entered the big data era, it is no longer necessary to determine the settings of machines or production lines purely by our experience. Through the recording values of related parameters from sensors, we are able to use data analysis methods to turn these values into helpful information, thereby achieving the goal of improving the efficiency of machines or production lines.
Air compressors are indispensable parts of semiconductor manufacturing. Air compressors are devices which transform mechanical energy into gas pressure energy. In particular, they compress gas and play the crucial role of providing energy to production lines. However, air compressors are quite energy-consuming, and their performance is affected by their parameters and environmental factors. In addition, slight changes in the parameter settings lead to substantial changes in the output. This paper examines a Taiwan semiconductor company as a case study. In this case, there are three different types of air compressors. For each type of air compressor, there is a different kind or different number of parameters which affect the output and performance. Also, there are two statuses for each compressor: full load and low load. We first set the relationships between the output and the parameters of each compressor using stepwise regression. Afterwards, a whole-system model which includes nine air compressors is built. With the objective function and subjective functions, we can find the configuration that optimizes the output and satisfies the company’s needs while using minimum electricity. We also develop dynamic programming combining the performance of machines to solve this problem in an efficient way. This paper can help semi-conductor companies save energy and reduce costs.

1. Abdul-Wahab, S. A., Bakheit, C. S., & Al-Alawi, S. M. (2005). Principal component and multiple regression analysis in modelling of ground-level ozone and factors affecting its concentrations. Environmental Modelling & Software, 20(10), 1263-1271.
2. Al-Alawi, S. M., Abdul-Wahab, S. A., & Bakheit, C. S. (2008). Combining principal component regression and artificial neural networks for more accurate predictions of ground-level ozone. Environmental Modelling & Software, 23(4), 396-403.
3. Ciampi, A., Thiffault, J., Nakache, J. P., & Asselain, B. (1986). Stratification by stepwise regression, correspondence analysis and recursive partition: a comparison of three methods of analysis for survival data with covariates. Computational Statistics & Data Analysis, 4(3), 185-204.
4. Hermiston, R. T., & Faulkner, J. A. (1971). Prediction of maximal oxygen uptake by a stepwise regression technique. Journal of Applied Physiology, 30(6), 833-837.
5. Kaytez, F., Taplamacioglu, M. C., Cam, E., & Hardalac, F. (2015). Forecasting electricity consumption: A comparison of regression analysis, neural networks and least squares support vector machines. International Journal of Electrical Power & Energy Systems, 67, 431-438.
6. Ku, W., Storer, R. H., & Georgakis, C. (1995). Disturbance detection and isolation by dynamic principal component analysis. Chemometrics and Intelligent Laboratory Systems, 30(1), 179-196.
7. Li, K., Hu, C., Liu, G., & Xue, W. (2015). Building's electricity consumption prediction using optimized artificial neural networks and principal component analysis. Energy and Buildings, 108, 106-113.
8. Liao, X., Li, Q., Yang, X., Zhang, W., & Li, W. (2008). Multiobjective optimization for crash safety design of vehicles using stepwise regression model. Structural and Multidisciplinary Optimization, 35(6), 561-569.
9. Nomikos, P., & MacGregor, J. F. (1994). Monitoring batch processes using multiway principal component analysis. AIChE Journal, 40(8), 1361-1375.
10. Tso, G. K., & Yau, K. K. (2007). Predicting electricity energy consumption: A comparison of regression analysis, decision tree and neural networks. Energy, 32(9), 1761-1768.
11. Wong, J. K. L., Mason, A. J., Neve, M. J., & Sowerby, K. W. (2006). Base station placement in indoor wireless systems using binary integer programming. IEE Proceedings-Communications, 153(5), 771-778.
12. Zou, H., Hastie, T., & Tibshirani, R. (2006). Sparse principal component analysis. Journal of Computational and Graphical Statistics, 15(2), 265-286.