本研究之目的為改善虛擬生產控制系統之在製品預測模組應用於半導體供應鏈中，透過收及在製品資訊進行產出推估。資料以snapshot傳送方式至系統中，由於生產週期時間資料傳送週期造成之部分資料遺失，針對無法對應其投料資訊之產出，稱為非成對資料，反之，稱為成對資料。針對這兩種資料型態所設計之不同預測架構，透過灰預測以及期望值預測方法，推估在製品產出時間。

The objective of this work is to enhance the prediction module of virtual production control system, which is used to forecast WIP outputs in a semiconductor supply chain. Due to work in process (WIP) information collected from suppliers by snapshot, the WIP data can be categorized by the corresponding move-in information, where the unpaired data fails to find the move-in time, but the paired data does. To process the different WIP data, two methods, i.e., Grey prediction model (GPM) and expected value (EV), are proposed in this prediction module to forecast WIP outputs which are evaluated the accuracy and reliance of forecasts by adopting an overlap area index.
After applied the enhanced module to forecast 3 days and 5 days outputs, an unpaired data case study shows that the average accuracies of the top 10 sizes of products by GPM are 55% and 81%, respectively. Meanwhile, a paired-data case illustrates that the accuracies of all products of by GPM are 85% and 86%, respectively. Moreover, the forecast reliance can be evaluated by using GPM and EV in this enhanced module.

[1] H.-C. Yang, Y.-L. Chen, M.-H. Hung and F.-T. Cheng, “Virtual Production Control System,” 6th annual IEEE Conference on Automation Science and Engineering, pp. 984–989, 2010.
[2] H.-C. Yang and T.-H. Tsai, “A Pareto-Optimal Fitting Scheme to Identify Cascaded Distributions,” The 10th International Conference on Automation Technology, pp. 55–57, 2009.
[3] D.-C. Li, C.-W. Yeh, and C.-J. Chang, “An improved grey-based approach for early manufacturing data forecasting,” Computers & Industrial Engineering, pp. 1161–1167, 2009.
[4] A. I. Dounis, P. Tiropanis, D. Tseles, G. Nikolaou, and G. P. Syrcos, “A Comparison of Grey Model and Fuzzy Predictive Model for Time Series,” International Journal of Engineering and Mathematical Sciences, pp. 176–181, 2006.
[5] E. Kayacan, B. Ulutas, and O. Kaynak, “Grey system theory-based models in time series prediction,” Expert Systems with Applications, pp. 1784–1789, 2010.
[6] F.-T. Cheng, Y.-T. Chen, Y.-C. Su and D.-L. Zeng, “Evaluating Reliance Level of a Virtual Metrology System,” IEEE Transactions on semiconductor, vol. 21, no. 1, pp. 92–103, 2008.
[7] T. Chen, “Predicting Wafer-Lot Output Time With a Hybrid FCM–FBPN Approach,” IEEE Transactions on systems, man, and cybernetics, vol. 37, no. 4, pp. 784–793, 2007.
[8] P.-C. Chang and C.-Y. Lai, “A hybrid system combining self-organizing maps with case-based reasoning in wholesaler’s new-release book forecasting,” Expert Systems with Applications, pp. 183–192, 2005.