過去採用外部資訊的研究顯示社群媒體資訊的應用對 於Business-to-customer (B2C) 企業的需求預測可有效降低其預測誤差。然而對於Business-to-business (B2B) 企業來說，銷售對象並非市場上使用社群媒體的終端消費者，因此社群媒體資訊在其銷售需求趨勢表現上相對不顯著。不僅如此，過去對於需求預測的研究中，大多將外部資訊指標結合歷史銷售數據做為新的訓練資料集，有鑒於此，本篇研究提出了一種智慧需求預測方法包含如何選取與利用外部資訊指標，並應用於B2B企業中的一間美國伺服器產業，以其真實銷售資料為需求預測之研究對象。本篇研究採用網路爬蟲技術與Google趨勢搜尋蒐集相關市場指標作為本篇伺服器產業之外部資訊指標，採用分群分類分析與迴歸分析處理外部資訊指標，所建立的模型對於突發性需求之高峰與低谷值能適時反映其結果。研究結果中顯示以隨機森林演算法預測之樣本外測試的均方誤差為19.772 (平均平方根百分比誤差為62.11%)，而透過本篇所提出的方法，可進一步降低預測均方誤差至11.87 (平均平方根百分比誤差降低至16.77%)。本研究提出的智慧預測方法相對於該公司的預測在成果上能顯著降低預測均方誤差並提升預測精確度超過63% (平均平方根百分比誤差降低61.25%)。因此，本篇研究揭示了外部資訊指標在B2B企業需求預測上的價值貢獻並提出一套方法有效利用外部資訊指標。

A fundamental aspect of inventory management is accurate demand forecasting. Forecasting is both challenging and difficult task when the demand suddenly increases or decreases. In terms of business-to-customer (B2C), prior researches have been adopting social media information to improve demand forecast accuracy. However, due to a lack of end-consumer evaluations, social media information is not significantly performed in business-to-business (B2B).
In summary, this research focus on interesting issues which is to capture and utilize external information to improve the B2B demand forecast. For this purpose, this research develops and utilizes an intelligent demand forecasting approach to a server of B2B company which is in the United States. Furthermore, we implement time series and machine learning methods and choose the best approach as a baseline model and use web crawler and Google Trends search to collect related market signals and google trends as the external information index. Finally, the external information indexes are incorporated into a machine learning method (i.e., classification and clustering) to adjust the demand peak or valley, which uses both internal (historical) data and external information. The results of this research demonstrate whether using the external information index or not yields different out‐of‐sample MSE with 11.87 and 19.77 by the random forest. The accuracy of the internal forecast is significantly improved more than 63.10% of MSE (44.1% of MAE and 61.26% of RMSPE) adopting intelligent forecasting in this research. In this research, the external information is considered as an efficient way to increase the accuracy of the machine learning model and sheds light on the rule of demand forecasting using external information index for the B2B industry.

Álvarez-López, V., Campomanes-Álvarez, B. R., & Quirós, P. (2019). Hybrid machine learning methods for demand forecasting: consecutive application of classification and regression methods for the forecasting of periodic and non-continuous demand. Paper presented at the Proceedings of the 2nd International Conference on Applications of Intelligent Systems.
Bagchi, S., Buckley, S. J., Ettl, M., & Lin, G. Y. (1998). Experience using the IBM supply chain simulator. Paper presented at the 1998 Winter Simulation Conference. Proceedings (Cat. No. 98CH36274).
Boone, T., Ganeshan, R., Hicks, R. L., & Sanders, N. R. (2018). Can Google Trends Improve Your Sales Forecast? Production and Operations Management, 27(10), 1770-1774.
Box, G. E., & Jenkins, G. M. (1976). Time series analysis: Forecasting and control San Francisco. Calif: Holden-Day.
Breiman, L. (2001). Random forests. Machine learning, 45(1), 5-32.
Chen, T., & Guestrin, C. (2016). Xgboost: A scalable tree boosting system. Paper presented at the Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining.
Cui, R., Gallino, S., Moreno, A., & Zhang, D. J. (2018). The operational value of social media information. Production and Operations Management, 27(10), 1749-1769.
Dai, W., Chuang, Y.-Y., & Lu, C.-J. (2015). A clustering-based sales forecasting scheme using support vector regression for computer server. Procedia Manufacturing, 2, 82-86.
Dimpfl, T., & Jank, S. (2016). Can internet search queries help to predict stock market volatility? European Financial Management, 22(2), 171-192.
Fu, W., Chien, C.-F., & Lin, Z.-H. (2018). A hybrid forecasting framework with neural network and time-series method for intermittent demand in semiconductor supply chain. Paper presented at the IFIP International Conference on Advances in Production Management Systems.
Goia, A., May, C., & Fusai, G. (2010). Functional clustering and linear regression for peak load forecasting. International Journal of Forecasting, 26(4), 700-711.
Goodwin, M., & Yazidi, A. (2014). A pattern recognition approach for peak prediction of electrical consumption. Paper presented at the IFIP International Conference on Artificial Intelligence Applications and Innovations.
Hsu, C.-Y., Lin, S.-C., & Chien, C.-F. (2015). A back-propagation neural network with a distributed lag model for semiconductor vendor-managed inventory. Journal of Industrial and Production Engineering, 32(3), 149-161.
Hyndman, R. J., & Athanasopoulos, G. (2017). Forecasting: principles and practice, 2013. URL: https://www. otexts. org/fpp [accessed 2018-02-15][WebCite Cache ID 6xFJlXCQI].
Mitchell, T. M. (1997). Does machine learning really work? AI magazine, 18(3), 11-11.
Norton, J. A., & Bass, F. M. (1987). A diffusion theory model of adoption and substitution for successive generations of high-technology products. Management science, 33(9), 1069-1086.
Saha, C., Lam, S. S., & Boldrin, W. (2014). Demand forecasting for server manufacturing using neural networks. Paper presented at the IIE Annual Conference. Proceedings.
Tashman, L. J. (2000). Out-of-sample tests of forecasting accuracy: an analysis and review. International journal of forecasting, 16(4), 437-450.
Taylor, R. (1990). Interpretation of the correlation coefficient: a basic review. Journal of Diagnostic Medical Sonography, 6(1), 35-39.
Wagstaff, K., Cardie, C., Rogers, S., & Schrödl, S. (2001). Constrained k-means clustering with background knowledge. Paper presented at the Icml.
Wei, Z., & Shan, X. (2019). Research on forecast method of railway passenger flow demand in pre-sale period. Paper presented at the IOP Conference Series: Materials Science and Engineering.
Whelan, C., Harrell, G., & Wang, J. (2015). Understanding the K-Medians Problem. Paper presented at the Proceedings of the International Conference on Scientific Computing (CSC).
Winters, P. R. (1960). Forecasting sales by exponentially weighted moving averages. Management Science, 6(3), 324-342.
Wu, S. D., Erkoc, M., & Karabuk, S. (2005). Managing capacity in the high-tech industry: A review of literature. The engineering economist, 50(2), 125-158.