過去的研究裡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並提升精確度。然而對於Business-to-business (B2B) 企業來說，社群媒體資訊在其銷售需求趨勢表現上相對不顯著。因為B2B企業產品銷售對象並非市場上使用社群媒體的終端消費者，所以不能以此社群媒體資訊有效代表B2B銷售需求的外部資訊。有鑒於此，本篇研究提出了一種智慧需求預測方法結合外部資訊，並應用於一間美國伺服器產業的B2B企業，以其真實銷售資料為需求預測之研究對象。本篇研究利用網路爬蟲與Google搜尋趨勢蒐集相關市場指標作為伺服器產業之外部資訊指標。在結合外部資訊指標於機器學習方法(隨機森林演算法)並作為需求感知調整參數後，所建立的模型對於突發需求之高峰與低谷質能適時反映其預測結果。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23.81 (平均平方根百分比誤差為64.38%)，而透過同時運用內部歷史資料與外部資訊指標，可進一步降低預測均方誤差至8.29 (平均平方根百分比誤差降低至43.50%)。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74% (平均平方根百分比誤差降低34.5%)。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。

Previous research has used social media information to improve the accuracy of demand forecast for business-to-customer (B2C) industry. However, social media information cannot be used for business-to-business (B2B) industry due to no end consumers’ evaluations. This paper develops an intelligent demand forecasting approach and applies it to an U.S. server company, which is a B2B company. Our approach firstly uses the web crawler and Google Trends search to collect the related market signals as the external information index for server industry. Then we incorporate the external information index into a machine learning method (i.e., random forests, RF) to adjust the demand peak or valley, which uses both internal (historical) data and external information. The results show that the random forest yields an out‐of‐sample MSE of 23.81 (RMSPE of 64.38%) when not using external information index, and 8.29 (RMSPE of 43.50%) when using external information index. Our approach significantly improves the accuracy of the company’s internal forecast more than 74% of MSE (34.5% of RMSPE). This research sheds light on the values of using external information index in demand forecasting for B2B industry.

Albey, E., Norouzi, A., Kempf, K. G., & Uzsoy, R. (2015). Demand modeling with forecast evolution: an application to production planning. IEEE Transactions on Semiconductor Manufacturing, 28(3), 374-384.
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.
Chen, T., & Wang, Y.-C. (2011). A hybrid fuzzy and neural approach for forecasting the book-to-bill ratio in the semiconductor manufacturing industry. The International Journal of Advanced Manufacturing Technology, 52(1-4), 377-389.
Chien, C.-F., Chen, Y.-J., & Peng, J.-T. (2010). Manufacturing intelligence for semiconductor demand forecast based on technology diffusion and product life cycle. International Journal of Production Economics, 128(2), 496-509.
Ching-Chin, C., Ieng, A. I. K., Ling-Ling, W., & Ling-Chieh, K. (2010). Designing a decision-support system for new product sales forecasting. Expert Systems with Applications, 37(2), 1654-1665.
Cui, R., Gallino, S., Moreno, A., & Zhang, D. J. (2017). The operational value of social media information. Production and Operations Management.
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.
Digital Market Outlook. (2018). Retrieved from https://www.statista.com/outlook/digital-markets
Ertugrul, Ö. F. (2016). Forecasting electricity load by a novel recurrent extreme learning machines approach. International Journal of Electrical Power & Energy Systems, 78, 429-435.
Gaur, V., Kesavan, S., Raman, A., & Fisher, M. L. (2007). Estimating demand uncertainty using judgmental forecasts. Manufacturing & Service Operations Management, 9(4), 480-491.
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.
Lee, C.-Y., & Chiang, M.-C. (2016). Aggregate demand forecast with small data and robust capacity decision in TFT-LCD manufacturing. Computers & Industrial Engineering, 99, 415-422.
Lolli, F., Gamberini, R., Regattieri, A., Balugani, E., Gatos, T., & Gucci, S. (2017). Single-hidden layer neural networks for forecasting intermittent demand. International Journal of Production Economics, 183, 116-128.
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.
SAMUEL, A. (1959). Some stuies on Machine Learning Using the Game of Checkers. IBM Journal on Research and Development, 3, 210-229.
Statistics. (2018). Retrieved from
https://www.itu.int/en/ITU-D/Statistics/Pages/stat/default.aspx
Tarng, M.-Y. (1998). IC Wafer Demand Estimation| A Multicriteria Approach with Modified Analytic Hierarchy Process. Information and Management, 9(3), 1-17.
Wang, C.-H., & Hsu, L.-C. (2008). Using genetic algorithms grey theory to forecast high technology industrial output. Applied mathematics and computation, 195(1),　256-263.
Worldwide Quarterly Server Tracker. (2018). Retrieved from https://www.idc.com/tracker/showtrackerhome.jsp
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.