對於各產業在需求規劃方面都會存在著許多的不確定性問題，而如何減少成本的損失及帶來利益，準確的需求預測更顯得重要。本研究就是希望透過多種預測方法來作為需求規劃時的決策，以利整個供應鏈的生產排程及產能供給上能有好的成效並減少長鞭效應的產生。而組合預測方法就是對於多個不同的個別預測方法做組合，並運用於需求規劃上來獲得更佳的預測績效。許多的學者也指出，組合預測的預測績效比任何個別預測方法的成效要來的好，且非線性組合預測方法又比線性組合預測方法績效來的好。所以在本研究中，以英國境內隨機的11組ATM現金需求量為預測標的，藉由組合預測結合兩個個別預測方法預測值所得結果，來達到對該產業進行需求規劃時的穩定性及精確度，並驗證非線性組合方法在預測的效果上更為顯著。而在組合預測的參數估計上，線性組合方面則使用k值權重法、調適權重法、線性迴歸組合法；非線性組合方面則使用適應性模糊類神經網路做訓練學習，期望可以找出最適合的權重來進行組合預測。

Demand planning in many industries exists uncertainty. To reduce the costs and increase benefits, the accuracy of demand forecasting becomes an important task. This research investigates the policy of demand planning through many kinds of forecasting methods, it will improve the performance in production schedule and productivity supply and reduce bullwhip effect. Combined forecasts method is to combine different forecasting methods. Many experts point out that combined forecast is more useful than any individual forecasting methods in prediction performance. In addition, nonlinear combined forecast is better than linear combined forecast. We use 11 groups of ATM cash demand at random within the territory of England as the target of prediction, by combining two individual forecasting methods’ predicted value to reach stability and accuracy of carrying on the demand while planning for this industry and prove that the nonlinear combined method is more apparent on the result that is predicted. To estimate the parameters of linear combined forecasts we use adaptive set of weights, k method and linear composite for nonlinear combined forecasts, we use the adaptive fuzzy neural networks to train and study and the results show that this method provides the most suitable weights for combine forecasts.

中文部分
[1] 于宗先，「經濟預測」，台北：大中國出版社，1972。
[2] 方世傑，「市場預測方法一百種」，台北：書泉出版社，1996。
[3] 王進德，「類神經網路與模糊控制理論入門與應用」，台北：全華科技出版社，2007。
[4] 徐義人，「應用模糊理論於颱風降雨量之推估」，國立成功大學水利及海洋工程學系碩士論文，2004。
[5] 張斐章、張麗秋、黃浩倫，「類神經網路理論與實務」，台北：東華書局，2003。
[6] 張致祥，「應用適應性模糊系統於磨後工件表面品質之預測」，國立高雄第一科技大學，2004。
[7] 蔡嘉文，「應用模糊神經網路於股價預測之研究」，國立成功大學工業管理研究所碩士論文，1996。

英文部分
[1] Bates, J. M. and Granger, C. W., “The combination of forecasts”, Operational Research Quarterly, Vol.20, No.4, pp.451-468, 1969.
[2] Bayus, B. L., “An analysis of product lifetimes in a technologically dynamic industry”, Management Science, Vol.44, No.6, pp.763-775, 1998.
[3] Box, G. E. P. and Jenkins, G. M., “Time series analysis forecasting and control”, 1st edition, San Francisco: Holden-Day, 1970.
[4] Bunn, D.W., “A comparative evaluation of the outperformance and minimum variance procedures for linear syntheses of forecasts”, Operational Research Quarterly, Vol.28, No.3, pp.653-662, 1977.
[5] Chase, C. W. Jr., “Composite forecasting : combining forecasts for improved accuracy”, The Journal of Business Forecasting, Vol.19, No.2, pp.2-22, Summer2000.
[6] Clemen, R. T. and Winkler, R. L., “Combining economics forecasts”, Journal of Business and Economics Statistics, Vol.4, No.1, pp.39-46, 1986.
[7] Clemen, R. T., “Combining forecasts: A review and annotated bibliography”, International Journal of Forecasting, Vol.5, No.4, pp. 559-583, 1989.
[8] Crane, D. B. and Crotty, J. R., “A two-stage forecasting model: Exponential smoothing and multiple regression”, Management Science, Vol.13, No.8, pp.501-507, 1967.
[9] Dickinson, J. P., “Some statistical results on the combination of forecasts”, Operational Research Quarterly, Vol.24, No.2, pp.253-260, 1973.
[10] Diebold, F. X. and Pauly, P., “Structural change and the combining forecasts”, Journal of Forecasting, Vol.6, No.1, pp.21-40, 1987.
[11] Diebold, F. X., “Serial correlation and the combination of forecasts”, Journal of Business and Economics Statistics, Vol.6, No.1, pp.105-111, 1988.
[12] Donaldson, R. G. and Kamstra, M., “Forecast combining with neural networks”, Journal of Forecasting, Vol.15, No.1, pp.49-61, 1996.
[13] Dong , J. R., “A nonlinear combining forecast method based on fuzzy neural network”, Proceedings of 2002 International Conference on Machine Learning and Cybernetics, Vol.4, pp.2160-2164, 2002.
[14] Engle, R. F., Granger, C. W. J. and Kraft, D., “Combining competing forecasts of inflation using a bivariate ARCH model”, Journal of Economic Dynamics and Control, Vol.8, No.2, pp.151-165, 1984.
[15] Fiordaliso, A., “A nonlinear forecasts combination method based on Takagi-Sugeno fuzzy systems”, International Journal of Forecasting, Vol.14, No.3, pp.367-379, 1998.
[16] Granger, C. W. J. and Ramanathan, R., “Improved method of combining forecasts”, Journal of Forecasting, Vol.3, No.2, pp.197-204, 1984.
[17] Hendry, D. F. and Clements, M. P., “Pooling of forecasts”, Econometrics Journal, Vol.7, No.1, pp.1-31, 2004.
[18] Jang, J. S. R., “ANFIS: Adaptive network-based fuzzy inference system”, IEEE Transactions on Systems, Man and Cybernetics, Vol.23, No.3, pp.665-685, 1993.
[19] Levenhach, H. and Cleary, J. P., “Forecasting: Practice and process for demand management”, Belmont: Duxbury, 2006.
[20] Lewis, C. D., “Industrial and business forecasting methods”, London: Butterworths, 1982.
[21] Li, F. and Tkacz, G., “Combining forecasts with nonparametric kernel regressions”, Studies in Nonlinear Dynamics and Econometrics, Vol.8, No.2, Article 2, 2002.
[22] Lo, S. P., “An adaptive-network based fuzzy inference system for prediction of workpiece surface roughness in end milling”, Material Processing Technology, Vol.142, No.3, pp.665-675, 2003.
[23] Makridakis, S., “Why combining works?”, International Journal of Forecasting, Vol.5, No.4, pp.601-603, 1989.
[24] Nelson, C. R., “A benchmark for the accuracy of econometric forecasts of GNP”, Business Economics, Vol.19, No.3, pp.52-58, 1984.
[25] Oller, L. E., “A method for pooling forecasts”, Journal of the Operational Research Society, Vol.29, No.1, pp.55-63, 1978.
[26] Reid, D. J., “A comparative study of time series prediction techniques on economic data”, University of Nottingham, Nottingham, 1969.
[27] Robert, T. C. and Robert, L. W., “Combining economic forecasts”, Journal of Business and Economic Statistics, Vol 4, No.1, pp.39-46, 1986.
[28] Rumelhart, D. E. and McClelland, J. L., “Parallel distributed processing : Explorations in the microstructure of cognition”, MIT press, Cambridge, MA, Vol.1, pp.547, 1986.
[29] Shi, S. M., Xu, L. D. and Liu, B., “Improving the accuracy of nonlinear combined forecasting using neural networks”, Expert Systems with Applications, Vol.16, No.1, pp.49-54, 1999.
[30] Stock, J. H. and Watson, M. W., “Combination forecasts of output growth in a seven country data set”, Journal of Forecasting, Vol.23, No.6, pp.405-430, 2004.
[31] Sugeno, M. and Kang, G. T., “Structure identification of fuzzy model”, Fuzzy Sets System, Vol.28, No.1, pp.15-33, 1988.
[32] Szecsi, T., “Cutting force modeling using artificial neural networks”, Journal of Materials Processing Technology, Vol.92, No.93, pp.344-349, 1999.
[33] Takagi, T. and Sugeno, M., “Fuzzy identification of systems and its applications to modeling and control”, IEEE Transactions on Systems, Man, Cybernetics, Vol.15, pp.116-132, 1985.
[34] Wheelwright, S. C. and Makridakis, S., “Examination of the use of adpative filtering in forecasting”, Operational Research Quarterly, Vol.24, No.1, pp.55-64, 1973.
[35] Wilson, J. H. and Keating, B., “Forecasting business”, 5th edition, McGraw-Hill Companies Inc., New York, 2007.
[36] Winkler, R. L. and Clemen, R. T., “Multiple experts vs. Multiple methods: Combining correlation assessments”, Decision Analysis, Vol.1, No.3, pp.167-176, 2004.
[37] Yao, Y., Lian, Z. W., Hou, Z. J. and Liu, W. W., “An innovative air-conditioning load forecasting model based on RBF neural network and combined residual error correction”, International Journal of Refrigeration, Vol.29, No.4, pp.528-538, 2006.
[38] Zadeh, L. A., “Fuzzy sets”, Information and Control, Vol.8, pp.338-353, 1965.
[39] Zamowitz, V., “An appraisal of short-term economic forecasts”, National Bureau of Economic Research, New York, pp.123-125, 1967.

網站部份
[1] BBCChinese.com /金融財經報導
http://news.bbc.co.uk/chinese/trad/low/newsid_6220000/newsid_6227900/6227950.stm ,2007.
[2] NN5 , http://www.neural-forecasting-competition.com , 2008.