世界經濟潮流導致經濟的競爭加劇，產品開始由大量規格化生產轉變為少量多樣
及客製化的生產，產品生命週期變的短暫且快速，在高投資成本的電子產業需求不再
是容易預測的長期線性趨勢，企業經營與獲利必須因應環境快速的變動做調整。在產
品生產初期由於成本考量及縮短開發階段時間，普遍只能獲得有限的資料，而傳統的
預測技術大多需要大量的資料分析再給予決策，已經無法符合現有的緊迫需求，因此
決策者在有限的樣本下做出迫切性判斷已是無法迴避，而小樣本資料的預測也變成重
要的分析工具。
灰預測模型為小樣本預測的重要方法之一，其中大多採用固定時間間距的建模方
式，而限制了模型的廣泛應用。本研究透過趨勢潛力追蹤法分析資料行為，以擷取非
等間距資料的隱含資訊找出趨勢潛力值，在灰色系統理論的架構下，發展出一種良好
適應性的灰預測模型，做為小樣本資料的分析預測工具。經過實例驗證，在本研究所
提出的方法能依照樣本特性建構適合的模型，更可同時成功地提高小樣本資料的預測
準確度。
關鍵字：灰預測、非等間距、小樣本、時間數列。

The trend of global economy increases industrial competition. Products experience
great changes from mass production into customization and lean production. The life cycle
of products is shortened and rapid. Demand is no longer a long-term linear trend easily
forecasted in high cost electronics industry. Industrial management and profits are in need
to adjust to the rapid change in the business nature. In the early stage of manufacturing, it
is ordinary that only limited data can be obtained due to the consideration of cost and time
spent on invention. However, traditional forecasting skills always require mass data for
analysis before making any decisions, which cannot meet current urgent demands. As a
result, it is unavoidable that decision makers are forced to make rush decisions with limited
samples. Therefore, the forecasting of small data sets is being valued as a vital analytical
method.
Grey forecasting model is one of the important approaches of small-data-set
forecasting. It always establishes models by fixing time span, which leads to the models’
limited application. This study analyzed data with trend and potency tracking method and
discovered the generation of trend and potency value by collecting extra information of
unequal span data. The study was aimed at developing an adaptive grey forecasting model
under the scheme of grey system theory to serve as a forecasting tool of small-data-set
analysis. Empirical evidence proved that the approach proposed in the study could not only
establish adaptive models in accordance with the characteristics of samples but
successfully improve the forecasting precision of small data sets.
Key words: Grey forecasting, Non-equigap, Small data set, Time series

Chang, C. J., Adaptive grey forecasting model. Master’s thesis, National Cheng Kung
University, Tainan, Taiwan, 2008.
Chang, S. C., Lai, H. C., Yu, H. C., A variable P value rolling grey forecasting model for
Taiwan semiconductor industry production. Technological Forecasting & Social
Change, 72(5), 623-640, 2005.
Chang, S. C., Wu, J. H., Lee, C. T., A study on the characteristics of α(k) of grey
prediction. The 4th National Conference on Grey Theory and Applications, Kaohsiung,
Taiwan, 291-296, 1999. (in Chinese)
Chen, C. I., Application of the novel nonlinear grey Bernoulli model for forecasting
unemployment rate. Chaos, Solitons and Fractals, 37(1), 278-287, 2006.
Chen, H. S., Chang, W. C., A study of optimal grey model GM(1,1). Journal of Grey
System, 1(2), 141-145, 1998. (in Chinese)
Chen, K. W., Lai, C. J., Optimal fixed α in Z (1) (k) for GM(1,1). The 6th National
Conference on Grey Theory and Applications, Yunlin, Taiwan, A26-A31, 2001. (in
Chinese)
Cheng, K. H., Shah, H. C., A new method for earthquake forecasting using gery theory:
Application to California. The Journal of Grey System, 11(3), 293-302, 1999.
Dai, W. I., Li, J. F., Modeling research on non-equidistance GM(1,1) model. Systems
Engineer-Theory & Practice, 25(9), 89-93, 2005. (in Chinese)
Deng, J. L., Control Problems of Grey Systems. Systems and Control Letters, 1(5),
288-294, 1982.
Deng, J. L., Grey System Fundamental Method. Huazhong University of Science and
Technology Press, Wuhan, China, 1987. (in Chinese)
Deng, J. L., Introduction to Grey System Theory. The Journal of Grey System, 1(1), 1-24,
1989.
Deng, J. L., A novel GM(1,1) model for non-equigap series. The Journal of Grey System,
9(2), 111-116, 1997.
He, X. G., Sun, G. Z., A non-equigap grey model NGM(1,1). The Journal of Grey System,
13(2), 189-192, 2001.
He, Y., Bao, Y. D., Grey-Markov forecasting model and its application. Systems
Engineer-Theory & Practice, 12(4), 59-63, 1992. (in Chinese)
Hsin, J. Y., Tsai, Y. P., The research of superposition method for α value in grey
forecasting. The 5th National Conference on Grey Theory and Applications, Taipei,
Taiwan, 305-308, 2000. (in Chinese)
Hsu, C. C., Chen, C. Y., Applications of improved grey prediction model for power
demand forecasting. Energy Conversion and Management, 44(14), 2241-2249, 2003a.
Hsu, C. C., Chen, C. Y., A modified Grey forecasting model for long-term prediction.
Journal of the Chinese Institute of Engineers, 26(3), 301-308, 2003b.
Hsu, C. I., Wen, Y. H., Improved grey prediction models for the trans-pacific air passenger
market. Transportation Planning and Technology, 22(2), 87-107, 1998.
Hsu, L. C., Applying the Grey prediction model to the global integrated ciruit industry.
Technological Forecasting & Social Change, 70(6), 563-574, 2003.
Huang, C., Principle of information diffusion. Fuzzy Sets and Systems, 91(1), 69-90, 1997.
Huang, C., Moraga, C., A diffusion-neural-network for learning from small samples.
International Journal of Approximate Reasoning, 35(2), 137-161, 2004.
Huang, S. X., Li, Z. C., Grey modeling of non-equidistant data sequent for forecasting
subsidence of engineering buildings. Geospatial Information, 2(1), 41-43, 2004. (in
Chinese)
Jang, J. S. R., ANFIS: Adaptive-network-based fuzzy inference system. IEEE Transactions
on Systems, Man and Cybernetics, 23(3), 665-685, 1993.
Jiang, S. S., E, J. Q., Li, J., Zhang, H. M., Gong, J. K., Yuan, W. H., Prediction on residual
service life of mine-vehicle engine cylinder wall by using of non-equidistance grey
forecasting model. Goal Mine Machinery, 29(7), 69-71, 2008. (in Chinese)
Li, D. C., Chen, L. S., Lin Y. S., Using functional virtual population as assistance to learn
scheduling knowledge in dynamic manufacturing environments. International Journal
of Production Research, 41(17), 4011-4024, 2003.
Li, D. C., Wu, C. S., Chang, F. M., Using data-fuzzification technology in small data set
learning to improve FMS scheduling accuracy. International Journal of Advanced
Manufacturing Technology, 27(3-4), 321-328, 2005.
Li, D. C., Lin, Y. S., Using virtual sample generation to build up management knowledge
in the early manufacturing stage. European Journal of Operational Research, 175(1),
413-434, 2006a.
Li, D. C., Wu, C. S., Tsai, T. I., Chang, F. M., Using mega-fuzzification and data trend
estimation in small data set learning for early FMS scheduling knowledge. Computer
& Operations Research, 33(6), 1857-1869, 2006b.
Li, D. C., Wu, C. S., Tsai, T. I., Lina, Y. S., Using mega-trend-diffusion and artifical
samples in small data set learning for early flexible manufacturing system scheduling
knowledge. Computer & Operations Research, 34(4), 966-982, 2007.
Li, D. C., Yeh C. W., A non-parametric learning algorithm for small manufacturing data
sets. Expert Systems with Applications, 34(1), 391-398, 2008.
Li, Y. G., Li, Q. F., Zhao, G. F., An improvement on Grey forecasting model. System
Engineering, 10(6), 27-31, 1992. (in Chinese)
Lin, C. T., Yang, S. Y., Forecast of the output value of Taiwan’s opto-electronics industry
using the Grey forecasting model. Technological Forecasting & Social Change, 70(2),
177-186, 2003.
Lin, C. T., Yeh, H. Y., The use of grey prediction to forecast Taiwan stock index option
prices. The Journal of Grey System, 18(4), 381-390, 2006.
Liu, S. F., Dang, Y. G., Fang, Z. G., The Theory of Grey System and Its Applications.
Science Press, Beijing, China, 2004. (in Chinese)
Liu, W. G., Jiang, L. H., Grey GM(1,1) model in ferroalloy burdening. Iron and Steel,
31(9), 52-56, 1996. (in Chinese)
Lu, H. C., Universal GM(1,1) model based on data mapping concept. The Journal of Grey
System, 8(4), 307-319, 1996.
Luo, D., Liu, S. F., Dang, Y. G., The optimization of grey model GM(1,1). Engineering
Science, 5(8), 50-53, 2003. (in Chinese)
Luo, E. X., Qian, X. S., Li, R., Construction and empirical research of the variable
parameter value rolling grey forecasting model. Journal of University of Shanghai for
Science and Technology, 28(5), 465-468, 2006. (in Chinese)
Mao, M., Chirwa, E. C., Combination of grey model GM(1,1) with three-point moving
average for accurate vehicle fatality risk prediction. International Journal of
Crashworthiness, 10(6), 635-642, 2005.
Mao, M., Chirwa, E. C., Application of grey model GM(1,1) to vehicle fatality risk
estimation. Technological Forecasting & Social Change, 73(5), 588-605, 2006.
Niyogi, P., Girosi, F., Poggin, T., Incorporating prior information in machine learning by
creating virtual examples. Proceedings of the IEEE, 86(11), 2196-2209, 1998.
Pan, C. L., Huang, Y. F., Lin, G., The study on α algorithm of grey prediction Z (1) (k)
with iterative method as basis. The 7th National Conference on Grey Theory and
Applications, Tainan, Taiwan, I27-I32, 2002.
Sheng, J. G., Improvement on and applying of GM(1,1). Mathematics in Practice and
Theory, 30(3), 10-15, 1990. (in Chinese)
Shi, B. Z., Modeling of non-equigap GM(1,1). The Journal of Grey System, 5(2), 105-113,
1993.
Song, Z. M., Tong, X. J. Xiao, X. P., Center approach grey GM(1,1) model. Systems
Engineer-Theory & Practice, 21(5), 110-113, 2001. (in Chinese)
Sun, G., Prediction of vegetable yields by grey model GM(1,1). The Journal of Grey
System, 3(2), 179-187, 1991.
Tan, G.. J., The structure method and application of background value in grey system
GM(1,1) model (Ⅰ). Systems Engineer-Theory & Practice, 20(4), 98-103, 2000. (in
Chinese)
Tien, T. L., Chen, S.P, Residual correction method of Fourier series to GM(1,1) Model. The
1st National Conference on Grey Theory and Applications, Kaohsiung, Taiwan,
93-101, 1996. (in Chinese)
Zhou, C. Y., Li, D. F., Liu, Z. X., Grey predicting the soft ground settlement via GM(1,1).
The Journal of Grey System, 11(4), 397-402, 1999.
Zhou, P., Ang, B. W., Poh K. L., A trigonometric grey prediction approach to forecasting
electricity demand. Energy, 31(14), 2839-2847, 2006.