Further Exmination to Fuzzy Exponential Smoothing Model

深度探討模糊指數平滑模式

曹銳勤(Ruey-Chyn Tsaur)

《工業工程學刊》 22卷6期 (2005/11) Pp. 521-542

https://doi.org/10.29977/JCIIE.200511.0008

摘要

模糊指數平滑模式對於資料量稀少或系統結構模糊不清之問題，能求解吻合資料未來趨勢的外差模糊預測值，以成功求解未來不確定高的短期預測問題。為了提高模糊指數平滑模式的可應用性，此模式已成功於研究並提出求解趨勢及季節型資料之問題。然而，模糊指數平滑模式有兩個特性尚未被清楚的討論。首先，以灰色模式GM(1,1)估計的前期平滑值需被進一步討論與確認。第二，模糊指數平滑模式中的模糊限制式之容忍值如何取捨需進一步的解釋。因此，在此研究中，我們針對上述之二問題加以深度探討以提高模糊指數平滑模式的可應用性。並以一範例舉例說明。

關鍵字

預測 ； 指數平滑模式 ； 模糊指數平滑模式 ； 灰色模式GM 1,1 ； 模糊限制式

並列摘要

The fuzzy exponential smoothing model has been used for short-term forecasting with a minimum of collected data or unknown system structure to determine a better extrapolative interval in a fuzzy set for an unknown future trend. In order to enlarge its applications, this model has been investigated and proposed for solving trend datum and seasonal datum. However, two characters in the fuzzy exponential smoothing model have never been clearly discussed. First, the old smoothed values estimated from the grey model GM(1,1) need further discussion and clarification. Second, the meanings of the tolerance values of the fuzzy constraints in the fuzzy exponential smoothing model require further elucidation. Therefore, in this study, in-depth examinations of fuzzy exponential smoothing model are carried out in order to elucidate the above two problems and to enhance the abilities of the fuzzy exponential smoothing model. An example is given for illustration.

並列關鍵字

Forecasting ； exponential smoothing model ； fuzzy exponential smoothing model ； grey model GM 1,1 ； fuzzy constraint

參考文獻

Deng, L.J.(1986).(Grey Forecasting and Decision).

Hanke, J.E.,A.G Reitsch(1998).(Business Forecasting).

Hsu, C.I.,Y.H. Wen(1998).Improved grey prediction models for the trans-pacific air passenger market.Transportation Planning and Technology.22(2),87-108.

Montgomery, D.C.,E.A. Peck(1982).(Introduction to Linear Regression Analysis).

Pantazopoulos, S.N.,C.P. Pappis(1996).A new adaptive method for extrapolative forecasting algorithms.European Journal of Operational Research.94,106-111.