手機五大品牌最適銷售預測法之研究-Nokia, Motorola, Samsung, LG, Sony Ericsson

Five Brand Mobile Phone Sales Forecast the Optimal Method of Research-Nokia, Motorola, Samsung, LG, and Sony Ericsson

陳林宗(Lin-Zong Chen)

指導教授 ： 江瑞清

中原大學/電機資訊學院/工業與系統工程研究所/碩士(2009年)

https://doi.org/10.6840/cycu200900779

摘要

手機產業在全球電子資訊市場中，扮演著相當重要的角色。許多零組件及代工廠會依照客戶所給的長期需求調整後進行季度及年度產能規劃並利用短期需求做備料計劃，在機構零組件上尤其重要。工廠端需要有預測的能力來判斷品牌業者的長期及短期需求量是否合理才能確保進料的價格低及不會造成大量庫存。
目前尚無單一的預測法可以完全來預測手機各品牌業者的實際需求，產業普遍存在庫存堆積與部分停工待料的問題。然而，透過適合的預測法需求方式可以幫助零組件及代工廠有效減少這類問題發生的機率與所造成的成本損失。因此，本研究提出由各式預測法中應用所收集的數據尋求五大品牌業適合的預測 模型。藉著分析預測模型來幫助零組件及代工廠做正確銷售預測、產能規劃及備料判斷。
數據量及數據趨勢去選取合適的預測方法，並參考文獻及書籍選取適合的預測模型進行比較，當平均絕對誤差百分比，平均絕對偏差 及平均誤差做為最小時我們判定為最佳預測模型。
實驗結果顯示，NOKIA 、SAMSUNG、MOTOROLA 及SONYERISSON 使用Winter ‘s Method 較準確而 LG 則以多項回歸為最佳，這樣的評估結果可以提供相關產業進行需求預測時之參考。

關鍵字

灰色預測 ； 銷售預測 ； 溫氏指數平滑法 ； 回歸預測

並列摘要

Mobile phones in the global electronic industry marketing play an important role. Components suppliers and OEM Company will be in accordance with customer long-term demand to adjusted quarterly and annual production capacity and use Short-term demand to do material planning and preparation .This is particular important for Components supplier and OEM Company.

Factory need to have forecast ability to judge the long-term & short term demand if reasonable that Brand Company provided to them. This is in order to guarantee lowest purchase prices and no over stock in Factory site.

Currently,there is no single forecasting method to predict the actual demand which the brand company needed. In this field has serious inventory problems & production suspension issue. Therefore, through appropriate demand forecasting method can help components & OEM Company effectively reduce the probability of occurrence of such problems and costs caused by losses.

Accordingly, study various types of prediction method by the application of the data collected to find out the best prediction model. Through the analysis and forecast models help Company to do the right judge in planning and preparation capacity.

Collecting data and doing data trends analysis to select appropriate forecasting methods, and reference literature and books to select the prediction model for comparison. When the percentage of the average absolute error, mean absolute deviation and the average error as the assessment criteria is all minimum that we say it is the best forecasting model.

Experimental results show that, NOKIA, SAMSUNG, MOTOROLA and SONYERISSON use winter’s Method and LG by more precise number of returns for the best, so that the results of the assessment of related industries can provide forecasts of the demand for reference.

並列關鍵字

Gray Forecast ； Sales Forecasting ； Exponential Smoothing Method Atsushi ； Regression

參考文獻

G. J., Tan, “The structure method and application of background value in grey system GM (1, 1)”Systems Engineering-theory and Practice, no. 4, pp.98-103, 2000.

Luxh, J. T., J O. Riis, B. Stensballe, 1996, “A hybrid econometric-neural network modeling approach for sales forecasting”, The International Journal of Production Economics, vol.43, pp.175-192.

Chin-Tsai Lin and Shih-Yu Yang, (2003), “Forecast the Output Value of Taiwan's Opto-electronics Industry Using the Grey Forecasting Model,” Technological Forecasting & Social Change 70, pp. 177-186, 2003.

Hsu, C. C., and Chen, C. Y., “Applications of improved grey prediction model for power demand forecasting,” Energy Conversion and Management 44, pp.2241-2249, 2003.

Markham, I. S. and Rakes, T. R. (1998).“The effect of sample size and variability of data on the comparative performance of artificial neural networks and regression”, Computers Operations Research, vol.25(4), pp.251-263