Applying Grey Theory to Building the Forecasting Model of Food Industry – A Case Study of Hai Ha Confectionary Company

• Main Authors: Vu Hao Tinh, 武好情
• Other Authors: Lai Wang Wang
• Format: Others
• Language: en_US
• Published: 2015
• Online Access: http://ndltd.ncl.edu.tw/handle/vt629f

碩士 === 國立高雄應用科技大學 === 製造與管理外國學生碩士專班 === 103 === Forecasting is the process and technology of projecting the future based on the previous data, which may vary from very simple to complex or sophisticated methods depending on the requirements needed. For example, a company’s financial forecasts focus on its future financial prospects derived from its historical data as well as the demand for capital necessary for the coming year. Similarly, a company also very emphasizes its sales. By the same token, it will pay attention to its historical sales records and come up with the projected forecasting that can be used as the sales goals and business operation basis. However, the forecasting cannot get a satisfactory result without a good forecasting method. Specific forecasting methods or techniques are also used which companies or industries own few data or lack enough parameters. Accordingly, in this study Grey system theory is employed to forecast the sales forecast where data are few and the behaviors of systems are unknown. Data used in this study are obtained from the annual financial statements of the Hai Ha confectionery company for the forecasting of net sales in the coming two years (i.e., 2014 and 2015). For the current research, in the first place the original predicted values of net sales is obtained individually by the GM(1,1) model, the Verhulst model and the DGM(2,1) model. Secondly, three models are used to test the unknown behaviors of the sales forecast. The results of these models in terms of predicting net sales show that the forecasting accuracy of the DGM (2, 1) is higher than the GM (1, 1) model and the Verhulst model. Keywords: Grey System Theory, GM (1.1) Model, Verhuslt Model, DGM (2, 1) Model