Using Aritifical Neural Network on the Sales forecast of beer

• Main Authors: Jang-Song Lin, 林振松
• Other Authors: Tung-Hsu Hou
• Format: Others
• Language: zh-TW
• Published: 2001
• Online Access: http://ndltd.ncl.edu.tw/handle/37828196744432331384

碩士 === 國立雲林科技大學 === 工業工程與管理研究所碩士班 === 90 === It is very important for the manager of an enterprise to make a right strategy for the change and demand in a competitive market. However, the indeterminable factors in the market bring the great risk when the manager makes strategy. Therefore, if the manager can accurately forecast the future of the market, he can reduce the risk and well manage in the market. Furthermore, the enterprise can operate the business permanently. In 1994, the Government started to release the import restriction of beer for certain country such as Japan. The sales volume of Taiwan Beer was deeply affected by the import beer and the market share decreased gradually. The first important mission of Taiwan tobacco & wine monopoly bureau is to accurately forecast the sales volume of beer market in Taiwan and makes a proper strategy to keep and increase the market share. For the reasons aforementioned, how to choose the accurate and suitable forecast tool to estimate the sales volume is extremely important for an enterprise. Therefore, this research is aimed to use the unique function of Artificial Neural Network to forecast the sales volume of beer in Taiwan. Besides, this research also use others forecasting tools to compared with Artificial Neural Network. The result of this research shows the forecasting accuracy of Artificial Neural Network is better than others. It is believed that Artificial Neural Network is an excellent forecast tool for the sales volume of beer.