| Summary: | 碩士 === 國立臺灣科技大學 === 工業管理系 === 95 === The attributes of the high-tech industries are high degree of uncertainty and short product life cycle. TFT-LCD industry is growing in Taiwan. LCD-monitor is a major component for a desktop computer. Demand planning can effectively reduce inventories and bring benefits. The difficulty of demand planning is uncertainty. We can use forecasting methods to improve the accuracy of demand planning. For example, we can use combining forecasts method which is associated with different forecasting methods. Many experts point out that combining forecast is more useful than any single forecasting method. There are many different ways to estimate the parameters of combining forecasts. In this study, we use genetic algorithm (GA) to find combining weights and compare with other traditional methods such as k method, adaptive set of weights and linear composite. The results show that GA approach out performs other method. We combine the global demand forecasts which is multiply by market share and the supply forecasts of company. Using combining forecasts, we can increase the stability and accuracy of the demand planning.
|
|---|
