| Summary: | 碩士 === 國立臺北科技大學 === 工業工程與管理系碩士班 === 101 === Steel industry is the fundamental industry of a nation, and it has a closed connection between upstream and downstream. Its development has a great concern with the economic stability and national defense independence of a nation; thus, it represents the nation’s power and is highly emphasized by governments. The related studies on domestic steel industry mostly focus on the current situation of steel industry, for instance, the cooperation in national economic development and the status of supply and demand of steel market. Therefore, via constructing the prediction systems, the findings of the study can show the prediction of the price fluctuation of steel products and provide the steel industry criteria of right purchasing time and quantity, enhancing the effects of the whole performance of Taiwan’s steel industry.
This study investigates that, under the two modes of price fluctuation, steel productions have different examinations and different comparisons as the results via two prediction systems: Honey-Bees Mating Optimization Self-Organizing Map (HBMOSOM) and Self-Organizing Map (SOM). The results prove that HBMOSOM prediction system could reach the higher prediction accuracy rate than SOM prediction system. Under the mixed mode, the prediction accuracy rate of rebar is much higher in HBMOSOM prediction system, while the key factors affecting the price are the price of raw materials, the price of crude oil and the stock price of Feng Hsin Iron & Steel Corporation. Under the divided mode, the prediction accuracy rate of HR stainless coil is much higher in HBMOSOM prediction system, while the key factors affecting the price fluctuation of HR stainless coil are the price of raw materials and its apparent consumption.
|
|---|
