摘要
Hsinchu Science Park (HSP) is a high-tech cluster where semiconductor industry plays a decisive role in Taiwan's economy as well as global supply chains. Semiconductor industry is capital intensive, in which capacity utilization significantly affects the capital effectiveness and profitability of semiconductor companies. Thus, demand forecasting provides critical input to support strategic decisions of capacity planning and the associated capital expenditure that require long lead-time. This study aims to predict the sales of semiconductor industry in HSP as a reference signal for supporting the decisions of individual companies and the government to maintain a healthy ecosystem. Empirical data of semiconductor industry in HSP from 1983 to 2010 was collected and analyzed. Furthermore, this study incorporated historical events to adjust the prediction. The results have shown practical viability of this research to support companies to improve the demand forecast as well as to make strategic decisions for semiconductor ecosystem as a whole. © 2012 Copyright Taylor and Francis Group, LLC.