An Analysis of Capacity Investment Strategies in Semiconductor Manufacturing
碩士 === 國立臺灣大學 === 工業工程學研究所 === 91 === Semiconductor manufacturing is a high-tech and capital-intensive industry. Market participants invest large amount of capital in capacity expansions and technology enhancements. One of the key profitability factors of this industry is to expand capacity at the r...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2003
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/38292141386730634457 |
| Summary: | 碩士 === 國立臺灣大學 === 工業工程學研究所 === 91 === Semiconductor manufacturing is a high-tech and capital-intensive industry. Market participants invest large amount of capital in capacity expansions and technology enhancements. One of the key profitability factors of this industry is to expand capacity at the right time with the right amount. Although larger capacity can meet higher demands and increase sales revenues in boom time, it also incurs undesirable depreciation costs in the subsequent periods. Huge fixed costs also expose the company to the risk of volatile demands and may jeopardize the company’s financial health in recession periods.
This research proposes two analysis methods of rational capacity investment and a capacity expansion decision model. The capacity expansion model is to find out the optimal capacity increment that maximizes the Return on Asset (ROA) under uncertain demands in each decision horizon. The model is tested on the empirical results (1994-2002) of a case company and other demand scenarios. This research shows that the new capacity expansion method improves the investment rationality of the case company. It enhances company’s return on investment and delivers conservative coverage ratio in all periods.
The conclusion of this study is that aggressively adding capacity to meet sudden demand jump will increase revenues in the short term at the expense of company’s ROA in the long term. An ROA based capacity expansion decision model avoids this problem and is a better alternative than traditional methods such as NPV, IRR, etc.
|
|---|