台灣大尺寸TFT-LCD產業經營績效評比之研究

A Comparative Study of Operational Performance for Large Size TFT-LCD Manufacturers in Taiwan

吳聲貴(Sheng-Kuei Wu)

指導教授 ： 徐旭昇

元智大學/工程學院/工業工程與管理學系/碩士(2009年)

摘要

面對TFT-LCD 面板產業屬於高度整合的產業，其未來的發展，著實存在著許多不確定的因素，實在難以掌控，此時提供台灣面板產業所需的產業資訊，使台灣TFT-LCD 產業在日、韓等國的激烈競爭之下脫穎而出，實屬重要，期能藉由兩種方法資料包絡分析法(DEA) 及層級程序分析法(AHP)，進行評比分析，比較兩種方法的差異處及結果，達到企業資源投入分配及經營方向之建議。
本研究在第一種方法DEA視窗分析評比結果，各家大尺寸TFT-LCD產業經營績效評比每階段未達到相對有效率，顯示各家之投入資源的運用效率可再改善，在整體之技術效率，則顯示出各家投入的資源管理不佳造成近3.63% 的技術無效率；此外，Malmquist 生產力指數評比結果，整體平均生產力衰退了1.2%，主要來自於技術效率不佳-0.8%，觀察個別公司的年平均變化值，除了群創的生產力提升外，其餘4 家面板廠的生產力皆呈現下降的趨勢。第二種方法AHP的探討結果，其評核項目與DEA方法中的重要指標項目大致雷同，增加主觀性的產品與服務之技術性資料，找出重要的評核項目及其相關權重，在綜合各評核項目之考量因素後，分析出台灣大尺寸TFT-LCD產業經營績效評比排序。
最後，本研究結論，在兩種不同的績效評比方法下，於不同評比觀點面向中分別訂定出TFT-LCD產業經營績效評比評核項目，並以實證分析結果提供業界參考。

關鍵字

資料包絡分析法 ； 層級程序分析法 ； 經營績效評比

並列摘要

TFT-LCD Manufacturing has become one of the most important industries over the past ten years. Observing the great potential of the TFT-LCD market, many large international manufacturers continue to expand their investment and strengthen their operational management to compete for the worldwide market share. This research aims at comparing the operational performance of the five largest TFT-LCD manufacturers in Taiwan – AUO, CMO, HannStar, CPT, and INNOLUX, all of which are producing large size TFT-LCD panels. Two methods are adopted in this study: data envelopment analysis (DEA) and analytical hierarchy process (AHP). Due to limited data, the DEA focuses on comparing these manufacturers’ performance from a financial standpoint. On the other hand, the performance attributes included in AHP can be more generally defined, but the results will depend heavily on the design of the questionnaire, as well as the selected respondents.
In employing DEA, eleven quarters of data from 2006 to 2008 were collected. Window analysis based on both CCR and BBC models is then performed. The results indicate that all manufacturers have not yet reached relative efficiency for all quarters. Further examination shows that their operational efficiency can be improved by reducing the input while keeping the output constant. Ineffective human resource management has led to a 3.63% loss of technical efficiency. Based on the average total efficiency in the window analysis, CMO performs the best. Additional analysis using Malmquist index (MI) shows that the average productivity in the eleven quarters has decreased by 1.2%, mainly due to technical inefficiency as well. Finally, the annual changes of MI indicate that INNOLUX increases in productivity but the others decrease during the whole period.
Other than the input and output items used in the DEA analysis, the major attributes of AHP involve “product and service”. The sub-attributes in “product and service” include technological innovation capability, technological collaboration potential, due-date delivery satisfaction, company reputation, introduction of green supply chain management, prompt reaction to customers’ complaints, and product yield. Our AHP study concludes that AUO is ranked first.

並列關鍵字

Data envelopment analysis ； analytical hierarchy process ； TFT-LCD industry ； operational performance

參考文獻

14.蔡佳惠(2007)，「從管理面建構網球選手績效評估之指標」，大專體育學刊，第九卷，第二期，頁27-38。

11.徐基生、李宗耀、史欽泰、洪志祥、虞孝成與曾國雄(2003)，「運用資料包絡分析法評量工業技術研究院研發組織之研發績效」，管理評論，第二十二卷，第二期，頁25-53。

15.柏謙基(2003)，「由供應鏈觀點探討台灣聚酯加工絲業之採購準則」，國立成功大學管理學院碩士論文。

7. 馮秀鳳(2002)，「半導體產業經營效率之研究：資料包絡分析法之應用」，中原大學會計學系碩士論文。

19.廖麗卿(2005)，「應用AHP 與DEA 模式評選供應商之研究-以戰甲砲車夜視裝備為例」，元智大學工業工程與管理所碩士論文。