模糊多準則決策方法評選TFT-LCD黃光製程之濕式機台

Applying Fuzzy Multi-Attribute Decision Making to Select Wet Process Equipment for TFT-LCD Manufacturing

余孟修(Meng-Hsiu Yu)

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(2011年)

https://doi.org/10.6838/YZU.2011.00183

摘要

TFT-LCD產業為台灣近年來最重要的高科技產業之一，業者爲增加市場佔有率及提高獲利無不以提升自身競爭力為目標。爲了實現此目標則必須採用最佳的TFT-LCD黃光製程溼式機台，因為良好的溼式機台將使生產過程更具效率和穩定，因此TFT-LCD生產之黃光製程中最具影響品質與效率為濕式設備。黃光製程溼式機台的單價高、需求數量較少，且替換性低變動難度高因此設備選擇購是非常重要的。

本研究採用模糊多屬性決策方法來解決TFT-LCD黃光製程濕式設備遴選問題，在決策過程中運用了兩個方法：Fuzzy AHP和Fuzzy TOPSIS，其中應用Fuzzy AHP來決定購買設備的各構面屬性與準則之間的相對權重，Fuzzy TOPSIS是用來評價各供應商設備的構面準則績效值。

本研究中，藉由五個部門經理，高級工程師 30人意見，以及相關研究建立雙層的屬性結構。以三個供應商被選作為研究。供應商1是一個韓國公司，供應商2是日本公司，供應商3是一家台灣公司。五大構面分別為：品質，成本，服務，技術和安全；各主要構面包括三到四個子屬性。各部門經理將分別爲各構面和準則訂定權重，並評估各供應商的準則績效。廠長則分別爲五位經理之意見訂定權重。因此，各供應商將分別獲得一個模糊的績效值。本研究運用五種解模糊化方法來排名三個供應商。 其中四種方法評定結果一致，一個方法只能評定出最佳解而無法決定第二與第三。

關鍵字

模糊理論 ； AHP ； TOPSIS ； 黃光製程 ； 濕式設備 ； 供應商遴選

並列摘要

In the past decades, TFT-LCD manufacturing is one of the most important high-tech industries in Taiwan. To increase competitiveness, every individual corporation strives to promote its market share and total profit. Using high quality equipment for the TFT-LCD manufacturing process is crucial in achieving this goal, since good equipment will ensure efficiency and reliability of the production process. The wet process in TFT-LCD manufacturing is an important stage that determines product quality. Selection of wet process equipment is essential, since the equipment is costly, subject to long-term use, and an imperative part of the production line.
In this research, we study a wet process equipment purchase decision problem for a major LCD manufacturing company in Taiwan, and employ fuzzy multi-attribute decision making method to solve the problem. The decision process integrates two methods: fuzzy analytic hierarchy process (FAHP) and fuzzy TOPSIS, where FAHP is applied to determine the relative weights between essential attributes in purchasing equipments, and fuzzy TOPSIS is employed to evaluate each candidate’s equipment performance. In the study, the information obtained from five department managers, 30 senior engineers, and relevant articles are used to establish a two-level attribute hierarchical structure. Three sample suppliers were selected for the case study. Supplier candidate 1 is a Korean company, candidate 2 is a Japanese company, and candidate 3 is a Taiwanese company.
The hierarchical structure has five major attributes (consideration factors): quality, cost, service, techniques, and safety; each major attribute comprises three to four sub-attributes. Each department manager will respectively determine his own weights for major attributes and sub-attributes, and evaluate the performance of each candidate supplier. The factory director is requested to assign his weighted values on the five managers’ opinions. As a consequence, each candidate supplier will receive a fuzzy score for his overall equipment performance evaluation. Five defuzzification methods are used to rank three candidates. Four of the methods conclude the same rank, and one method can only conclude rank 1 candidate, but is inconclusive in ranking the remaining two candidates.

並列關鍵字

TFT-LCD manufacturing ； wet process equipment ； fuzzy AHP ； fuzzy TOPSIS ； multi-attribute decision making ； supplier selection

參考文獻

36. 莊宗南、張嘉良「應用模糊理論於港埠地區觀光發展方向之研究」，環境與生態學報，第2卷，第2期，61~89頁，2009

30. 畢威寧，「結合AHP與TOPSIS 法於供應商績效評估之研究」，科學與工業技術期刊，第一卷第一期，73~83頁，2005

29. 王天津，劉素苓，「應用模糊多準則決策於入口網站服務績效之研究」，電子商務研究，第二卷第三期，235~258頁，2004

41. 光電科技工業協進會http://www.pida.org.tw

35. 陳璋玲，「應用模糊層級分析法建構漁港多功能發展之評估準則模式」，公共事務評論第9卷，第1期，2008