English  |  正體中文  |  简体中文  |  全文筆數/總筆數 : 49378/84106 (59%)
造訪人次 : 7378071      線上人數 : 79
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
搜尋範圍 查詢小技巧:
  • 您可在西文檢索詞彙前後加上"雙引號",以獲取較精準的檢索結果
  • 若欲以作者姓名搜尋,建議至進階搜尋限定作者欄位,可獲得較完整資料
  • 進階搜尋
    請使用永久網址來引用或連結此文件: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/93567


    題名: 生產部門KPI管理之良率指標變化對客訴產生之非線性影響關係探討 : PCB產業F公司大陸廠為例
    其他題名: Nonlinear analysis for the impact of yield rate variation on the custom complain : evidence from the China branch of F company in PCB industry
    作者: 詹詠淇;Chan, Yung Chi
    貢獻者: 淡江大學全球華商經營管理數位學習碩士在職專班
    聶建中;莊孟翰
    關鍵詞: 關鍵績效指標;PCB產業;良率;客訴;平滑移轉迴歸模型;Key Performance Index;PCB Industry;Yield Rate;Compensation Rate of Custom Complain;Smooth transition regression
    日期: 2013
    上傳時間: 2014-01-23 13:11:46 (UTC+8)
    摘要: 本研究採非線性模式,以產品良率作為門檻變數,並以產品良率、離職率、嫁動率三者KPI管理指標作為解釋變數,進行產品良率、離職率、嫁動率與客訴產生間之影響關係探討。研究所使用方法為Granger and Teräsvirta(1993)和Teräsvirta(1994)所提出的「平滑移轉迴歸模型」(Smooth Transition AutoRegression, STAR)非線性模型。
    研究結果,首先發現四筆變數資料均為穩定的I(0)序列。進以實證得知在移轉變數之良率,落後2期時使變數間互動顯著呈現非線性關係;而模型設定得到移轉函數符合羅吉斯模型。最後,在研究以羅吉斯模型為移轉函數,進行各解釋變數對客訴賠償折讓率的影響估計,得到支持研究初始之臆測現象。亦即當產品良率在一定門檻水準以下時,公司經營力求良率之增加,有效助於客訴數量之減少;離職率之增加,則引起公司生產作業流程的瑕疵品增加,進而引致較多的客訴數;生產稼動率的增加,當產品良率在一定門檻水準以下時,生產稼動率即便增加而讓生產部門對客戶商品需求之產量增產,但因良率之不足使瑕疵品明顯產生,而無法滿足客戶之品質所需,進而招致客訴不減反增,因此使得稼動率的增加使客訴數量增加。相反的,當產品良率在一定門檻水準以上時,由於為了提升產品良率,造成公司額外負擔,因之招致另類客訴之產生,使得最終良率增加反使客訴數量增加,然而,客戶在接受並滿足公司產品之交貨品質下之客訴之增加量並不顯著;最後,由於產品良率已達一定水準,客訴對於離職率與稼動率增減變化之敏感度已不明顯。
    本研究以非線性角度出發,相信較能夠捕捉公司經營之真實市場情境。期所得結果,能夠協助企業管理效率之提升,並減少客訴數,更期能提供公司生產部門未來經營之參考依據。
    This research based on non-linear model, adopting product yield rate as threshold variable and using three regressors of product yield rate, dimission rate and utilization rate to investigate the nonlinear effect of all three regressors on the compensation rate of custom complain. The methodology method is Smooth Transition AutoRegression (STAR) elaborated by Granger and Teräsvirta (1993) and Teräsvirta (1994). The empirical result first shows that all four index date are I(0) series. Further results show that, if product yields rate change, after 2 periods (2 months), variables interaction shows a significant non-linear relation and the appropriate transition function is Logistic type from our model specification. The final examination on the effect of all three regressors on the compensation rate of custom complain based on Logistic smooth transition function supports initial research conjecture. When product yield rate is below threshold level, if company seeks to increase yield, it is effectively decrease issue claims quantity; if dimission rate increase, it will cause produce much more defective product, increase issue claim quantity; when increase utilization rate, product yield rate under one level, even meet customer product quantity requirement, but because yield is low cause produce much more defective products, can not meet customer quality requirement, issue claims quantity increase, so utilization rate increase cause issue claims increase. When product yield rate is above one level, still wants to improve products yield cause additional burden, lead to other claims, final increase yield but increase issue claims, but the issue claims increase quantity (customer accept and meet customer quality requirement ) is not significant Finally, when the products yield rate already meet level, issue claims are influenced by dimission rate and utilization rate not obvious.
    This research is in the perspective of non-linear, believe can capture the true marketsituation of company. Hope the research result can assist improves the efficiency of enterprise management, reduce issue claims quantity. Also hope this research can provide a reference for manufacturing department running in the future.
    顯示於類別:[全球華商經營管理數位學習碩士在職專班] 學位論文

    文件中的檔案:

    檔案 大小格式瀏覽次數
    index.html0KbHTML259檢視/開啟

    在機構典藏中所有的資料項目都受到原著作權保護.

    TAIR相關文章

    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - 回饋