• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • 1
  • Tagged with
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

多個指數母群體之分類研究

朱靜眉, ZHU,JING-MEI Unknown Date (has links)
在產品研究開發階段, 決策者常面臨的分類問題是在給定一標準( 控制 )設計(Stand -ard Design)及K 個可行設計(Competitive Designs) 下, 如何將此K 個設計分類成 三個部分, 分別為:(1)比控制母體好(better than the standard);(2)與控制母體無 差異(indifference to the standard);(3)比控制母體差(Worse than the standard )。 當產品分析為韋伯分配(Weibull distribution),Patel and Kingston(1982) 在型Ⅱ 截略計劃(Type Ⅱcensoring plan) 下提出一直觀的分類法則(Classification Rule ),此法則之優點是很簡捷, 但所需之樣本數及截略數需要經由模擬求得。Tseng and Wu(1989)提出挑選較佳設計之局部最適的法則(Local Optimal Selection Rule)。本 論文擬修改此挑選法則, 來探討上述分類問題。 本文首先, 提出一演算法(Algorithm) 來計算型Ⅱ抽樣計劃所需之樣本數及截略數。 其次, 以模擬試驗來驗證此法則之穩定性。最后和Patel and Kingston方法作一比較 。

Page generated in 0.0275 seconds