Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/2644
標題: 模糊推論於製程能力評估之研究
Application of Fuzzy Inference to Process Capability Assessment
作者: 陳再萬
Chen, Tasi-Wan
關鍵字: 製程能力指標;process capability indices;望大型;望目型品質特性;製程良率;模糊推論;bigger-the-best and norminal-;the-best type quality characteristics;process yield;fuzzy inference
出版社: 機械工程學系
摘要: 
在產品的品質特性中,有些是屬於單邊規格特性(望大型、望小型),而有些則是屬於雙邊規格特性(望目型)。製程能力指標 或 是評估單邊規格製程能力之有效工具,而指標 則是評估雙邊規格製程能力之有效工具,因為這些指標都能充分的反應製程良率,因此被當作評估製程能力之優良指標。然而在實務上,製程平均數與標準差皆無法輕易獲得,以計算這些指標值,因此通常採用統計的檢定方法,以較客觀方法來評估製程能力。由於樣本有抽樣誤差,如此亦將不確定引入,倘若僅以指標估計值判斷製程能力足與不足,則顯得太過尖銳。因此本論文將結合統計檢定和模糊/類神經模糊推論方法,並利用簡明易懂之得分觀念,用以評估製程能力是否達到產品規格的要求。另外本論文也利用 估測值結合模糊推論方法,評估供應商製程能力高低,以從中選出最佳之供應商。同時本論文也提供一套評估程序,並由應用實例中說明,本論文所提供之方法,能有效評估產品的製程能力。

Many industrial products can be characterized as bigger-the-best type, smaller-the-best type or nominal-the-best type. Quality characteristics and process yields of unilateral or bilateral specification products can be evaluated by using process capability indices. Process capability indices , or are utilized to evaluate quality characteristics and process yields of unilateral and bilateral specification products due to the fact that the formulas for these indices are easy to understand and straightforward to apply. Since sample data must be collected in order to calculate these indices a great degree of uncertainty may be introduced into capability assessments due to sampling errors. Thus, a statistic test is frequently used to afford an objective evaluation of process capability. However, the description in which the boundary between meeting and rejecting the quality requirement is too sharp through this way. In this paper, a method to incorporate the fuzzy/neuro-fuzzy inference with process capability indices in the quality characteristics assessments is presented. A concise score concept is used to represent the grade of process capability. In addition, a fuzzy inference method is also proposed in our study to select the best one among the competing suppliers based on an estimated capability index of calculated from sampled data. Presented fuzzy evaluation procedures and illustrated examples show that the proposed method is effective for application and thus supports its feasibility.
URI: http://hdl.handle.net/11455/2644
Appears in Collections:機械工程學系所

Show full item record
 
TAIR Related Article

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.