DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Yum, Bong-Jin | - |
dc.contributor.advisor | 염봉진 | - |
dc.contributor.author | Lee, Pal-Hun | - |
dc.contributor.author | 이팔훈 | - |
dc.date.accessioned | 2011-12-14T02:39:15Z | - |
dc.date.available | 2011-12-14T02:39:15Z | - |
dc.date.issued | 2001 | - |
dc.identifier.uri | http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=165645&flag=dissertation | - |
dc.identifier.uri | http://hdl.handle.net/10203/40515 | - |
dc.description | 학위논문(박사) - 한국과학기술원 : 산업공학과, 2001.2, [ v, 102 p. ] | - |
dc.description.abstract | The objective of parameter design is to make the performance characteristics robust against various noises by determining the optimal levels of product or process design parameters. However, most of the existing parameter design techniques deal with the case of a single performance characteristic, although the case of multiple performance characteristics appears more commonly in product or process design. A difficult problem in dealing with multiple performance characteristics is how to compromise the conflicts among the selected levels of the design parameters for each performance characteristic. This thesis deals with three approaches for resolving such conflicts. Two approaches among them are concerned with developing aggregated performance measures representing the total quality of all the performance characteristics, and the third one is concerned with reducing the dimension of the performance characteristics. For the loss function approach, this thesis first shows that the existing multi-characteristic loss functions have some irrational properties, and then a rational loss function is developed. The properties of the developed loss function are investigated in the light of the basic ideas of parameter design. Finally, a multi-characteristic parameter design (MCPD) technique is developed using the new loss function. Another aggregated performance measure developed is a weighted geometric mean of the desirabilities based on process capability indices. Three types of process capability indices are employed depending on the type of performance characteristics, and a new desirability function (i.e., a logistic function) is proposed based on the consideration that the behavior of a desirability function for a process capability index needs to be S-shaped. The developed approach is relatively easier for engineers to apply since it is not computationally demanding and the process capability indices are familiar to engineers. The third approach developed ... | eng |
dc.language | eng | - |
dc.publisher | 한국과학기술원 | - |
dc.subject | Process Capability Index | - |
dc.subject | Loss Function | - |
dc.subject | Mutiple Performance Characteristics | - |
dc.subject | Parameter Design | - |
dc.subject | Principal Component Analysis | - |
dc.subject | 주성분분석 | - |
dc.subject | 공정능력지수 | - |
dc.subject | 손실함수 | - |
dc.subject | 다성능특성치 | - |
dc.subject | 파라미터 설계 | - |
dc.title | Parameter design in the case of multiple performance characteristics | - |
dc.title.alternative | 다성능특성치에 대한 파라미터 설계 | - |
dc.type | Thesis(Ph.D) | - |
dc.identifier.CNRN | 165645/325007 | - |
dc.description.department | 한국과학기술원 : 산업공학과, | - |
dc.identifier.uid | 000965325 | - |
dc.contributor.localauthor | Yum, Bong-Jin | - |
dc.contributor.localauthor | 염봉진 | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.