尚未登入 請由此登入
 
 
 










知識分享平台Eshare檢視資訊View
返回前一頁Back
      [檢舉]

標題Title: Constructing a hybrid Kansei engineering system based on multiple affective responses: application to product form design
作者Authors: 楊智傑,Kuang-Hsiung Chen..等
上傳單位Department: 多媒體與電腦娛樂科學系
上傳時間Date: 2010-8-9
上傳者Author: 楊智傑
審核單位Department: 多媒體與電腦娛樂科學系
審核老師Teacher: 楊智傑
檔案類型Categories: 論文Thesis
關鍵詞Keyword: Product form design, Kansei engineering, Support vector regression, Multi-objective genetic algorithm
摘要Abstract: This study proposes an expert system, which is called the hybrid Kansei engineering system (HKES) based on multiple affective responses (MARs), to facilitate the development of product form design. An HKES is consists of two sub-systems, namely the forward Kansei engineering system (FKES) and the backward Kansei engineering system (BKES). The FKES is utilized to generate product alternatives and the BKES is utilized to predict affective response of new product designs. Although the idea of the HKES and similar hybrid systems have been applied in various research fields, such as product design, engineering design, and system optimization etc., most of existing methodologies are limited to search optimal design solutions using single-objective optimization (SOO), instead of multi-objective optimization (MOO). Hence the applicability of the HKES is limited when adapted to real-world applications, such as product form design discussed in this paper. To overcome this shortcoming, this study integrates the methodologies of support vector regression (SVR) and multi-objective genetic algorithm (MOGA) into the scheme of HEKS. The BKES was constructed by training SVR prediction model of every single affective response (SAR) using basic product form samples as input data and the affective response score gathered from the questionnaires as output values. The FKES generates optimal design alternatives based on MOGA-based searching method according to the MARs specified by a product designer as the system supervisor. A case study of mobile phone design was given to demonstrate the analysis results. The proposed HKES based on MARs can also be applied to a wide variety of product design problems, as well as other MOO problems involving with subjective human perceptions.

檔案名稱
FileName
檔案大小
Size
檔案格式
Format
瀏覽次數
Browses
下載次數
Downloads
2010_8_336b4b3a.pdf 313Kb pdf 386 287
文件中檔案:
 

開啟檔案Download
 
 
返回前一頁