Antonio Lanzotti
Department of Aerospace Engineering, University of Naples "Federico II", Italy
Pietro Tarantino
Department of Aerospace Engineering, University of Naples "Federico II", Italy
Giovanna Matrone
Department of Aerospace Engineering, University of Naples "Federico II", Italy
Download article
Published in: 11th QMOD Conference. Quality Management and Organizational Development Attaining Sustainability From Organizational Excellence to SustainAble Excellence; 20-22 August; 2008 in Helsingborg; Sweden
Linköping Electronic Conference Proceedings 33:14, p. 163-173
Published: 2008-12-09
ISBN:
ISSN: 1650-3686 (print), 1650-3740 (online)
Kansei Engineering (KE) is a methodology able to incorporate; systematically and concretely; people’s emotions into product design solution; above all in concept design phase. This work aims at testing the appropriateness and the robustness of statistical methods which are at this moment new in KE applications. In particular; in order to reduce the length and the reliability of the evaluation session in Virtual Reality environment; optimal experimental designs and methods of analysis are suggested.
Methodology: Statistical methods are used in each phase of KE. In this work; we focus our attention on the choice of experimental design for the synthesis phase and on the analysis methods for the model building. In particular; we compare the KE results by using classical fractionated designs; with the efficiency of saturated designs and supersaturated designs. Two methods of analysis are tested: categorical regression analysis (CATReg) and Ordinal Logistic regression (OLR). Moreover; a comparison between the results of ranking and rating procedures are discussed for the saturated design.
Findings: The comparison among the suggested statistical methods is performed through a study on railway seats design in a virtual reality environment. The results of the analysis support the use of Fractional Factorial Design instead of saturated and supersaturated design. The two methods of data analysis give the same results. No evident differences emerge from the comparison of rating and ranking procedures.
Value of paper: This paper propose optimal experimental design selection strategies to reduce the number of product concepts to design; test and evaluate; and data collection analysis strategies in order to improve the appropriateness and the robustness of model building phase at the end of the synthesis phase. If applied faster and more reliable; a KE approach can overcome the distrust of industrial designers toward the methods belong to the emotional design area.
Kansei Engineering; Saturated Design; Nested Experimental Design; Ordinal Logistic Regression; Categorical Regression
Barone; S.; Lombardo; A.; Tarantino; P.; (2008); “Analysis of user needs for re-design a Wheelchair”; in Erto; P.; Statistical Design for Innovation; Springer; Italy.
Barone; S.; Lombardo; A.; Tarantino; P.; (2007); “A Weighted Logistic Regression for Conjoint Analysis and Kansei Engineering”; Quality and Reliability Engineering International; 23; 689-706.
Box; G.E.P.; Hunter J.S.; Hunter W.G.; (2005); Statistics for Experimenters; John Wiley and Sons; New Jersey.
Di Gironimo; G.; Lanzotti; A.; Matrone; G.; Patalano; S.; Renno; F.; (2008) “Modellazione di assiemi ed esperimenti in Realtà Virtuale per migliorare la qualità di nuovi sedili ferroviari”. Proceedings of 20th INGEGRAF Conference; Valencia; Spain; June 04-06.
Kruskal; J. B.; (1965); “Analysis of factorial experiments by estimating monotone transformations of the data”; Journal of the Royal Statistical Society Series B; 27; 251-263.
Lanzotti; A.; Tarantino; P. (2007); “Kansei Engineering Approach for Identifying Total Quality Elements: A Case Study on Train Interior Design”; Proceedings of 10th QMOD Conference; Helsingborg; June 18-20.
Lin; D.J.K.; (1993a); “ A new class of Supersaturated Design”; Technometrics; 35; 28-31.
Lin; D.J.K. (1993b); “Another look at first-order Saturated Designs: The p-efficient Designs”; Technometrics; 35; 284-292.
McCullagh; P.; Nelder; J.A.; (1989); Generalized Linear Models; Chapman and Hall; London.
Meulman; J. J.; (2003); “Prediction and classification in nonlinear data analysis: Something old; something new; something borrowed; something blue”; Psychometrika; 4; 493-517.
