Konferensartikel

Comparison between Statistical and Lower / Upper Approximations Rough Sets Models for Beer Can Design and Prototype Evaluation

Ricardo Hirata Okamoto
Keisen Consultores, Clavería, Mexico City, Mexico

Tatsuo Nishino
Dept. of Kansei Design, Hiroshima International University, Japan

Mitsuo Nagamachi
User Science Institute, Kyushu University, Siobara, Minami-ku, Fukuoka, Japan

Ladda ner artikelhttp://www.ep.liu.se/ecp_article/index.en.aspx?issue=026;article=045

Ingår i: 10th QMOD Conference. Quality Management and Organiqatinal Development. Our Dreams of Excellence; 18-20 June; 2007 in Helsingborg; Sweden

Linköping Electronic Conference Proceedings 26:45, s.

Visa mer +

Publicerad: 2008-02-15

ISBN:

ISSN: 1650-3686 (tryckt), 1650-3740 (online)

Abstract

Rough sets theory was introduced as mathematical theory to handle uncertain or inconsistent data (Pawlak; 1998). Since it is superior in dealing with linearly inseparable data; it has been used to extract the decision rules in many application areas; and its effectiveness has been shown.

On the other hand; one of the core technologies in Kansei Engineering (KE) is to identify the relational rules between design elements of products and human evaluation data such as sense and feeling (Nagamachi; 1996). Rough set methods have been used to extract decision rules between human Kansei evaluation experimental data set and design elements (Nishino; 2001; 2003). The extracted rules would enable product designer to design the products fitted to the sense of human.

The purpose of this paper is to apply Rough Sets lower / upper approximations for the definition of decision rules for the design elements of beer cans and compare them to the results obtained through statistical analysis of same experimental data set. Data sets from 2 previous studies in Japan and Mexico were used (Hirata; 2004a; 2004b).

Nyckelord

Product design; Rough set theory; Consumer feeling (Kansei); package design; Kansei market segmentation

Referenser

Inga referenser tillgängliga

Citeringar i Crossref