A KE-LSA approach for user-centered design

Abstract

In today's highly competitive business environment, most successful companies use `user-centered' design techniques, to create products that meet user needs. Designers have used traditional user-centered design techniques, with some degree of success. However, in today's market, companies need more accurate, more advanced user-centered design techniques, to mass produce customized products for individual users. This study introduces a new KE-LSA approach for user-centered design. Prior techniques used Kansei engineering (KE) for product design. Prior techniques used latent semantic analysis (LSA) for document indexing. The KE-LSA approach uses KE and a KE-LSA semantic space for individually customized product design. A case study is used to describe the KE-LSA approach. The case study uses the KE-LSA approach to match Kansei values for individual users' needs to cell phone designs for each user. The case study also determines accuracy for matching KE-LSA designs to each user's actual choice for a cell phone. In the case study, an iterative KE-LSA approach matched designs to users' choices, with 82.3% overall accuracy. The iterative KE-LSA approach improved overall accuracy 11.4%, compared to a traditional iterative KE-QT1 approach. The KE-LSA approach improves matching accuracy for mass producing individually customized cell phone designs. Therefore, this study improves individually customized product design and product quality.