The method proposed in this dissertation addresses the need to relate product features to customer expectations. This is particularly difficult given the variety of consumer perspectives and the uncertainty in their assessments. Current statistical methods may not relate all of the market research information available to customer-oriented product-development approaches. Rough set-based Kansei Engineering (RSBKE) is an approach for reasoning under uncertainty and deals with imperfect information originating from the imprecision of human assessment. This mathematically powerful approach extracts knowledge from customer survey data and develops product design rules based upon single or multiple subjective impressions (Kansei) from single or multiple users. A two-stage user-oriented product development approach generates market segmentation rules and product design rules for either a single or multiple Kansei(s). RSBKE provides an enhanced means of defining primary customer groupings and automatically generating design rules. Several extensions to target marketing, lead-user identification, and Kano model applications are presented. RSBKE can be extended to the decision attributes of functional customer requirements. The approach presented here is compared to statistical methods. A case study involving a website design was used to illustrate this approach. The results identified distinctive classes of users who had the same perception of a set of websites. The system generated a set of strict design rules for each class.