Abstract
To develop competitive products, we have to select important quality elements and set levels for satisfying customer needs. However, quality elements whose values can not be obtained may exist in the case different kinds of products are included. A method for identifying important quality elements and determining levels to achieve in such a situation has not been proposed. The aim of the study is to propose a method for obtaining a set of quality elements and preference rules that represents customer needs when different kinds of products are competing. In this research, we proposed a method in which neighborhood rough sets are applied to selecting quality elements with missing values and classification algorithms such as decision trees are applied to extracting preference rules.
