On Weights Determination in Ideal Point Multiattribute Decision-Making Model

Abstract

TOPSIS is a commonly used method in multiattribute decision making. With the weight standardization matrix, it ranks the projects through/by calculating the distances of each project to the positive ideal point and to the negative ideal point. In such a method, the key problem is how to decide the weight of each attribute. First, the chapter analyzes the deficiency in former studies about attribute weights determination; and second, the author proposes a weight determination method based on the principal components. This method decides weights of attributes according to their contribution in sample data. So, the influence of subjective factors can be reduced, the deviation between projects choice can be avoided, and the real importance of any attribute can be reflected objectively. This method entrusts great weights to indexes which synthesize much sample information and small weights to indexes which synthesize little sample information. It conforms to the basic meaning of indexes weights.