Multivariate Statistical Methods in Multi-Criteria Decision-Making

• Activity Number: 373
• Type: Contributed
• Date/Time: Tuesday, August 11, 2015 : 10:30 AM to 12:20 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Fassil Nebebe* and Tak Kwan Mak
• Companies: Concordia University and Concordia University
• Keywords: analytic hierarchy ; data envelopment ; dimensionality reduction ; multi-attribute rating ; multi-attribute rating

Abstract

We address in this study the issue of multi-criteria decision making involving a large number of attributes in the decision process. This problem typically arises, among other applications, in human resources recruitment and supplier selection in Supply Chain management. As many as over a hundred selection and supplier performance criteria have been developed for consideration by the manager in this complex decision process. Various methods, such as the analytic hierarchy process, data envelopment analysis, simple multi-attribute rating technique and grey relational analysis and their ramifications have been proposed in the literature. These methods may be automatic without the intervention of the purchasing manager, or rely on subjective weightings of the various selection criteria, which may be complex and less robust. Relatively less is being dedicated to the applications of multivariate statistical methods to tackle this important issue. We examine the uses of dimensionality reduction methods which allow the manger to focus on a few dimensions and examine the roles of the remaining uniqueness factors that may have little practical significances on supplier evaluation.