A novel MGDM method based on information granularity under linguistic setting

Abstract

The aim of this paper is to investigate the multiple attribute group decision making(MGDM) problems under linguistic information, in which attribute weights and the expert weights are completely unknown, and the attribute values take the form of linguistic variables. Firstly, an objective method based on information granularity and entropy is proposed for acquiring attribute weights. The expert weights by use of attribute weights and the relative entropy are obtained. Secondly, we utilize the numerical weighting linguistic average operator to aggregate the linguistic variables corresponding to each alternative, and rank the alternatives according to the linguistic information. Finally, an illustrative example is given to verify practicality and effectiveness of the developed approach. © 2013 Springer-Verlag.