The robust multiple criteria group decision aiding model with risk aversion

Multiple criteria group decision aiding (MCGDA) provides a good insight for experts to make decision. However, the previous MCGDA model had only focused on strategic interaction between decision makers (DMs) and ignored the adverse effects of uncertain decision environment and corresponding decision risk. This paper proposes a novel MCGDA model to address these issues to reduce cost uncertainty and decision risk. First, mean-variance (MV) theory is employed to construct multiple criteria group decision aiding model with risk aversion. In addition, a novel robust multiple criteria group decision aiding model with risk aversion is presented to handle the mean estimation error and covariance estimation error of unit compensation cost. Furthermore, the proposed robust model is equivalently transformed into the robust counterpart problem through duality theory. We extend the robust model to three uncertain scenarios to reduce decision bias. Finally, the numerical case study on academic ranking of world universities demonstrates the applicability of the proposed approach. Comparative analysis and sensitivity analysis confirm the good performance of the proposed model in handling risk preference and uncertainty.