DS-SMAA: An Extension of the SMAA Method Based on Evidence Theory

Multicriteria decision-making methods are widely used in different contexts, with weight elicitation being an important step in the decision-making process. In this context, partial elicitation methods stand out, in which only indirect information is provided to explore the space of feasible weights. Among these methods, the Stochastic Multicriteria Acceptability Analysis (SMAA) considers uncertainty in the weights and performances of the alternatives, allowing the use of rankings and intervals of weights as input. However, the literature is still limited regarding the treatment of imprecision and vagueness associated with the decision-maker’s partial judgments. Thus, this work proposes an integration between SMAA and Dempster-Shafer theory, aiming to model such uncertainties. As a result, a new acceptability index is presented based on the combination of the mass functions of Evidence Theory with the traditional SMAA Acceptability Index.