Interval cross-efficiency for ranking decision making units using the stochastic multicriteria acceptability analysis-evidential reasoning approach

Abstract The interval cross-efficiency approach is an effective technique to rank decision making units (DMUs). It somewhat alleviates the selection difficulties between benevolent and aggressive secondary goal models when the cross-efficiency score is not unique. Present studies offer more insight into the exploration of aggregation weights for interval cross-efficiency aggregation, sometimes with little attention given to the rationality of the aggregation methods. When a decision-maker’s preference structure does not satisfy the “additive independence” condition, a new aggregation method is required. This study employs the stochastic multicriteria acceptability analysis-evidential reasoning (SMAA-ER) approach to provide a model for interval cross-efficiency aggregation. First, the interval cross-efficiency of each DMU is calculated using aggressive and benevolent formulations and then transformed into the belief structure. The analytical ER algorithm is then applied to aggregate the belief structure. Finally, reliable ranking results and other rich decision information (e.g., probability of ranking and degree of predominance) are obtained using the SMAA method. An example is presented to verify the effectiveness of the proposed method.