Supplier Selection Based on Optimistic and Pessimistic Perspectives

Discretionary models for supplier efficiency evaluation assume that all criteria are controllable, i.e. they are controlled by the supplier and can change at the discretion of the management. These models do not consider supplier selection in circumstances where some factors are non-discretionary. In this paper, we present a data envelopment analysis (DEA) approach for selection of the best suppliers in the presence of non-discretionary factors and imprecise data that considers both optimistic and pessimistic perspectives. The optimistic perspective evaluates suppliers using the optimistic efficiency and can be used for identifying optimistic efficient suppliers, whereas the pessimistic perspective evaluates suppliers using the pessimistic efficiency and can be used for identifying pessimistic inefficient suppliers. These distinct efficiencies are integrated in the form of geometric average efficiencies that provide an assessment of the overall performance of each supplier in multiple criteria compared to other rival suppliers in the same market and therefore can be used as a basis for comparison of suppliers. We show that geometric average efficiencies have more discriminative power than either of the two efficiencies. A numerical example shows the advantages, capabilities, and applications of the proposed DEA approach.