Decision models for evaluation and selection of suppliers in the presence of ‎cardinal and ordinal data, weight restrictions, and non-discretionary factors: ‎An approach based on DEA with double frontiers

Selection of suppliers for outsourcing is now one of the most important decisions of the purchasing ‎department. These decisions constitute an important part of the production and logistics ‎management in many firms. On the other hand, suppliers can be evaluated and selected from ‎optimistic and pessimistic perspectives. There is an argument that both points of view must be ‎considered simultaneously, and any approach that considers only one perspective is biased. This ‎paper proposes a new “data envelopment analysis (DEA) with double frontiers” approach for ‎evaluation and selection of suppliers. The DEA with double frontiers approach can identify the best ‎supplier in the presence of weight restrictions, non-discretionary factors, and cardinal and ordinal ‎data. This paper proposes to integrate both efficiencies in the form of a geometric mean efficiency ‎that measures the overall performance of each supplier. It is shown that geometric mean efficiency ‎has more discriminative power than any of the optimistic and pessimistic efficiencies. A numerical ‎example illustrates the application of the proposed method.‎