Supplier Selection in Volume Discount Environments in the Presence of Both Cardinal and Ordinal Data: A New Approach Based On Double Frontiers DEA

Nowadays, simultaneous consideration of cardinal and ordinal data in supplier selection process is highlighted more than ever as production philosophies such as just-in-time production are being widely used. Many supplier selection optimization models traditionally presume that the average price of respective expenditures is constant. However, the reality is far from this. As a matter of fact, the suppliers often offer volume discounts in order to encourage buyers to place more orders. This paper presents an innovative approach for selecting the best suppliers in volume discount environments and in the presence of both cardinal and ordinal data. The efficiencies of the suppliers under evaluation are, at first, obtained using optimistic and pessimistic views. The optimistic view evaluates each supplier via a set of most favorable weights. The efficiencies measured by the optimistic view are called “optimistic efficiencies”. The pessimistic view evaluates each supplier via a set of most unfavorable weights. The efficiencies measured by this view are called “pessimistic efficiencies”. It is then demonstrated that these two evaluation results contradict one another, and are undoubtedly biased, non-realistic, and unconvincing. To overcome this problem, a new general performance measure is proposed, which is utilized for integrating the measures obtained from the optimistic and pessimistic views. It will be used for identifying the supplier with the best performance. A numerical example will show the application of the proposed method.