A multiple criteria data envelopment analysis approach to a two-stage process

In the conventional DEA models, the operating multi-input and multi-output production and service is assumed to treat as a black-box for evaluating the technical efficiency. However, the focus on the internal structure leads to the deep perception of a process with the aim of identifying the source of inefficiencies. In the relational two-stage DEA models, the intermediate measures that generally play double input and output roles simultaneously are characterized by identical weights. Therefore, the total flexibility characteristic of multipliers in ordinary DEA can be also observed in two-stage DEA models. In this paper, we develop an integrated multiple criteria DEA (MCDEA) model for the two-process structure so as to measure the efficiency of the whole process and its two sub-process. Our method utilizes a common set of weights to reduce the total freedom of multipliers as well as to improve the discriminating power of DEA. We present a case study in banking industry to show the applicability and efficacy of the proposed approach.