Measuring the Relative Efficiency inMulti-Component Decision Making Units andits Application to Bank Branches

In many cases of data envelopment analysis (DEA), decision making units (DMUs) can be separated into different components. These DMUs are called multi-component DMUs, and studying them is known as multi-component DEA. In multi-component DEA some inputs are shared among the components of a DMU, and some components involve into producing some outputs of the DMU. In this paper, we survey measuring the relative efficiency in multi-component DEA. It is shown that using common idea for measuring the efficiency of multi-component DMUs, the relative efficiency of an evaluating DMU may be not obtained. Therefore, present paper proposes a new DEA model which can obtain the relative efficiencies of multi-component DMUs. Some facts about the proposed approach are also provided by theorems. Moreover, the proposed DEA model is compared to another approach in literature utilizing a set of data about 19 bank branches.