An interactive approach for finding target unit in the output-oriented BCC model in the present of undesirable outputs

Based on information about existing data on the performance of the units and some preliminary assumptions, Data envelopment analysis (DEA) forms an empirical efficient frontier. Once the efficient frontier is determined, inefficient decision making units (DMUs) can improve their performance to reach the efficient frontier by either increasing their current (desirable) output levels or decreasing their current inputs levels. DEA also determines a virtual unit on the efficient surface as a target unit for the inefficient unit by projecting an inefficient DMU radially to the efficient surface that usually does not include a decision makers (DM) preference structure. Moreover, this approach is not applicable when some outputs (undesirable output) need to be decrease to improve the performance. This paper attempts to establish an equivalent relation between multiple objective linear programming (MOLP) and the output-oriented BCC model in the present of undesirable outputs. We will use the method of satisfactory goals for reflecting the DM’s preference in the process of obtaining target unit after transforming the DEA model into MOLP formulation. Finally, a study on measuring the carbon emission performance of the eight world region is presented.

Data envelopment analysis, multiple objective linear programming, undesirable output