Merging two-stage series network structures: A DEA-based approach

Merging decision-making units (DMUs) is one of the most important issues in data envelopment analysis (DEA). Hitherto, several merging approaches have been presented in DEA; however, none of them can be used in network DEA. Because they do not consider intermediate products of two-stage DMUs (or two-stage processes) in the merging process. To tackle this problem, this study contributes to network DEA by introducing a novel merging approach. In this approach, we first survey the situations of the first and second stages of the candidate two-stage DMUs relative to the efficient frontiers and then obtain the merged two-stage DMU based on these situations. In other words, our proposed approach estimates the appropriate inputs and intermediate products for merging the candidate two-stage DMUs so that the merged two-stage DMU gets its favorable efficiency score. This research also explains the managerial and economic implications of merging two-stage DMUs. Finally, a numerical example and an empirical application to the US commercial banks are provided to show the use of the proposed approach.