Centralized Resource Allocation in Two-Stage Production Systems: A Network DEA Approach

Abstract

Network data envelopment analysis (NDEA) is recently developed to explore the internal structure of network production systems so that the efficiencies are measured more precisely. Centralized Resource Allocation (CRA) is a method in which all DMUs are projected onto the efficiency frontier through solving just one DEA mode. This paper proposed a centralized network data envelopment analysis model that combines the centralized data envelopment analysis model and network data envelopment analysis to allocate resources among sub-units. The novel proposed non-radial centralized NDEA approach provides improvement of all inputs and outputs in a unified model. A simple numerical example is presented to illustrate the applicability of the approach. An empirical application on the 8 Chinese commercial banks is also provided for illustration and analysis.