A NEW APPROACH TO CALCULATING EFFICIENCY FOR NETWORK DEA

Network Data Envelopment Analysis (Network DEA) enables evaluating decision-making units (DMUs) considering internal processes. However, many NDEA models often fail to identify globally efficient DMUs, hindering benchmarking. This study proposes a novel approach for calculating efficiency, utilizing internal stage efficiency values to construct a global efficiency frontier. In this frontier, DMUs that have the efficiency values ​​of their internal stages that are not dominated by the stages of others DMUs are efficient. A case study demonstrated the approach's efficacy, identifying seven efficient DMUs. The model ensures the existence of efficient DMUs, aligns with the non-parametric nature of DEA, identifies benchmarks based on internal performance similarity, and is flexible, being applicable to various NDEA models.