In this paper, we use the context-dependent method for ranking decision making units (DMUs) with non-discretionary data in data envelopment analysis (DEA). This methodology partitions a set DMUs into the several levels of efficient frontier, then obtains the efficiency of each DMU against of every calculate efficient frontier. We calculate a ranking score for each DMU by this efficiency indexes. Therefore we can rank DMUs by this score. تفاصيل المقالة

DEA is a nonparametric method for calculating the relative efficiency of a DMU that yields to a reference target for an inefficient DMU. However, it is very hard for inefficient DMUs to be efficient by benchmarking a target DMU which has different inputs. Finding appropriate benchmarks based on the similarity of inputs makes it easier for an inefficient DMU to try to be like its target DMUs. But it is rare to discover a target DMU, which is both the most efficient and similar in inputs, in real situation. Therefore, it is necessary to find the most similar and closest real DMU in terms of inputs on the strong efficiency frontier, which has the highest possible output. In this paper, a combination of the Enhanced Russell model and the additive model is proposed as a new model to improve the efficiency of the inefficient DMUs. The proposed model is applied on a dataset of a large Canadian Bank branches. The target introduced by the proposed method is more practical target for the evaluated unit. The inefficient unit can improve its efficiency more easily by this benchmark. تفاصيل المقالة