The Overall Efficiency in the Presence of Imprecise Adaptable Measures

Traditional data envelopment analysis (DEA) models usually evaluate the efficiency scores of decision making units (DMUs) with precise data from an optimistic point of view where the status of each measure (i.e. input/output) is certain. However, there are occasions in real world applications that measures can play both input and output roles in an imprecise environment. In the current study, measures with two roles, input and output, are called “adaptable measures”. This paper proposes a DEA-based approach for estimating the performance of DMUs where adaptable and fuzzy data exist. Indeed, efficiency scores are calculated from two aspects, optimistic and pessimistic, when there are adaptable and fuzzy data. Two different efficiency scores are integrated into a geometric average efficiency. Thus, the overall efficiency is calculated and adaptable variables are split into input and output variables in evaluating the efficiency of each DMU. A numerical example is used to illustrate the approach.