Abstract:
Data envelopment analysis (DEA) is a technique that uses data to evaluate the relative efficiencies of decision-making units (DMUs). In real life, the collection of crisp data is onerous; some vagueness can occur due to inconvenient data. With vague data, conventional DEA models cannot be used, as DEA is very sensitive to data. To overcome this problem, the fuzzy theory is integrated with DEA. The new slack DEA model (NSM) concerns straight away with input and output slacks. To handle vague or qualitative data, a fuzzy new slack DEA model can be used. In this study, the fuzzy NSM technique with the expected credits approach is used to calculate the scale efficiencies of DMUs under the constant returns to scale (CRS) and variable returns to scale (VRS) assumptions. This approach converts the fuzzy NSM into a crisp linear programming model and provides a single, crisp efficiency score for each DMU. To illustrate the proposed fuzzy NSM technique with the expected credits approach, the scale efficiency of the Indian oil refineries is measured.