Designing a Hybrid Approach to Predict the Performance of Decision Making Units Based on Fuzzy Stochastic DEA and PCA

Data Envelopment Analysis is a management method which applied to performance analysis for decision making units. This paper presents a new hybrid approach based on fuzzy stochastic DEA (FSDEA) and principal component analysis (PCA) to predict efficiency of DMUs. Since the proposed model contains the credibility of fuzzy events in the constraints and the expected value of fuzzy variable in objective function, the solution process is complex. Thus, with the assumption that inputs and outputs are mutually independent triangular fuzzy variables, we then transform the FSDEA model to its equivalent deterministic programming. Then we do PCA on the initial predicted efficiency scores which are obtained from the equivalent deterministic programming under various confidence levels. In order to deal with undesirable initial predicted efficiencies, the required principal components have been selected from the generated ones according to the given choice principal. Then, the principal components are treated as outputs into DEA model to obtain the final predicted efficiencies of DMUs. Finally, the proposed hybrid     PCA-FSDEA approach is applied on real exampleand the computational results will be compared with Fuzzy-SBM model which was proposed by Chen et al. (2013).