Fuzzy Expectation-Based Data Envelopment Analysis Model

Based on credibility theory, this paper presents a new class of fuzzy expectation-based data envelopment analysis (DEA) model, in which we maximize the expected value of relative efficiency subject to credibility constraints at given credibility levels. When the fuzzy inputs and fuzzy outputs are trapezoidal fuzzy variables, we transform the credibility constraints to their crisp equivalents and discuss some basic properties of the fuzzy expectation-based DEA model. After that, we deal with the approximation approach of the fuzzy expected value function and design a heuristic algorithm, which integrates the approximation approach, neural network (NN) and genetic algorithm (GA), to solve the proposed DEA model. Finally, a numerical example is given to illustrate the effectiveness of the designed heuristic algorithm.