Possibilistic Mean Models for Linear Programming Problems with Discrete Fuzzy Random Variables

This paper considers linear programming problems where objective functions involve fuzzy random variables. New decision making models, called possibilistic mean model, are proposed in order to maximize the mean (expectation) of the degrees of possibility and necessity with respect to attained objective function values. It is shown that the original fuzzy random programming problems are transformed into deterministic nonlinear ones which can be solved by conventional nonlinear programming techniques.