Stochastic versus possibilistic programming (Q584073)

A distribution problem of stochastic linear programming is compared with that of finding the possibility distribution of the optimal value for a linear program with fuzzy coefficients. For understanding the arguments one has to follow carefully the evolution in the meaning of the symbols and attributes. For choosing a decision x, the author considers two objective functions to be maximized: The probability/possibility that x is feasible and the probability/possibility that the optimal value exceeds a given target value. Without any details concerning numerical tractability he suggests to use standard techniques of multicriterial optimization to find a compromise solution. The verbal description of a possible decision procedure (as spelled out in conclusions) is misleading.