Considering that the probability distribution of random variables in stochastic\nprogramming usually has incomplete information due to a perfect sample\ndata in many real applications, this paper discusses a class of two-stage stochastic\nprogramming problems modeling with maximum minimum expectation\ncompensation criterion (MaxEMin) under the probability distribution\nhaving linear partial information (LPI). In view of the nondifferentiability of\nthis kind of stochastic programming modeling, an improved complex algorithm\nis designed and analyzed. This algorithm can effectively solve the nondifferentiable\nstochastic programming problem under LPI through the variable\npolyhedron iteration. The calculation and discussion of numerical examples\nshow the effectiveness of the proposed algorithm.
Loading....