Regularization methods for optimization problems with probabilistic constraints

摘要: We analyze nonlinear stochastic optimization problems with probabilistic constraints on inequalities random right hand sides. develop two numerical methods regularization for their solution. The are based first order optimality conditions and successive inner approximations of the feasible set by progressive generation p-efficient points. algorithms yield an optimal solution involving α-concave probability distributions. For arbitrary distributions, solve convex hull problem provide upper lower bounds value nearly solutions. compared numerically to cutting plane methods.