Integrated Value Function Global Optimization Approaches for Two-Stage Stochastic Programs
Antley, Eric M
Schaefer, Andrew J; Hicks, Illya V; Huchette, Joey; Ozaltın, Osman Y
Master of Science
We consider a class of nonnegative two-stage stochastic integer programs with discretely distributed right-hand sides. We develop a global optimization algorithm that exploits the structure of the superadditive dual for mixed-integer programs to search for the optimal solution. We demonstrate that our algorithm is capable of solving instances with large scenario counts to optimality.
Stochastic Programming·Mixed-Intger Programming·Super-additive Duality·Non-convex Optimization