Computing Two-Stage Robust Optimization with Mixed Integer Structures

Mixed integer sets have a strong modeling capacity to describe practical systems. Nevertheless, incorporating a mixed integer set often renders an optimization formulation drastically more challenging to compute. In this paper, we study how to effectively solve two-stage robust mixed integer programs built upon decision-dependent uncertainty (DDU). We particularly consider those with a mixed integer DDU set that has a novel modeling power on describing complex dependence between decisions and uncertainty, and those with a mixed integer recourse problem that captures discrete recourse decisions. Note that there has been very limited research on computational algorithms for these challenging problems. To this end, several exact and approximation solution algorithms are presented, which address some critical limitations of existing counterparts (e.g., the relatively complete recourse assumption) and greatly expand our solution capacity. In addition to theoretical analyses on those algorithms' convergence and computational complexities, a set of numerical studies are performed to showcase mixed integer structures' modeling capabilities and to demonstrate those algorithms' computational strength.