Robustifying against Event-driven and Attribute-driven Uncertainties

Over five decades have passed since the first wave of robustoptimization studies conducted by Soyster [1] and Falk [2].It is outstanding that real-life applications of robustoptimization are still swept aside; there is much morepotential for investigating the exact nature of uncertaintiesto have intelligent robust models. For this purpose, in thisstudy, we investigate a more refined description of theuncertain events including (1) Event-driven and (2)Attribute-driven. Instead of model-based calibration ofrobustness, we analyze the structural properties of uncertainevents to obtain a more refined description of the uncertaintypolytopes. Hence, we introduce tractable robust models witha decent degree of conservatism and aversion from overprotectioncaused by the classic cardinality-restricteduncertainty polytopes.

Keywords:

Robust optimization , Convex optimization , Uncertainty sets , Uncertainty events