Sufficient Conditions for Metric Subregularity of Constraint Systems with Applications to Disjunctive and Ortho-Disjunctive Programs

This paper is devoted to the study of metric subregularity constraint qualification for general optimization problems, with emphasis on nonconvex setting. We elaborate notions directional pseudo- and quasi-normality, recently introduced by Bai et al., which combine standard approach via quasi-normality modern tools variational analysis. focus applications disjunctive programs, where (directional) pseudo-normality characterized an extremal condition. This, in turn, yields efficient verify qualification, include, but are not limited to, Robinson’s result polyhedral multifunctions Gfrerer’s second-order sufficient condition subregularity. Finally, we refine our defining new class ortho-disjunctive programs comprises prominent problems such as mathematical complementarity, vanishing or switching constraints.