pipette
ENEnglish

Boxing Problem

April Niu, Santanu Dey, Oktay G\"unl\"uk, Jisun Lee

Preprint

En palabras de los autores

We study the Boxing Problem, which seeks a subset of points in \(\mathbb{R}^d\) maximizing the total reward of a selection of points minus the width of its axis-aligned bounding box. We also investigate the associated Boxing Set, defined by the epigraph of the width function over binary selection variables. Although the width function is not submodular in general, we restore submodularity by fixing an anchor point. After anchoring, we establish a connection to the intersection of mixing sets with a linking constraint, which yields a polynomial-time optimization algorithm and a polynomial-size extended formulation. We then study the polyhedral structure of the Boxing Set and show that its facets admit laminar families of tight sets and can be obtained by sequential lifting. For the one-dimensional case, we provide a complete description of facet-defining inequalities together with a strongly polynomial-time separation algorithm.

Resultado principalEl resumen no menciona limitaciones.

Apareció: viernes, 30 de octubre. arXiv. Preprint, todavía sin revisión por pares.