Polyhedral combinatorics of the K-partitioning problem with representative variables

Research output: Contribution to journalArticlepeer-review

Abstract

The K-partitioning problem consists in partitioning the vertices of a weighted graph in K sets in order to minimize a function related to the edge weights. We introduce a linear mixed integer formulation with edge variables and representative variables. We investigate the polyhedral combinatorics of the problem, study several families of facet-defining inequalities and evaluate their efficiency on the linear relaxation.

Original languageEnglish
Pages (from-to)1-14
Number of pages14
JournalDiscrete Applied Mathematics
Volume211
DOIs
Publication statusPublished - 1 Oct 2016
Externally publishedYes

Keywords

  • Combinatorial optimization
  • Graph partitioning
  • Polyhedral approach

Fingerprint

Dive into the research topics of 'Polyhedral combinatorics of the K-partitioning problem with representative variables'. Together they form a unique fingerprint.

Cite this