Combining lift-and-project and reduce-and-split

Egon Balas, Gérard Cornuéjols, T. Kis, Giacomo Nannicini

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Split cuts constitute a class of cutting planes that has been successfully employed by the majority of branchand- cut solvers for mixed-integer linear programs. Given a basis of the linear programming (LP) relaxation and a split disjunction, the corresponding split cut can be computed with a closed-form expression. In this paper, we use the lift-and-project framework introduced by Balas and Perregaard to provide the basis, and the reduceand- split algorithm as described by Cornuéjols and Nannicini to compute the split disjunction. We propose a cut generation algorithm that starts from a Gomory mixed-integer cut and alternates between lift-and-project and reduce-and-split in order to strengthen it. This paper has two main contributions. First, we extend the Balas and Perregaard procedure for strengthening cuts arising from split disjunctions involving one variable to split disjunctions on multiple variables. Second, we apply the reduce-and-split algorithm to nonoptimal bases of the LP relaxation. We provide detailed computational testing of the proposed methods.

Original languageEnglish
Pages (from-to)475-487
Number of pages13
JournalINFORMS Journal on Computing
Volume25
Issue number3
DOIs
Publication statusPublished - Jun 2013

Fingerprint

Linear programming
Testing
Integer
Linear program
Cutting planes

Keywords

  • Branch-and-cut
  • Computational analysis
  • Integer
  • Lift-and-project
  • Programming

ASJC Scopus subject areas

  • Software
  • Information Systems
  • Computer Science Applications
  • Management Science and Operations Research

Cite this

Combining lift-and-project and reduce-and-split. / Balas, Egon; Cornuéjols, Gérard; Kis, T.; Nannicini, Giacomo.

In: INFORMS Journal on Computing, Vol. 25, No. 3, 06.2013, p. 475-487.

Research output: Contribution to journalArticle

Balas, E, Cornuéjols, G, Kis, T & Nannicini, G 2013, 'Combining lift-and-project and reduce-and-split', INFORMS Journal on Computing, vol. 25, no. 3, pp. 475-487. https://doi.org/10.1287/ijoc.1120.0515
Balas, Egon ; Cornuéjols, Gérard ; Kis, T. ; Nannicini, Giacomo. / Combining lift-and-project and reduce-and-split. In: INFORMS Journal on Computing. 2013 ; Vol. 25, No. 3. pp. 475-487.
@article{a0eae69369914fc9877bb3e1017fa50c,
title = "Combining lift-and-project and reduce-and-split",
abstract = "Split cuts constitute a class of cutting planes that has been successfully employed by the majority of branchand- cut solvers for mixed-integer linear programs. Given a basis of the linear programming (LP) relaxation and a split disjunction, the corresponding split cut can be computed with a closed-form expression. In this paper, we use the lift-and-project framework introduced by Balas and Perregaard to provide the basis, and the reduceand- split algorithm as described by Cornu{\'e}jols and Nannicini to compute the split disjunction. We propose a cut generation algorithm that starts from a Gomory mixed-integer cut and alternates between lift-and-project and reduce-and-split in order to strengthen it. This paper has two main contributions. First, we extend the Balas and Perregaard procedure for strengthening cuts arising from split disjunctions involving one variable to split disjunctions on multiple variables. Second, we apply the reduce-and-split algorithm to nonoptimal bases of the LP relaxation. We provide detailed computational testing of the proposed methods.",
keywords = "Branch-and-cut, Computational analysis, Integer, Lift-and-project, Programming",
author = "Egon Balas and G{\'e}rard Cornu{\'e}jols and T. Kis and Giacomo Nannicini",
year = "2013",
month = "6",
doi = "10.1287/ijoc.1120.0515",
language = "English",
volume = "25",
pages = "475--487",
journal = "INFORMS Journal on Computing",
issn = "1091-9856",
publisher = "INFORMS Inst.for Operations Res.and the Management Sciences",
number = "3",

}

TY - JOUR

T1 - Combining lift-and-project and reduce-and-split

AU - Balas, Egon

AU - Cornuéjols, Gérard

AU - Kis, T.

AU - Nannicini, Giacomo

PY - 2013/6

Y1 - 2013/6

N2 - Split cuts constitute a class of cutting planes that has been successfully employed by the majority of branchand- cut solvers for mixed-integer linear programs. Given a basis of the linear programming (LP) relaxation and a split disjunction, the corresponding split cut can be computed with a closed-form expression. In this paper, we use the lift-and-project framework introduced by Balas and Perregaard to provide the basis, and the reduceand- split algorithm as described by Cornuéjols and Nannicini to compute the split disjunction. We propose a cut generation algorithm that starts from a Gomory mixed-integer cut and alternates between lift-and-project and reduce-and-split in order to strengthen it. This paper has two main contributions. First, we extend the Balas and Perregaard procedure for strengthening cuts arising from split disjunctions involving one variable to split disjunctions on multiple variables. Second, we apply the reduce-and-split algorithm to nonoptimal bases of the LP relaxation. We provide detailed computational testing of the proposed methods.

AB - Split cuts constitute a class of cutting planes that has been successfully employed by the majority of branchand- cut solvers for mixed-integer linear programs. Given a basis of the linear programming (LP) relaxation and a split disjunction, the corresponding split cut can be computed with a closed-form expression. In this paper, we use the lift-and-project framework introduced by Balas and Perregaard to provide the basis, and the reduceand- split algorithm as described by Cornuéjols and Nannicini to compute the split disjunction. We propose a cut generation algorithm that starts from a Gomory mixed-integer cut and alternates between lift-and-project and reduce-and-split in order to strengthen it. This paper has two main contributions. First, we extend the Balas and Perregaard procedure for strengthening cuts arising from split disjunctions involving one variable to split disjunctions on multiple variables. Second, we apply the reduce-and-split algorithm to nonoptimal bases of the LP relaxation. We provide detailed computational testing of the proposed methods.

KW - Branch-and-cut

KW - Computational analysis

KW - Integer

KW - Lift-and-project

KW - Programming

UR - http://www.scopus.com/inward/record.url?scp=84881180512&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84881180512&partnerID=8YFLogxK

U2 - 10.1287/ijoc.1120.0515

DO - 10.1287/ijoc.1120.0515

M3 - Article

AN - SCOPUS:84881180512

VL - 25

SP - 475

EP - 487

JO - INFORMS Journal on Computing

JF - INFORMS Journal on Computing

SN - 1091-9856

IS - 3

ER -