A framework for constructing general integer problems with well-determined duality gaps

Anito Joseph, S. I. Gass

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The paper is concerned with constructing general integer programming problems (GIP) with well-determined duality gaps. That is, given an integer solution vector, X*, our problem is to develop a set of integer linear inequalities AX ≤ b and an objective function c such that X* lies within some known objective function distance of the optimal solution of the relaxed linear-programming problem. By well-determined, we mean that on completion an upper bound on the problem duality gap and an integer solution (optimal or best known) are available to the problem developer. Such a procedure can, therefore, be used to develop test problems to support the research effort in the area of general IP.

Original languageEnglish (US)
Pages (from-to)81-94
Number of pages14
JournalEuropean Journal of Operational Research
Volume136
Issue number1
DOIs
StatePublished - Jan 1 2002

Fingerprint

Duality Gap
Integer
Integer programming
Linear programming
Objective function
Optimal Solution
Integer Programming
Test Problems
Completion
Linear Inequalities
Framework
Duality
Upper bound
Optimal solution

Keywords

  • Duality gap
  • General integer programming
  • Linear programming
  • Test problem generation

ASJC Scopus subject areas

  • Modeling and Simulation
  • Management Science and Operations Research
  • Information Systems and Management

Cite this

A framework for constructing general integer problems with well-determined duality gaps. / Joseph, Anito; Gass, S. I.

In: European Journal of Operational Research, Vol. 136, No. 1, 01.01.2002, p. 81-94.

Research output: Contribution to journalArticle

@article{cb55a99d10944d56ac64795344ab7d62,
title = "A framework for constructing general integer problems with well-determined duality gaps",
abstract = "The paper is concerned with constructing general integer programming problems (GIP) with well-determined duality gaps. That is, given an integer solution vector, X*, our problem is to develop a set of integer linear inequalities AX ≤ b and an objective function c such that X* lies within some known objective function distance of the optimal solution of the relaxed linear-programming problem. By well-determined, we mean that on completion an upper bound on the problem duality gap and an integer solution (optimal or best known) are available to the problem developer. Such a procedure can, therefore, be used to develop test problems to support the research effort in the area of general IP.",
keywords = "Duality gap, General integer programming, Linear programming, Test problem generation",
author = "Anito Joseph and Gass, {S. I.}",
year = "2002",
month = "1",
day = "1",
doi = "10.1016/S0377-2217(01)00035-2",
language = "English (US)",
volume = "136",
pages = "81--94",
journal = "European Journal of Operational Research",
issn = "0377-2217",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - A framework for constructing general integer problems with well-determined duality gaps

AU - Joseph, Anito

AU - Gass, S. I.

PY - 2002/1/1

Y1 - 2002/1/1

N2 - The paper is concerned with constructing general integer programming problems (GIP) with well-determined duality gaps. That is, given an integer solution vector, X*, our problem is to develop a set of integer linear inequalities AX ≤ b and an objective function c such that X* lies within some known objective function distance of the optimal solution of the relaxed linear-programming problem. By well-determined, we mean that on completion an upper bound on the problem duality gap and an integer solution (optimal or best known) are available to the problem developer. Such a procedure can, therefore, be used to develop test problems to support the research effort in the area of general IP.

AB - The paper is concerned with constructing general integer programming problems (GIP) with well-determined duality gaps. That is, given an integer solution vector, X*, our problem is to develop a set of integer linear inequalities AX ≤ b and an objective function c such that X* lies within some known objective function distance of the optimal solution of the relaxed linear-programming problem. By well-determined, we mean that on completion an upper bound on the problem duality gap and an integer solution (optimal or best known) are available to the problem developer. Such a procedure can, therefore, be used to develop test problems to support the research effort in the area of general IP.

KW - Duality gap

KW - General integer programming

KW - Linear programming

KW - Test problem generation

UR - http://www.scopus.com/inward/record.url?scp=0036027417&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036027417&partnerID=8YFLogxK

U2 - 10.1016/S0377-2217(01)00035-2

DO - 10.1016/S0377-2217(01)00035-2

M3 - Article

AN - SCOPUS:0036027417

VL - 136

SP - 81

EP - 94

JO - European Journal of Operational Research

JF - European Journal of Operational Research

SN - 0377-2217

IS - 1

ER -