Computational development of a lagrangian dual approach for quadratic networks

Research output: Contribution to journalArticle

18 Citations (Scopus)

Abstract

A Lagrangian dual approach for large‐scale quadratic programs due to Lin and Pang is adapted to the quadratic cost network flow problem. The unconstrained dual problem is solved iteratively by combining conjugate gradient directions with finite line‐search methods. By exploiting the graph structure, matrix and vector operations are streamlined so that the technique can be used for large problems. A computational study of randomly generated networks with up to 500 nodes and 20,000 arcs gives a comparison between combinations of different conjugate direction formulas with three line‐search techniques. The results demonstrate the clear superiority of the Polak‐Ribiere direction formula combined with a variant of a line‐search technique originally developed by Bitran and Hax for convex knapsack problems.

Original languageEnglish (US)
Pages (from-to)469-485
Number of pages17
JournalNetworks
Volume21
Issue number4
DOIs
StatePublished - Jul 1991

Fingerprint

Costs

All Science Journal Classification (ASJC) codes

  • Software
  • Information Systems
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

@article{03c0457abdbf4ac9a78673f53f071067,
title = "Computational development of a lagrangian dual approach for quadratic networks",
abstract = "A Lagrangian dual approach for large‐scale quadratic programs due to Lin and Pang is adapted to the quadratic cost network flow problem. The unconstrained dual problem is solved iteratively by combining conjugate gradient directions with finite line‐search methods. By exploiting the graph structure, matrix and vector operations are streamlined so that the technique can be used for large problems. A computational study of randomly generated networks with up to 500 nodes and 20,000 arcs gives a comparison between combinations of different conjugate direction formulas with three line‐search techniques. The results demonstrate the clear superiority of the Polak‐Ribiere direction formula combined with a variant of a line‐search technique originally developed by Bitran and Hax for convex knapsack problems.",
author = "Ventura, {Jose A.}",
year = "1991",
month = "7",
doi = "10.1002/net.3230210407",
language = "English (US)",
volume = "21",
pages = "469--485",
journal = "Networks",
issn = "0028-3045",
publisher = "Wiley-Liss Inc.",
number = "4",

}

Computational development of a lagrangian dual approach for quadratic networks. / Ventura, Jose A.

In: Networks, Vol. 21, No. 4, 07.1991, p. 469-485.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Computational development of a lagrangian dual approach for quadratic networks

AU - Ventura, Jose A.

PY - 1991/7

Y1 - 1991/7

N2 - A Lagrangian dual approach for large‐scale quadratic programs due to Lin and Pang is adapted to the quadratic cost network flow problem. The unconstrained dual problem is solved iteratively by combining conjugate gradient directions with finite line‐search methods. By exploiting the graph structure, matrix and vector operations are streamlined so that the technique can be used for large problems. A computational study of randomly generated networks with up to 500 nodes and 20,000 arcs gives a comparison between combinations of different conjugate direction formulas with three line‐search techniques. The results demonstrate the clear superiority of the Polak‐Ribiere direction formula combined with a variant of a line‐search technique originally developed by Bitran and Hax for convex knapsack problems.

AB - A Lagrangian dual approach for large‐scale quadratic programs due to Lin and Pang is adapted to the quadratic cost network flow problem. The unconstrained dual problem is solved iteratively by combining conjugate gradient directions with finite line‐search methods. By exploiting the graph structure, matrix and vector operations are streamlined so that the technique can be used for large problems. A computational study of randomly generated networks with up to 500 nodes and 20,000 arcs gives a comparison between combinations of different conjugate direction formulas with three line‐search techniques. The results demonstrate the clear superiority of the Polak‐Ribiere direction formula combined with a variant of a line‐search technique originally developed by Bitran and Hax for convex knapsack problems.

UR - http://www.scopus.com/inward/record.url?scp=84987041036&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84987041036&partnerID=8YFLogxK

U2 - 10.1002/net.3230210407

DO - 10.1002/net.3230210407

M3 - Article

AN - SCOPUS:84987041036

VL - 21

SP - 469

EP - 485

JO - Networks

JF - Networks

SN - 0028-3045

IS - 4

ER -