Insertion heuristics for central cycle problems

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

A central cycle problem requires a cycle that is reasonably short and keeps the maximum distance from any node not on the cycle to its nearest node on the cycle reasonably low. The objective may be to minimize maximum distance or cycle length. Most classes of central cycle problems are NP-hard. This article investigates insertion heuristics for central cycle problems, drawing on insertion heuristics for p centers and travelling salesman tours. It shows that a modified farthest insertion heuristic has reasonable worst-case bounds. It then compares the performance of two farthest insertion heuristics on a range of problems from TSPLIB. It shows that a simple farthest insertion heuristic performs well in practice.
Original languageEnglish
Pages (from-to)70–80
Number of pages11
JournalNetworks
Volume56
Issue number1
Early online date19 Sep 2009
DOIs
Publication statusPublished - Aug 2010

Fingerprint

Computational complexity

Keywords

  • tour
  • cycle center
  • eccentricity
  • cycle-length

Cite this

Insertion heuristics for central cycle problems. / Lamb, John D.

In: Networks, Vol. 56, No. 1, 08.2010, p. 70–80.

Research output: Contribution to journalArticle

Lamb, John D. / Insertion heuristics for central cycle problems. In: Networks. 2010 ; Vol. 56, No. 1. pp. 70–80.
@article{891fce619f874ae49a866434cfe09b6e,
title = "Insertion heuristics for central cycle problems",
abstract = "A central cycle problem requires a cycle that is reasonably short and keeps the maximum distance from any node not on the cycle to its nearest node on the cycle reasonably low. The objective may be to minimize maximum distance or cycle length. Most classes of central cycle problems are NP-hard. This article investigates insertion heuristics for central cycle problems, drawing on insertion heuristics for p centers and travelling salesman tours. It shows that a modified farthest insertion heuristic has reasonable worst-case bounds. It then compares the performance of two farthest insertion heuristics on a range of problems from TSPLIB. It shows that a simple farthest insertion heuristic performs well in practice.",
keywords = "tour, cycle center, eccentricity, cycle-length",
author = "Lamb, {John D}",
year = "2010",
month = "8",
doi = "10.1002/net.20348",
language = "English",
volume = "56",
pages = "70–80",
journal = "Networks",
issn = "0028-3045",
publisher = "Wiley-Liss Inc.",
number = "1",

}

TY - JOUR

T1 - Insertion heuristics for central cycle problems

AU - Lamb, John D

PY - 2010/8

Y1 - 2010/8

N2 - A central cycle problem requires a cycle that is reasonably short and keeps the maximum distance from any node not on the cycle to its nearest node on the cycle reasonably low. The objective may be to minimize maximum distance or cycle length. Most classes of central cycle problems are NP-hard. This article investigates insertion heuristics for central cycle problems, drawing on insertion heuristics for p centers and travelling salesman tours. It shows that a modified farthest insertion heuristic has reasonable worst-case bounds. It then compares the performance of two farthest insertion heuristics on a range of problems from TSPLIB. It shows that a simple farthest insertion heuristic performs well in practice.

AB - A central cycle problem requires a cycle that is reasonably short and keeps the maximum distance from any node not on the cycle to its nearest node on the cycle reasonably low. The objective may be to minimize maximum distance or cycle length. Most classes of central cycle problems are NP-hard. This article investigates insertion heuristics for central cycle problems, drawing on insertion heuristics for p centers and travelling salesman tours. It shows that a modified farthest insertion heuristic has reasonable worst-case bounds. It then compares the performance of two farthest insertion heuristics on a range of problems from TSPLIB. It shows that a simple farthest insertion heuristic performs well in practice.

KW - tour

KW - cycle center

KW - eccentricity

KW - cycle-length

U2 - 10.1002/net.20348

DO - 10.1002/net.20348

M3 - Article

VL - 56

SP - 70

EP - 80

JO - Networks

JF - Networks

SN - 0028-3045

IS - 1

ER -