Polynomial-time approximation schemes for the euclidean survivable network design problem

Document Type

Conference Proceeding

Publication Date

1-1-2002

Abstract

The survivable network design problem is a classical problem in combinatorial optimization of constructing a minimum-cost subgraph satisfying predetermined connectivity requirements. In this paper we consider its geometric version in which the input is a complete Euclidean graph. We assume that each vertex v has been assigned a connectivity requirement rv. The output subgraph is supposed to have the vertex- (or edge-, respectively) connectivity of at least min{rv, ru} for any pair of vertices v, u. We present the first polynomial-time approximation schemes (PTAS) for basic variants of the survivable network design problem in Euclidean graphs. We first show a PTAS for the Steiner tree problem, which is the survivable network design problem with rv ∈ {0, 1} for any vertex v. Then, we extend it to include the most widely applied case where rv ∈ {0, 1, 2} for any vertex v. Our polynomial-time approximation schemeswork for both vertex- and edge-connectivity requirements in timeO(n log n), where the constants depend on the dimension and the accuracy of approximation. Finally, we observe that our techniques yield also a PTAS for the multigraph variant of the problem where the edge-connectivity requirements satisfy rv ∈ {0, 1,.. ., κ} and κ = O(1). © 2002 Springer-Verlag Berlin Heidelberg.

Identifier

84869152022 (Scopus)

ISBN

[3540438645, 9783540438649]

Publication Title

Lecture Notes in Computer Science Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics

External Full Text Location

https://doi.org/10.1007/3-540-45465-9_83

e-ISSN

16113349

ISSN

03029743

First Page

973

Last Page

984

Volume

2380 LNCS

This document is currently not available here.

Share

COinS