### Abstract

Given an undirected and edge-weighted graph G together with a set of ordered vertex-pairs, called st-pairs, we consider the problems of finding an orientation of all edges in G: min-sum orientation is to minimize the sum of the shortest directed distances between all st-pairs; and min-max orientation is to minimize the maximum shortest directed distance among all st-pairs. In this paper, we first show that both problems are strongly NP-hard for planar graphs even if all edge-weights are identical, and that both problems can be solved in polynomial time for cycles. We then consider the problems restricted to cacti, which form a graph class that contains trees and cycles but is a subclass of planar graphs. Then, min-sum orientation is solvable in polynomial time, whereas min-max orientation remains NP-hard even for two st-pairs. However, based on LP-relaxation, we present a polynomial-time 2-approximation algorithm for min-max orientation. Finally, we give a fully polynomial-time approximation scheme (FPTAS) for min-max orientation on cacti if the number of st-pairs is a fixed constant.

Original language | English |
---|---|

Title of host publication | Algorithms and Computation - 20th International Symposium, ISAAC 2009, Proceedings |

Pages | 403-412 |

Number of pages | 10 |

DOIs | |

Publication status | Published - 1 Dec 2009 |

Event | 20th International Symposium on Algorithms and Computation, ISAAC 2009 - Honolulu, HI, United States Duration: 16 Dec 2009 → 18 Dec 2009 |

### Publication series

Name | Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) |
---|---|

Volume | 5878 LNCS |

ISSN (Print) | 0302-9743 |

ISSN (Electronic) | 1611-3349 |

### Conference

Conference | 20th International Symposium on Algorithms and Computation, ISAAC 2009 |
---|---|

Country | United States |

City | Honolulu, HI |

Period | 16/12/09 → 18/12/09 |

## Fingerprint Dive into the research topics of 'Route-enabling graph orientation problems'. Together they form a unique fingerprint.

## Cite this

*Algorithms and Computation - 20th International Symposium, ISAAC 2009, Proceedings*(pp. 403-412). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 5878 LNCS). https://doi.org/10.1007/978-3-642-10631-6_42