An exact method for the biobjective shortest path problem for large-scale road networks

•Recursive exact algorithm that implicitly enumerates all solutions.•Fast method that works well on graphs of up to 1.2 million nodes and 2.8 million arcs.•Consistently outperformed a top-performer benchmark algorithm for real road networks.•Easily extensible to the multiobjective shortest path prob...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:European journal of operational research Ročník 242; číslo 3; s. 788 - 797
Hlavní autoři: Duque, Daniel, Lozano, Leonardo, Medaglia, Andrés L.
Médium: Journal Article
Jazyk:angličtina
Vydáno: Amsterdam Elsevier B.V 01.05.2015
Elsevier Sequoia S.A
Témata:
Algorithms
Biobjective shortest path
Enumeration
Handles
Mathematical functions
Multiobjective combinatorial optimization (MOCO)
Multiobjective shortest path
Network flow problem
Networks
Operational research
Pulse algorithm
Recursive methods
Roads
Roads & highways
Routing
Shortest path algorithms
Shortest-path problems
Studies
Multiobjective combinatorial optimization (MOCO)
Routing
Biobjective shortest path
Multiobjective shortest path
Pulse algorithm
ISSN:0377-2217, 1872-6860
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:•Recursive exact algorithm that implicitly enumerates all solutions.•Fast method that works well on graphs of up to 1.2 million nodes and 2.8 million arcs.•Consistently outperformed a top-performer benchmark algorithm for real road networks.•Easily extensible to the multiobjective shortest path problem with three or more objectives. The Biobjective Shortest Path Problem (BSP) is the problem of finding (one-to-one) paths from a start node to an end node, while simultaneously minimizing two (conflicting) objective functions. We present an exact recursive method based on implicit enumeration that aggressively prunes dominated solutions. Our approach compares favorably against a top-performer algorithm on two large testbeds from the literature and efficiently solves the BSP on large-scale networks with up to 1.2 million nodes and 2.8 million arcs. Additionally, we describe how the algorithm can be extended to handle more than two objectives and prove the concept on networks with up to 10 objectives.
Bibliografie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ObjectType-Article-1
ObjectType-Feature-2
content type line 23
ISSN:0377-2217
1872-6860
DOI:10.1016/j.ejor.2014.11.003