Ordered Weighted Average optimization in Multiobjective Spanning Tree Problem

•Analysis of Multiobjective Spanning Tree Problems.•Different formulations of the Ordered Median Average.•Polyhedral analysis of formulations.•Extensive computational results comparing formulations and methods. Multiobjective Spanning Tree Problems are studied in this paper. The ordered median objec...

Full description

Saved in:
Bibliographic Details
Published in:European journal of operational research Vol. 260; no. 3; pp. 886 - 903
Main Authors: Fernández, Elena, Pozo, Miguel A., Puerto, Justo, Scozzari, Andrea
Format: Journal Article Publication
Language:English
Published: Elsevier B.V 01.08.2017
Elsevier
Subjects:
05 Combinatorics
05C Graph theory
60 Probability theory and stochastic processes
60C05 Combinatorial probability
Classificació AMS
Combinatorial analysis
Combinatorial optimization
Combinatorial probabilities
Combinatòria
Matemàtica discreta
Matemàtiques i estadística
Multiobjective optimization
Ordered median
Ordered Weighted Average
Probabilitat
Probabilitats
Spanning trees
Àrees temàtiques de la UPC
Combinatorial optimization
Ordered median
Spanning trees
Multiobjective optimization
Ordered Weighted Average
ISSN:0377-2217, 1872-6860
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•Analysis of Multiobjective Spanning Tree Problems.•Different formulations of the Ordered Median Average.•Polyhedral analysis of formulations.•Extensive computational results comparing formulations and methods. Multiobjective Spanning Tree Problems are studied in this paper. The ordered median objective function is used as an averaging operator to aggregate the vector of objective values of feasible solutions. This leads to the Ordered Weighted Average Spanning Tree Problem, a nonlinear combinatorial optimization problem. Different mixed integer linear programs are proposed, based on the most relevant minimum cost spanning tree models in the literature. These formulations are analyzed and several enhancements presented. Their empirical performance is tested over a set of randomly generated benchmark instances. The results of the computational experiments show that the choice of an appropriate formulation allows to solve larger instances with more objectives than those previously solved in the literature.
ISSN:0377-2217
1872-6860
DOI:10.1016/j.ejor.2016.10.016