Multi-objective evolutionary algorithms for a reliability location problem

•Two new multiobjective evolutionary algorithms have been designed to solve the RFLP.•The Pareto frontiers given by the metaheuristics have been compared with the Pareto-optimal frontier.•A computational study demonstrates the outstanding efficiency of the proposed techniques. Some location problems...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:European journal of operational research Jg. 283; H. 1; S. 83 - 93
Hauptverfasser: Alcaraz, Javier, Landete, Mercedes, Monge, Juan F., Sainz-Pardo, José L.
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Elsevier B.V 16.05.2020
Schlagworte:
Location
Metaheuristics
Multi-objective evolutionary algorithms
Multi-objective location problems
Pareto frontier
Reliability models
Multi-objective location problems
Pareto frontier
Reliability models
Multi-objective evolutionary algorithms
Metaheuristics
Location
ISSN:0377-2217, 1872-6860
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:•Two new multiobjective evolutionary algorithms have been designed to solve the RFLP.•The Pareto frontiers given by the metaheuristics have been compared with the Pareto-optimal frontier.•A computational study demonstrates the outstanding efficiency of the proposed techniques. Some location problems with unreliable facilities present two different objectives, one consisting of minimizing the opening and transportation costs if none of the facilities fail and another consisting of minimizing the expected transportation costs. Usually, these different targets are combined in a single objective function and the decision maker can obtain some different solutions weighting both objectives. However, if the decision maker prefers to obtain a diverse set of non-dominated optimal solutions, then such procedure would not be effective. We have designed and implemented two multi-objective evolutionary algorithms for the realibility fixed-charge location problem by exploiting the peculiarities of this problem in order to obtain sets of solutions that are properly distributed along the Pareto-optimal frontier. The computational results demonstrate the outstanding efficiency of the proposed algorithms, although they present clear differences.
ISSN:0377-2217
1872-6860
DOI:10.1016/j.ejor.2019.10.043