Mi SciELO
Servicios Personalizados
Articulo
Indicadores
- Citado por SciELO
Links relacionados
- Similares en SciELO
Compartir
Revista Cubana de Ciencias Informáticas
versión On-line ISSN 2227-1899
Resumen
SALGUEIRO SICILIA, Yamisleydi; TORO POZO, Jorge L y BELLO PEREZ, Rafael. Performance assessment of MOVMO metaheuristic on constrains test functions. Rev cuba cienc informat [online]. 2016, vol.10, n.1, pp. 182-193. ISSN 2227-1899.
ABSTRACT Classical mathematical programming methods have limitations solving multi-objective optimization problems. These drawbacks are mainly evident in real problems with multiple functions in conflict and complex solutions spaces. That is why the use of meta-heuristics has extended a great deal at present due to its ability to deal with such problems. But as meta-heuristics do not guarantee finding the optimal solution for a problem, new methods are being created either by means of the incorporation of new strategies or by hybridization of the existing ones, to obtain better approximations to Pareto front. This is the case of MOVMO, created by the authors of this work, which is a multi-objective version of VMO meta-heuristic. The objective of this present research was to evaluate the performance of MOVMO on constrains test problems. The experimental studies allowed us to assess the competence of MOVMO in comparison with NSGA-II, SPEA2 and SMPSO methods on ConstrEx, Golinski, Osyczka, Srinivas, Tanaka, and Water functions. Results achieved by MOVMO in Epsilon and Hypervolume quality indicators were higher with significant statistically differences in comparison with those results from other methods in several test functions. These results prove the competitiveness of operators and techniques used in MOVMO on constrains multi-objective optimization problems.
Palabras clave : Multiobjective Metaheuristics; Multiobjective Optimization; Variable Mesh Optimization.