A simulated annealing method based on a specialised evolutionary algorithm

• Published in: Applied soft computing Vol. 12; no. 2; pp. 573 - 588
• Main Authors: García-Martínez, C., Lozano, M., Rodríguez-Díaz, F.J.
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 01.02.2012
• Subjects: Combinatorial optimisation; Evolutionary computation; Hybrid algorithms; Simulated annealing; Hybrid algorithms; Combinatorial optimisation; Simulated annealing; Evolutionary computation
• ISSN: 1568-4946; 1872-9681
• DOI: 10.1016/j.asoc.2011.11.007

[Display omitted] ► We propose a hybrid algorithm for binary combinatorial problems. ► It is an evolutionary algorithm performing like a simulated annealing model. ► We develop experiments on a representative set of problems. ► The model is competitive regarding similar hybrids and state-of-the-art methods. The flexible architecture of evolutionary algorithms allows specialised models to be obtained with the aim of performing as other search methods do, but more satisfactorily. In fact, there exist several evolutionary proposals in the literature that play the role of local search methods. In this paper, we make a step forward presenting a specialised evolutionary approach that carries out a search process equivalent to the one of simulated annealing. An empirical study comparing the new model with classic simulated annealing methods, hybrid algorithms and state-of-the-art optimisers concludes that the new alternative scheme for combining ideas from simulated annealing and evolutionary algorithms introduced by our proposal may outperform this kind of hybrid algorithms, and achieve competitive results with regard to proposals presented in the literature for binary-coded optimisation problems.