Feature cluster on adaptation of discrete metaheuristics to continuous optimization

Many metaheuristic methods have been originally proposed for solving hard discrete combinatorial problems. Practical applications in engineering, however, usually involve continuous variables, or combination of continuous and discrete variables. As a consequence, a significant research effort has been made for adapting such metaheuristics to the continuous domains; this was the case, for example, with genetic algorithms (GA), ant colony (AC) algorithms, Tabu search (TS), simulated annealing (SA). The goal of this special issue was to collect state-of-the-art research papers that discussed recent developments in that area and highlighted some general ideas that proved effective for adapting such metaheuristics to continuous domains. This feature issue comprises of 11 papers that can be divided roughly into three groups. The first group is composed of four papers on real-coded genetic algorithms, in the context of continuous optimization. The first paper in this group, by K. Deb and S. Tiwari, describes a generic evolutionary algorithm, called ‘‘Omni-Optimizer,’’ that automatically adapts to the type of problem at hand, thus solving efficiently a wide variety of single-objective or multiobjective optimization problems. The second paper, by C. Garcia-Martinez, M. Lozano, F. Herrera, D. Molina, and A.M. Sanchez, introduces three processes for improving the behaviour of parent-centric real-parameter crossover operators that can be used in real-coded genetic algorithms. The two remaining papers in this group describe real-life applications. The paper by A. Troncoso, J.C. Riquelme, J.S. Aguilar-Ruiz, and J.M. Riquelme Santos describes evolutionary techniques applied to the optimal