Genetic Algorithms (GAs) are among the most popular techniques to solve multi-objective optimization problems, with NSGA-II being the most well-known algorithm in the field. Although most of multi-objective GAs (MOGAs) use a generational scheme, in the last few years some proposals using a steady-state scheme have been developed. However, studies about the influence of using those selection strategies in MOGAs are scarce. In this chapter we implement a steady-state version of NSGA-II, which is a generational MOGA, and we compare the two versions with a set of four state-of-the-art multi-objective metaheuristics (SPEA2, OMOPSO, AbYSS, and MOCell) attending to two criteria: the quality of the resulting approximation sets to the Pareto front and the convergence speed of the algorithms. The obtained results show that search capabilities of the steady-state version of NSGA-II significantly improves the original version, providing very competitive results in terms of the quality of the obtained Pareto front approximations and the convergence speed. © 2009 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Nebro, A. J., & Durillo, J. J. (2009). On the effect of applying a steady-state selection scheme in the multi-objective genetic algorithm NSGA-II. Studies in Computational Intelligence, 193, 435–456. https://doi.org/10.1007/978-3-642-00267-0_16
Mendeley helps you to discover research relevant for your work.