Genetic Algorithms: The Crossover-Mutation Debate
Nuwan I. Senaratna · 2005
Crossover and mutation are two of the most important genetic operators found in genetic algorithms. There has been much debate as to which of these is practically and theoretically more effective. This literature review highlights the principal milestones of this debate. The conclusion we reach is that there is no evidence to show that either operator is better than the other, and that both operators have their own useful qualities. We also highlight some important new trends that we think might influence the debate in the future. This literature review will be particularly useful for those who would like to theoretically study the relative roles of the crossover and mutation operators and for those who practically work with genetic algorithms and would like to learn more about how best to utilize crossover and mutation.