Thematic issue on hybrid nature-inspired algorithms: concepts, analysis and applications

Excerpt

Optimization represents a crucial task in many application domains; in science, engineering, business and social activities, industrial designs, and so on. Usually, it is necessary to find the appropriate values of variables and parameters to improve systems’ performance under different working conditions. Nevertheless, since computational resources, time, and money are always limited in real-world applications, the optimization task could be particularly hard to be accomplished and, as a consequence, it is necessary to identify opportune strategies to “optimize” the optimization processes. In this context, bio-inspired algorithms are assuming an increasingly important role thanks to their capabilities to emulate nature in its organizational complexity and in creating the dynamics, interactions and feedbacks to produce desired outcomes capable of maintaining the right balance in a given ecosystem. Genetic algorithm, particle swarm optimization, ant colony optimization, bee colony optimization are only few examples of bio-inspired optimization algorithms successfully applied to numerous real-world applications. However, these algorithms can suffer from premature convergence due to their strong “exploration” features. During the past few years, a novel research trend is focusing on hybrid nature-inspired algorithms. These innovative strategies are aimed at merging together the benefits provided by different bio-inspired optimization algorithms in order to find the appropriate trade-off between exploration and exploitation features, and identify a breakthrough optimization process in a given application scenario. …