Article Preview
TopIntroduction
Population-based swarm intelligence algorithms have been universally allowed and strongly utilized to work in various optimization difficulties. Unlike common single-point based methods such as hill-climbing algorithms, a population-based swarm intelligence algorithm is based on a set of objects which deals with the difficulty by providing knowledge to support and/or compete between themselves (Shi, An optimization algorithm based on brainstorming process, 2011) (Shi, Brain storm optimization algorithm Conference on Swarm Intelligence, 2011).
During the earlier decades, the procedures of meta-heuristic optimization look pretty familiar. Moreover, besides a large number of general studies, there are several statements on the optimization methods within various areas. Here, the author discusses meta-heuristics as a common strategy (A. Kaveh, M. Khayatazad, 2010). This technique is adopted due to its simplicity, flexibility, derivation-free tool, and it’s avoiding local optima (K. Ahmed, B. Al-Khateeb and M. Mahmood, 2018). Some straightforward concepts can be implemented using computers. Moreover, integrity leads to this potential for proposing beyond new checks, hybridization of two or more meta-heuristics, or improvement of recent meta-heuristics. Moreover, integrity leads to a quick and simple ephemeral inference on different specialists and can be used for their difficulties.
Swarm Intelligence is a problem-solving behavior that emerges as a consequence of the multiplicity of connections that make up the whole system, i.e. the swarm, between individual components. The algorithms focused on cooperative behavior in nature, as in social insect colonies, are based on artificial agent swarms and were initially applied to solve the problems of combinatorial optimization. These algorithms are iterative methods of computation which at the cost of longer computation cycles, provide improved solutions. The agents as in swarm are not aware of the global purpose, so the algorithm can be tailored to address multiple issues by changing the rules of local interactions (Obradović, 2018). A metaheuristic is a problem-independent algorithmic structure at a high level that offers a collection of instructions or techniques to create algorithms for heuristic optimization. The concept is often used according to the recommendations expressed in such a context to refer to a problem-specific application of a heuristic optimization algorithm.
The motivation beyond optimization algorithms influenced by nature is very simple: to take advantage of natural processes or structures to solve problems with optimization. Exploitation and exploration are major concerns to certain meta-heuristic swarm intelligence methods. Currently, there are several population-based swarm intelligence methods including particle swarm optimization(PSO), ant colony optimization (ACO) (Kennedy, J. and Eberhart, R., 1995), artificial bee colony algorithm(ABC) (Karaboga, 2005) (Karaboga, D. and Basturk, B., 2007), imperialist competitive algorithm (Esmaeil, A. and Lucas, C., 2007) and brainstorm optimization (Shi, An optimization algorithm based on brainstorming process, 2011) (Shi, Brain storm optimization algorithm Conference on Swarm Intelligence, 2011).