Description and Implementation of Swarm Intelligence Algorithms Using the system Approach

P. V. Matrenin, Postgraduate Student, e-mail: pavel.matrenin@gmail.com, Novosibirsk State Technical University

The article results analysis algorithms Swarm intelligence as a special class of optimization algorithms. It shows fundamental differences between Swarm Intelligence algorithms and other stochastic optimization algorithms. UML-diagrams illustrating the unified approach proposed by author to swarm intelligence algorithms are given.

Swarm Intelligence algorithms should be considered as a special group among population-based optimization algorithms, since they share the same characteristic idea, based on the collective movements of decentralized agents and the indirect information exchange. This distinguishes Swarm Intelligence of evolutionary algorithms that mimics the process of natural selection.

Proposed unified description of Swarm Intelligence algorithms in case of their implementation provides standardization, enhance flexibility and portability of software and increasing the speed of development.

The scheme is applied to describe the Artificial Bee Colony Optimization and the Fish School Search. Analogous descriptions of the Particle Swarm Optimization and Ant Colony Optimization have been given in this journal, N. 12, 2013.

Keywords: artificial bee colony optimization, fish school search, global optimization, probabilistic methods, Swarm Intelligence, system approach

pp. 27–34