Introductory Review of Swarm Intelligence Techniques

TL;DR

This paper provides an overview of swarm intelligence algorithms, highlighting their natural inspiration, mathematical models, and their role in optimizing complex, multi-dimensional problems with minimal human intervention.

Contribution

It offers a comprehensive review of various swarm intelligence algorithms, detailing their structures and mathematical foundations, which is valuable for researchers and practitioners.

Findings

Swarm intelligence algorithms effectively solve complex optimization problems.

Natural phenomena inspire efficient algorithm design.

Mathematical models underpin the behavior of SI algorithms.

Abstract

With the rapid upliftment of technology, there has emerged a dire need to fine-tune or optimize certain processes, software, models or structures, with utmost accuracy and efficiency. Optimization algorithms are preferred over other methods of optimization through experimentation or simulation, for their generic problem-solving abilities and promising efficacy with the least human intervention. In recent times, the inducement of natural phenomena into algorithm design has immensely triggered the efficiency of optimization process for even complex multi-dimensional, non-continuous, non-differentiable and noisy problem search spaces. This chapter deals with the Swarm intelligence (SI) based algorithms or Swarm Optimization Algorithms, which are a subset of the greater Nature Inspired Optimization Algorithms (NIOAs). Swarm intelligence involves the collective study of individuals and their mutual interactions leading to intelligent behavior of the swarm. The chapter presents various population-based SI algorithms, their fundamental structures along with their mathematical models.