Artificial Bee Colony Algorithm Performances in Solving Constraint-Based Optimization Problem

Abstract

Swarm intelligence (SI) can be described as the behavior of a collection of self-organized
swarms. There are plenty of examples of such swarms in real life. Such as bird flocks, fish
schools, bee colonies, termite colonies, etc. Researchers in the 1990s were highly attracted to
two approaches of swarm intelligence. Namely, ant colony and bird flocking/fish schooling.
However, in the beginning of the 21st century, researchers have expressed large interest in
honey bee colony swarm intelligence approach. Since features of self-organization can very
clearly be seen in the honey bee colonies. In the next 10 years, several honey bee swarm
algorithms were developed depending on the intelligence behaviors evident in those colonies.
One such algorithm is the artificial bee colony algorithm or ABC is short. This algorithm has
been largely examined and proven in solving real-world problems.
In this paper, ABC has been used to solve A well-known engineering optimization problem
called the tension/compression spring design problem belongs to Single-Objective Constrained
Optimization Problems, has been used to evaluate the effectiveness of recently developed
metaheuristics.
In this paper, artificial bee colony algorithm is used to solve the tension/compression spring
design problem. This problem is a well-known engineering optimization problem that is part
of the single objective-constrained optimization problems. ABC has been used to assess the
effectiveness of newly implemented metaheuristics.
The solution includes providing various initial values for the artificial bee colony algorithm.
This is done in an attempt to find the best initial values for the algorithm. Afterwards, the
minimum, maximum and mean result of the penalized objectives functions (PFit) calculated.