Artificial Bee Colony algorithm with Self-Adaptive Mutation: A novel approach for numeric optimization

Abstract

This paper introduces a variant of Artificial Bee Colony algorithm and compares its results with a number of swarm intelligence and population based optimization algorithms. The Artificial Bee Colony (ABC) is an optimization algorithm based on the intelligent food foraging behavior of honey bees. The proposed variant, Artificial Bee Colony with Self-Adaptive Mutation (ABC-SAM) makes attempts to dynamically adapt the mutation step size with which the artificial bees explore the search space. Mutation with small step size produces small variations of existing solutions which is better for exploitations, while large mutation steps are likely to produce large variations that facilitate better explorations of the search space. ABC-SAM fosters both large and small mutation steps as well as adaptively controls the step lengths based on their effectiveness to produce better solutions. ABC-SAM has been evaluated and compared on a number of benchmark functions with the basic ABC algorithm, Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Particle Swarm Inspired Evolutionary Algorithm (PS-EA). Results indicate that the proposed adaptation scheme facilitates more effective mutations and performs better optimization outperforming all other algorithms in comparison.