Incorperating Levy flight method on the bees algorithm with pseudo-gradient towards numerical benchmark test functions

Zakaria, Muhammad 'Izzat (2016) Incorperating Levy flight method on the bees algorithm with pseudo-gradient towards numerical benchmark test functions. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

	Text (24 Pages) Incorperating Levy Flight Method On The Bees Algorithm With Pseudo-Gradient Towards Numerical Benchmark Test Functions 24 Pages.pdf - Submitted Version Download (724kB)
	Text (Full Text) Incorperating Levy flight method on the bees algorithm with pseudo-gradient towards numerical benchmark test functions.pdf - Submitted Version Restricted to Repository staff only Download (2MB)

Abstract

The Bees Algorithm (BA) is a new population-based swarm intelligence inspired by foraging behaviour of honey in nature. BA has recognised by many researcher as a robust and efficient optimisation tool in combinatorial and functional in optimisation fields. Nevertheless, the convergence rate of BA to the optimal solution still requires further development and need a mechanism to avoid from getting trapped in local optima. Therefore, this study concentrates on the improvement of BA performance in term of convergence speed to solve numerical benchmark function. The enhancement is accomplished by incorporating the Levy flight distribution in the neighbourhood search on BA with pseudo-gradient method. The pseudo-gradient method is implemented as a guidance strategy for the scout bees. Meanwhile, the Levy flight is proposed to speed-up the searching process of algorithm. The improved algorithm is subjected to several high dimensional numerical benchmark functions. The result of the experiments confirmed that the proposed algorithm is significantly outperformed the other population-based optimisers and the standard version of BA with an average percentage improvement of 71.42% in the term of speed-up the convergence rate to optimum solution.

Actions (login required)

View Item