A comparative study of cuckoo search and flower pollination algorithm on solving global optimization problems
| Date | 20 November 2017 |
| Pages | 588-601 |
| DOI | https://doi.org/10.1108/LHT-04-2017-0077 |
| Published date | 20 November 2017 |
| Author | Mohamed Abdel-Basset,Laila A. Shawky,Arun Kumar Sangaiah |
| Subject Matter | Library & information science,Librarianship/library management,Library technology,Information behaviour & retrieval,Information user studies,Metadata,Information & knowledge management,Information & communications technology,Internet |
A comparative study of cuckoo
search and flower pollination
algorithm on solving global
optimization problems
Mohamed Abdel-Basset
Faculty of Computers and Informatics, Zagazig University, Zagazig, Egypt
Laila A. Shawky
Zagazig University, Zagazig, Egypt, and
Arun Kumar Sangaiah
VIT University, Vellore, India
Abstract
Purpose –The purpose of this paper is to present a comparison between two well-known Lévy-based
meta-heuristics called cuckoo search (CS) and flower pollination algorithm (FPA).
Design/methodology/approach –Both the algorithms (Lévy-based meta-heuristics called CS and Flower
Pollination) are tested on selected benchmarks from CEC 2017. In addition, this study discussed all CS and
FPA comparisons that were included implicitly in other works.
Findings –The experimental results show that CS is superior in global convergence to the optimal solution,
while FPA outperforms CS in terms of time complexity.
Originality/value –This paper compares the working flow and significance of FPA and CS which seems to
have many similarities in order to help the researchers deeply understand the differences between both
algorithms. The experimental results are clearly shown to solve the global optimization problem.
Keywords Optimization, Metaheuristic, Cuckoo search, Flower pollination algorithm,
Lévy-based global search, Local search
Paper type Research paper
1. Introduction
Optimization is the attainment of the available best alternatives that minimize or maximize
the outcome. Thereby, it is a very old science that extends into daily life (Neumaier, 2006).
Nowadays, new trends direct toward computational intelligence for solving various real-life
optimization problems through the simulation of mother nature’s power and harmonization
of its residential creatures. Thus, many algorithms were proposed with different
implementations of the selected phenomena, i.e. the mathematical representation of agents’
stochastic behaviors was done through the formulas that coped with proper parameters and
probability distributions. In addition, they should follow up a general algorithmic
framework that is a trade-off between two major functions of a metaheuristic (Blum and
Roli, 2003; Črepinšek et al., 2013): exploration and exploitation. Exploitation (intensification)
leans toward search through the best solution’s neighbors, while exploration
(diversification) makes sure that the algorithm can navigate new areas of the search
space efficiently, often randomly. If metaheuristic has a balanced search that blend
exploration with exploitation, it can ensure high-quality solutions.
In the Particle Swarm Optimization (Eberhart and Kennedy, 1995), Uniform Distribution
(U(a, b)) was used for updating velocity then the new particle’s position was calculated, while
U(a, b) was used in Wolf Search Algorithm (Tang et al., 2012) for simulating wolfs escaping,
passive preying, and initiative preying behaviors. Also, it was used for mimicking fish
behaviors (swarming, searching, chasing and leaping) in Artificial Fish Swarm Algorithm
Library Hi Tech
Vol. 35 No. 4, 2017
pp. 588-601
© Emerald PublishingLimited
0737-8831
DOI 10.1108/LHT-04-2017-0077
Received 18 April 2017
Revised 18 May 2017
Accepted 3 June 2017
The current issue and full text archive of this journal is available on Emerald Insight at:
www.emeraldinsight.com/0737-8831.htm
588
LHT
35,4
To continue reading
Request your trial