Knowledge-Based Systems 165 (2019) 169–196
Contents lists available at ScienceDirect
Knowledge-Based Systems
journal homepage: www.elsevier.com/locate/knosys
Seagull optimization algorithm: Theory and its applications for
large-scale industrial engineering problems
Gaurav Dhiman
∗
, Vijay Kumar
Computer Science and Engineering Department, Thapar Institute of Engineering & Technology, Patiala 147004, Punjab, India
a r t i c l e i n f o
Article history:
Received 11 May 2018
Received in revised form 15 November 2018
Accepted 18 November 2018
Available online 26 November 2018
Keywords:
Optimization
Bio-inspired metaheuristics
Industrial problems
Benchmark test problems
a b s t r a c t
This paper presents a novel bio-inspired algorithm called Seagull Optimization Algorithm (SOA) for
solving computationally expensive problems. The main inspiration of this algorithm is the migration and
attacking behaviors of a seagull in nature. These behaviors are mathematically modeled and implemented
to emphasize exploration and exploitation in a given search space. The performance of SOA algorithm
is compared with nine well-known metaheuristics on forty-four benchmark test functions. The analysis
of computational complexity and convergence behaviors of the proposed algorithm have been evaluated.
It is then employed to solve seven constrained real-life industrial applications to demonstrate its appli-
cability. Experimental results reveal that the proposed algorithm is able to solve challenging large-scale
constrained problems and is very competitive algorithm as compared with other optimization algorithms.
© 2018 Elsevier B.V. All rights reserved.
1. Introduction
Optimization is a process of determining the decision variables
of a function to minimize or maximize its values. Most of the
real world problems have non-linear constraints, high computa-
tional cost, non-convex, complicated, and large number of solution
spaces [1]. Therefore, solving such problems with large number
of variables and constraints is very tedious and complex [2–6].
Secondly, there are many local optimum solutions that do not
guarantee the best solution using classical numerical methods.
To overcome these problems, metaheuristic optimization algo-
rithms are introduced which are capable of solving such complex
problems [7,8]. Recently, there is a lot of interest in developing
metaheuristic algorithms that are computationally inexpensive,
flexible, and simple by nature [9–12].
Metaheuristics are broadly classified into two categories such
as single solution and population based algorithms [13]. Single
solution based algorithms are in which a solution is randomly
generated and improved until the best result is obtained. Popula-
tion based algorithms are in which a set of solutions is randomly
generated in a given search space and solution values are updated
during course of iterations until the best solution is found.
However, single solution based algorithms may trap into local
optima that preventing us to find global optimum. This is because it
reforms only one solution which is randomly generated for a given
problem. On the other hand, population based algorithms are able
∗
Corresponding author.
E-mail addresses: gaurav.dhiman@thapar.edu, gdhiman0001@gmail.com
(G. Dhiman).
to find the global optimum. Due to this, researchers have attracted
towards population based algorithms nowadays.
The categorization of population based algorithms is done that
is based on the theory of evolutionary algorithms, logical behav-
ior of physics algorithms, swarm intelligence of particles, and
biological behavior of bio-inspired algorithms. Evolutionary algo-
rithms are inspired by the evolutionary processes such as repro-
duction, mutation, recombination, and selection. These algorithms
are based on the survival fitness of candidate in a population (i.e., a
set of solutions) for a given environment [14]. The physics law
based algorithms are inspired by physical processes according to
some physics rules such as gravitational force, electromagnetic
force, inertia force, heating, and cooling of materials. Swarm intel-
ligence based algorithms are inspired by the collective intelligence
of swarms. The collective intelligence is found among colonies of
flocks, ants, and so on.
Generally, swarm intelligence based algorithms are easy to
implement than the evolutionary algorithms due to less number
of parameters required. The well-known swarm intelligence based
algorithms are Particle Swarm Optimization (PSO) [15] and Ant
Colony Optimization [16]. The reason behind the popularity of
these algorithms is that only few parameters are required for fine
tuning.
Every optimization algorithm needs to address the exploration
and exploitation of a search space [17] and maintains a good bal-
ance between exploration and exploitation. The exploration phase
in an algorithm investigates the different promising regions in a
search space whereas exploitation is able to search the optimal
https://doi.org/10.1016/j.knosys.2018.11.024
0950-7051/© 2018 Elsevier B.V. All rights reserved.
评论0