Engineering Applications of Artificial Intelligence 82 (2019) 148–174
Contents lists available at ScienceDirect
Engineering Applications of Artificial Intelligence
journal homepage: www.elsevier.com/locate/engappai
STOA: A bio-inspired based optimization algorithm for industrial
engineering problems
✩
Gaurav Dhiman, Amandeep Kaur
∗
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
Keywords:
Optimization
Bio-inspired metaheuristic techniques
Constrained problems
Benchmark test problems
A B S T R A C T
This paper presents a bio-inspired algorithm called Sooty Tern Optimization Algorithm (STOA) for solving
constrained industrial problems. The main inspiration of this algorithm is the migration and attacking behaviors
of sea bird sooty tern in nature. These two steps are implemented and mathematically modeled to emphasize
exploitation and exploration in a given search space. The proposed algorithm is compared with nine well-
known bio-inspired algorithms over 44 benchmark test functions. The analysis of convergence behaviors and
computational complexity of the proposed algorithm have been evaluated. Furthermore, to demonstrate its
applicability it is then employed to solve six constrained industrial applications. The outcomes of experiment
reveal that the proposed algorithm is able to solve challenging constrained problems and is very competitive
compared with other optimization algorithms.
1. Introduction
Optimization is the process of defining the decision variables of
a function to minimize or maximize its values (Dhiman and Kumar,
2018a). Most of the real world problems (Chandrawat et al., 2017;
Singh and Dhiman, 2017; Kaur and Dhiman, 2019; Singh and Dhiman,
2018b; Singh et al., 2018b,a; Kaur et al., 2018; Dhiman et al., 2018,
2019; Dhiman and Kumar, 2018) have high computational cost, non-
linear constraints, non-convex, and huge amount of solution spaces and
are complicated. Therefore, solving such problems with large number
of variables and constraints is very tedious and complex (Spears et al.,
2012). Secondly, there are many local optimum solutions that do not
guarantee the best overall solution using classical numerical methods.
To overcome these problems, metaheuristic optimization algorithms
(Dhiman and Kaur, 2019; Dhiman and Kumar, 2019b; Dhiman and
Kaur, 2017) are introduced which are capable of solving such complex
problems throughout course of iterations. Recently, researchers are
gaining interest in developing metaheuristic algorithms (Dhiman and
Kumar, 2018c; Singh and Dhiman, 2018a; Dhiman and Kaur, 2018; Dhi-
man and Kumar, 2018b, 2019a) that are computationally inexpensive,
flexible, and simple by nature. Metaheuristics are broadly classified into
two categories namely single solution and population based algorithms.
Algorithms based on a single solution are those in which a solution
is generated randomly and improved until the best result is obtained.
Population-based algorithms are those in which a set of solutions is
✩
No author associated with this paper has disclosed any potential or pertinent conflicts which may be perceived to have impending conflict with this work.
For full disclosure statements refer to https://doi.org/10.1016/j.engappai.2019.03.021.
∗
Corresponding author.
randomly generated in a given search space and the solution values
are updated during the iterations until the best solution is found.
However, algorithms based on a single solution can be trapped in
local optima, thus not allowing the discovery of the global optimum.
This is because, for a given problem, it reforms only one randomly
generated solution. On the other hand, population based algorithms
are able to find the global optimum. Due to this, researchers have been
attracted towards population based algorithms nowadays.
The categorization of population based algorithms is based on the
theory of evolutionary algorithms, logical behavior of physics algo-
rithms, swarm intelligence of particles, and biological behavior of
bio-inspired algorithms. Evolutionary algorithms are inspired by the
evolutionary processes such as reproduction, 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 envi-
ronment. 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 intelligence 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 very easier to
implement than evolutionary algorithms due to the smaller number
of parameters required. The best known algorithms of the swarm in-
telligence technique are Particle Swarm Optimization (PSO) (Kennedy
https://doi.org/10.1016/j.engappai.2019.03.021
Received 18 June 2018; Received in revised form 22 January 2019; Accepted 19 March 2019
Available online xxxx
0952-1976/© 2019 Published by Elsevier Ltd.
