A
More Efficient MOPSO for Optimization
Walid Elloumi
REGIM: Research Group
on Intelligent Machines,
University of Sfax,
ENIS, BP - 1173
Sfax 3038, Tunisia
Email: elloumiwalid@ieee.org
Adel M. Alimi
REGIM: Research Group
on Intelligent Machines,
University of Sfax,
ENIS, BP - 1173
Sfax 3038, Tunisia
Email: Adel.alimi@ieee.org
Abstract—Swarm-inspired optimization has become very pop-
ular in recent years. The multiple criteria nature of most
real world problems has boosted research on multi-objective
algorithms that can tackle such problems effectively, with the
computational burden and colonies. Particle Swarm Optimization
(PSO) and Ant colony Optimization (ACO) have attracted the
interest of researchers due to its simplicity, effectiveness and
efficiency in solving optimization problems. We use the notion
of multi-objective Particle Swarm Optimization (MOPSO) for
few methods; and we find in most of the results; more the
number of the swarm increases more the accuracy of object is
achieved with greater accuracy. Performance of the basic swarm
for small problems with moderate dimensions and searching
space is satisfactory.
Keywords-Swarm intelligence, Particle swarm Optimization,
Multi-objective Optimization
I. INTRODUCTION
Solving Multi-Objective (MO) problems using
Computational Intelligence (CI) techniques such as Genetic
Algorithms (GA), Particle Swarm Optimization (PSO),
Artificial Immune Systems (AIS), are a fast developing field
of research. Similar to other optimization, MO algorithms
using CI techniques.
Problems are examples of a special class of optimization
problems called multi-objective optimization problems
(MOPs). The question is; what is an optimal solution for
a multi-objective problem? In general, it is called a Pareto
optimal solution if there exists no other feasible solution [1].
However, users practically need only one solution from
the set of optimal trade-off solutions. Therefore, solving
MOPs can be seen as the combination of both searching and
decision-making [2].
Evolutionary algorithms [3][4] have emerged as heuristic
and global alternatives with their most striking characteristic
being: using a population for the search in each iteration. This
makes them suitable for solving multi-objective problems.
Today, the rise of evolutionary multi-objective optimization
can be seen by the number of publications produced over time
[5]. It is worthwhile to note that there are several paradigms
that have emerged as alternatives for the conventional EAs,
such as; Particle Swarm Optimization (PSO) [6], Ant Colony
Optimization (ACO) [7], Differential Evolution (DE) [8].
A bibliographical survey on methods reveals that various
numerical optimization techniques have been employed to
approach the problem over the last three decades. Among
these methods, priority list methods [9] are very fast;
however, they are highly heuristic and generate schedules
with relatively higher operation cost.
From the traffic management on a foraging network [10]
to the building of efficient structures, along with the dynamic
task allocation between workers, examples of complex and
sophisticated behaviors are numerous and diverse among
social insects [11].
Swarm intelligence, as a scientific discipline including
research fields such as; swarm optimization or distributed
control in collective robotics, is born from the incredible
abilities of social insects to solve their everyday-life
problems. Their colonies ranging from few animals to
millions of individuals, display fascinating behaviors that
combine efficiency with both flexibility and robustness [12].
We are working toward a model that describes peoples’
thoughts as a social phenomenon. Thinking differs from the
choreographed behaviors of fish and birds in two major ways.
First, thinking takes place in a space of many more than
three dimensions, as we have seen it in graphs or matrices
and high-dimensional analogues of language. Second, when
two minds converge on the same point in cognitive space, we
call it agreement, not collision.
In flocking simulation the important thing to simulate is
coordinated movement of the organisms, whether flocks,
herds, or schools. Some motives for studying such a topic
include the desire to understand biological aspects of social
behavior and the wish to create interesting and lucrative
graphical effects.
This paper is organized as follows: the second section
gives an overview of accomplishments of the social insects