Volume-3-Issue-6-June,2025 International Journal of Modern Science and Research Technology

ISSN NO-2584-2706

Integration of the Genetic Algorithm with

Mountain Gazelle Optimizer

Harshita Shivankar; Khushi Patle; Sandhya Dahake

Dept. of MCA, G H Raisoni College of Engineering & Management,
Nagpur, India

Abstract We created a hybrid optimization

This proposed algorithm provides a algorithm by combining the Genetic
promising direction for improving the Algorithm (GA) with the Mountain
performance of nature-inspired optimization Gazelle Optimizer (MGO). The Genetic
algorithms. In this approach, the Genetic Algorithm works by selecting the best
Algorithm is used to address the limitations solutions, mixing them (crossover), and
of the Mountain Gazelle Optimizer, which making small changes (mutation) to
may suffer from premature convergence and improve results. When we added GA to
a lack of exploitation in certain scenarios. By the MGO process, the hybrid algorithm
combining both algorithms, the hybrid became better at both exploring new
approach often finds more accurate solutions solutions and improving existing ones.
to complex optimization problems. We This helps it find good answers more
conducted tests on benchmark functions, and quickly and accurately.
the results show that the hybrid GAME To see how well the hybrid method
algorithm outperforms the standalone GO works, we ran tests using standard
and GA in terms of convergence speed, benchmark problems. The results showed
solution accuracy, and robustness. that the hybrid approach outperforms
using just GA or MGO by themselves. It
Keywords— Genetic Algorithm, was faster, more accurate, and more
Hybridization, multimodal, optimization, reliable, especially for solving complex
meta-heuristic. optimization problems. This study
suggests using the Mountain Gazelle
1. Introduction Optimizer and the Genetic Algorithm
In recent years, there has been a lot of (GA) to get over these restrictions . With
interest in the ability of optimization its selection, crossover, and mutation
algorithms inspired by nature to tackle processes, GA, a well-known
difficult optimization issues [3]. Non- evolutionary algorithm, is excellent at
conventional meta-heuristic algorithms preserving genetic variety. The hybrid
inspired by natural phenomena have been strategy, known as GA-MGO, seeks to
utilized recently to handle a number of increase convergence dependability,
challenging non-linear optimization problems prevent local optima, and improve global
because regular algorithms often fail in search capabilities by integrating GA
certain situations [1]. The great majority of components into the MGO framework.
MAs can be categorized into two main
groups: those that are influenced by
biological processes in nature and those that
are entirely dependent on natural
occurrences.

IJMSRT25JUN013 www.ijmsrt.com 81
DOI: https://doi.org/10.5281/zenodo.15605503
Volume-3-Issue-6-June,2025 International Journal of Modern Science and Research Technology
ISSN NO-2584-2706

This study's goal is to assess how well the situations where the fitness landscape is rough
suggested GA-MGO hybrid algorithm or misleading [3][4].
performs on a collection of popular high-
dimensional benchmark functions. The goal 2.3 Comparative Performance Analysis
of the integration is to improve MGO's To address these challenges, researchers
exploration capabilities without sacrificing its have proposed the integration of the
rate of convergence. To verify the efficacy of Genetic Algorithm (GA) with the
the suggested approach, experimental findings Mountain Gazelle Optimizer, resulting in
are contrasted with those of other conventional a hybrid algorithm known as GA-MGO.
and hybrid optimization algorithms. The incorporation of GA’s evolutionary
operators such as selection, crossover,
2. Literature Review and mutation enhance population
2.1FoundationsofDevelopmentandlgorithms diversity and reinforces the global search
A metaheuristic optimization method called capability of the hybrid model [2].
the Mountain Gazelle Optimizer was presented This integration helps the algorithm
with the goal of increasing convergence speed escape local optima and maintain a
and solution accuracy. The program healthier exploration–exploitation
successfully strikes a balance between balance throughout the optimization
exploration and exploitation since it is process [5].
designed around the dynamic and adaptable Compared to the standard MGO, the
movement patterns of gazelles [5]. By GA-MGO hybrid demonstrates
imitating gazelles' adaptive evasive strategies improved convergence stability, solution
in the face of threats or barriers, MGO aims to accuracy, and robustness when applied
decrease the probability of becoming caught in to high-dimensional and complex
local optima. behavior into the process of optimization problems. Several studies
optimization. By encouraging varied have highlighted that this hybrid
exploration in the early phases of optimization approach outperforms standalone MGO
and fine-tuning in the latter stages, this and other conventional algorithms across
integration greatly improves convergence various benchmark functions.
speed and accuracy [2]. To further improve Algorithms and Authors
MGO's capacity to break out of local optima,
especially in highly multimodal problem Table 1: Algorithm, Authors & Year
environments, spiral dynamics have also been ofpublishing
added. Sr. Algorithm Author Year
No name name

2.2 Challenges and Limitations of MGO 1 Sine Cosine Seyedali 2016

Despite its promising results, the MGO Algorithm Mirjalili
algorithm has several drawbacks, especially
2 Equilibrium Abdollah 2020
when dealing with complex multimodal or Optimizer Asghari
high-dimensional optimization problems. The Varzaneh
premature trend of achieving the ideal MGO et al
3 Differential Rainer 1997
solution is one of the major drawbacks, Equation Stom et al
particularly when population diversity
decreases at the start of the search process. For 4 Backtracking P 2013
this reason, the algorithm is trapped in a local Search Civicioglu
Algorithm
optimizer and cannot explore sufficiently 5 Particle James 1995
different and sometimes better regions of the Swarm Kennedy et
search space. It is not very effective as it does Optimization al
6 Slime Moul Mohammed 2020
not provide a mechanism for robust global Algorithm H Saremi
exploration and diversity authority in

IJMSRT25JUN013 www.ijmsrt.com 82
DOI: https://doi.org/10.5281/zenodo.15605503
Volume-3-Issue-6-June,2025 International Journal of Modern Science and Research Technology
ISSN NO-2584-2706

7 Sunflower Osman K 2021

Evolutionary Erol
Optimization
Algorithm
8 Teaching Rao et al 2011
Learning
Based
Optimization

Fig 1: Classification diagram for Integration of the

Genetic Algorithm with Mountain Gazelle Optimizer

3. Methodology
Actual Value
The Actual Value represents the original or true
value of a solution, reflecting the outcome from a
real-world scenario you’re trying to optimize.
Hybrid Value
On the other hand, the Hybrid Value is the result
that the Mountain Gazelle Optimizer Produces
after running the algorithm to find an optimal.

Table 2 : Standard UM benchmark function

fig 2: Flowchart

IJMSRT25JUN013 www.ijmsrt.com 83
DOI: https://doi.org/10.5281/zenodo.15605503
Volume-3-Issue-6-June,2025 International Journal of Modern Science and Research Technology
ISSN NO-2584-2706

4. Result
By using existing algorithms, we can apply
new approaches to achieve better
improvements. After applying these
approaches, they provide optimized results
and enhance the algorithm's performance. By
implementing hybridization with a genetic
algorithm, we can further improve efficiency
In the diagram below, we perform the test
function on the existing Mountain Gazelle
Optimizer algorithm. Additionally, we
perform the test function on the integration of
the Genetic Algorithm with Mountain
Gazelle Optimizer.
By comparing both, we can conclude that
the reviewed Mountain Gazelle Optimizer
algorithm provides better and more
optimized results.

Fig: 3 Parameter Spaces for the

Function 1 to function 23

IJMSRT25JUN013 www.ijmsrt.com 84
DOI: https://doi.org/10.5281/zenodo.15605503
Volume-3-Issue-6-June,2025 International Journal of Modern Science and Research Technology
ISSN NO-2584-2706

While not as good as F8, other functions

including F1, F22, and F23 also showed good
performance with hybrid values around -10.5.
Function Name Actual Value Hybrid Value

F1 1.3746e-75 -10.5364

F2 1.4299e-45 0.015036

F3 1.3693e-09 0.015036

F4 4.304e-26 29.9455

F5 1.2485e-28 28.0437

F6 4.1225e-10 0.0041524

F7 4.1225e-10 0.053953

F8 4.1225e-10 -9429.329

F9 4.1225e-10 21.3584

F10 4.4409e-16 0.061984

F11 0 0.061984

F12 0 7.1728

F13 0 0.004021

F14 0 6.9033
Fig 4: Graphs for the Function 1 to F15 0.00030749 0.00096214
function 2
Table 3: Result of Function 1 to Function F16 0.00030749 -1.0316
23
F17 0.00030749
5. Conclusion
Both Actual and Hybrid Values were used in F18 0.00030749 3.0001
this study to evaluate the performance of 23
test functions; the Hybrid Value showed how F19 -3.8628 -3.8628
well a hybrid optimization algorithm that F20 -3.8628 -3.322
blends local and global search techniques
worked. With the lowest Hybrid Value of - F21 -10.1532 -3.322
9429.329, the results show that Function F8
F22 -10.1532 -10.4029
produced the best optimal solution under a
minimization objective. F23 -10.1532 -10.5364

IJMSRT25JUN013 www.ijmsrt.com 85
DOI: https://doi.org/10.5281/zenodo.15605503
Volume-3-Issue-6-June,2025 International Journal of Modern Science and Research Technology
ISSN NO-2584-2706

Reliable convergence was highlighted by Fatawu Seini Yussif2, Iddrisu Mohammed

functions F19 to F23, which demonstrated Katali3
consistency between Actual and Hybrid [6] W. Y. Lin, “A novel 3D fruit fly
Values. optimization algorithm and its
The best performances in a maximizing setting applications in economics,” Neural
were F4 and F5. Considerably, the results Comput. Appl., 2016, doi:
demonstrate how well the hybrid strategy 10.1007/s00521-015-1942-8.
works to produce excellent results in a variety [7] Y. Cheng, S. Zhao, B. Cheng, S. Hou, Y.
of optimization scenarios. Shi, and J. Chen, “Modeling and
optimization for collaborative business
SummaryofOptimalFunctions process towards IoT applications,” Mob.
(Minimization Focus): Inf. Syst., 2018, doi:
10.1155/2018/9174568.
Function Hybrid Value [8] X. Wang, T. M. Choi, H. Liu, and X. Yue,
“A novel hybrid ant colony optimization
F8 -9429.329
algorithm for emergency transportation
F1 -10.5364 problems during post-disaster scenarios,”
F22 -10.4029
IEEE Trans. Syst. Man, Cybern. Syst.,
2018, doi: 10.1109/TSMC.2016.2606440.
F23 -10.5364 [9] I. E. Grossmann, Global Optimization in
F19 -3.8628 Engineering Design (Nonconvex
Optimization and Its Applications), vol. 9.
F20 -3.322
1996.
F21 -3.322 [10] R. V. Rao and G. G. Waghmare, “A new
optimization algorithm for solving
complex constrained design optimization
problems,” vol. 0273, no. April, 2016,
6. References doi: 10.1080/0305215X.2016.1164855.
[1] Chaotic-Based Mountain Gazelle Optimizer [11] R. Al-Hajj, A. Assi, and F. Batch, “An
for Solving Optimization Problems evolutionary computing approach for
Priteesha Sarangi1 · Prabhujit Mohapatra1 estimating global solar radiation,” in 2016
Received: 17 October 2023 / Accepted: 3 IEEE International Conference on
March 2024 © The Author(s) 2024 Renewable Energy Research and
[2] AnAccurate Metaheuristic Mountain Gazelle
Applications, ICRERA 2016, 2017. doi:
Optimizer for Parameter Estimation of Single- 10.1109/ICRERA.2016.7884553.
and Double-Diode Photovoltaic Cell Models [12] R. Al-Hajj and A. Assi, “Estimating solar
Rabeh Abbassi 1,2,3,* , Salem Saidi 2,4 and irradiance using genetic programming
Manoharan Premkumar 8 1 , Shabana Urooj technique and meteorological records,”
5,* , Bilal Naji Alhasnawi 6 , Mohamad AIMS Energy, 2017, doi:
Alawad 10.3934/energy.2017.5.798.
[3] An improved mountain gazelle optimizer
[13] R. Al-Hajj, A. Assi, and F. Batch, “An
based on chaotic mapand spiral disturbance for evolutionary computing approach for
medical feature selection Ying Li estimating global solar radiation,” in 2016
ID1,2,YanyuGengID1,2*,HuankunSheng1, IEEE International Conference on
[4] Modified Mountain Gazelle Optimizer Based
Renewable Energy Research and
on Logistic Chaotic Mapping and Truncation Applications, ICRERA 2016, 2017. doi:
Selection Abdul-Fatawu Seini Yussif1, Elvis 10.1109/ICRERA.2016.7884553.
Twumasi2, Emmanuel Asuming Frimpong3
[5] Enhancing Mountain Gazelle Optimizer
(MGO) with an Improved F Parameter for
Global Optimization Toufic Seini1, Abdul-

IJMSRT25JUN013 www.ijmsrt.com 86
DOI: https://doi.org/10.5281/zenodo.15605503
Volume-3-Issue-6-June,2025 International Journal of Modern Science and Research Technology
ISSN NO-2584-2706

[14] R. A. Meyers, “Classical and Nonclassical

Optimization Methods Classical and
Nonclassical Optimization Methods 1
Introduction 1 1.1 Local and Global
Optimality 2 1.2 Problem Types 2 1.3
Example Problem: Fitting Laser-induced
Fluorescence Spectra 3 1.4 Criteria for
Optimization 4 1.5 Multicriteria Optimization
4,” Encycl. Anal. Chem., pp. 9678–9689,
2000, [Online]. Available:
https://pdfs.semanticscholar.org/5c5c/908bb00
a54439dcee50ec1ada6b735694a94.pdf
[15] N. Steffan and G. T. Heydt, “Quadratic
programming and related techniques for the
calculation of locational marginal prices in
distribution systems,” in 2012 North American
Power Symposium (NAPS), 2012, pp. 1–6.
doi: 10.1109/NAPS.2012.6336310.
[16] M. Mafarja et al., “Evolutionary Population
Dynamics and Grasshopper Optimization
approaches for feature selection problems,”
Knowledge-Based Syst., vol. 145, pp. 25–45,
2018, doi: 10.1016/j.knosys.2017.12.037.
[17] A. A. Heidari, R. Ali Abbaspour, and A.
Rezaee Jordehi, “An efficient chaotic water
cycle algorithm for optimization tasks,” Neural
Comput. Appl., vol. 28, no. 1, pp. 57–85,
2017, doi: 10.1007/s00521-015-2037-2.

IJMSRT25JUN013 www.ijmsrt.com 87
DOI: https://doi.org/10.5281/zenodo.15605503