Received 11 April 2023, accepted 28 May 2023, date of publication 6 June 2023, date of current version 13 June 2023.

Digital Object Identifier 10.1109/ACCESS.2023.3283422

Red Panda Optimization Algorithm: An Effective

Bio-Inspired Metaheuristic Algorithm for Solving
Engineering Optimization Problems
HADI GIVI 1, (Member, IEEE), MOHAMMAD DEHGHANI 2, AND ŠTĚPÁN HUBÁLOVSKÝ3
1 Department of Electrical Engineering, Shahreza Campus, University of Isfahan, Iran
2 Department of Mathematics, Faculty of Science, University of Hradec Králové, 50003 Hradec Králové, Czech Republic
3 Department of Applied Cybernetics, University of Hradec Králové, 50003 Hradec Králové, Czech Republic

Corresponding author: Hadi Givi ([email protected])

This work was supported by the Excellence project of Faculty of Science, University of Hradec Králové, Czech Republic, 2214/2023-2024.

ABSTRACT This paper presents a new bio-inspired metaheuristic algorithm called Red Panda Optimization
(RPO) that imitates the natural behaviors of red pandas in nature. The main design idea of RPO is derived
from two characteristic natural behaviors of red pandas: (i) foraging strategy, and (ii) climbing trees to rest.
The proposed RPO approach is mathematically modeled in two phases of exploration based on the simulation
of red pandas’ foraging strategy and exploitation based on the simulation of red pandas’ movement in
climbing trees. The main advantage of the proposed approach is that there is no control parameter in
its mathematical modeling, and for this reason, it does not need a parameter adjustment process. The
performance of RPO is evaluated on fifty-two standard benchmark functions including unimodal, high-
dimensional multimodal, and fixed-dimensional multimodal types as well as CEC 2017 test suite. The
optimization results obtained by the proposed RPO approach are compared with the performance of twelve
well-known metaheuristic algorithms. The simulation results show that RPO, by maintaining the balance
between exploration and exploitation, is effective in solving optimization problems and its performance is
superior over competitor algorithms. Based on the analysis of the optimization results, RPO has provided
more successful performance compared to the competitor algorithms in 100% of unimodal functions, 100%
of high-dimensional multimodal functions, 100% of fixed-dimensional multimodal functions, and 86.2% of
CEC 2017 test suite benchmark functions. Also, the statistical analysis of the Wilcoxon rank sum test shows
that the superiority of RPO in the competition with the compared algorithms is significant from a statistical
point of view. In addition, the results of implementing RPO on four engineering design problems confirms
the ability of the proposed approach to handle real-world optimization applications.

INDEX TERMS Optimization, bio-inspired, red panda, metaheuristic, exploration, exploitation.

I. INTRODUCTION constraints, and objective function. The main goal in opti-

Optimization problems are a type of problems that have more mization is to set values for decision variables such that the
than one feasible solution. According to this definition, opti- objective function is optimized according to the constraints of
mization is the process of finding the best feasible solution the problem [2]. There are numerous optimization problems
among the available solutions for a problem [1]. From a in science that have become more complex with the advance-
mathematical point of view, an optimization problem can ment of technology, and this is the reason for the need to
be modeled considering three main parts: decision variables, powerful tools for solving optimization problems [3].
Problem solving methods in optimization studies are
The associate editor coordinating the review of this manuscript and classified into two groups: deterministic and stochastic
approving it for publication was Xiong Luo . approaches [4]. Deterministic approaches are effective tools

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
VOLUME 11, 2023 57203
 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

for solving linear, convex, continuous, differentiable, and not a guarantee for the similar performance of that algorithm
low-dimensional problems [5]. However, in case of more in solving other optimization problems. According to NFL
complex optimization problems, deterministic approaches theorem, a successful algorithm in solving an optimization
lose their efficiency due to getting stuck in local optima. problem may even fail in solving another problem. Therefore,
This is while many of today’s optimization problems and there is no guarantee for the success or failure of implement-
real-world applications are non-linear, non-convex, discon- ing an algorithm on a problem. By keeping open the field
tinuous, non-differentiable, and high-dimensional [6]. These of metaheuristic algorithms study, NFL theorem encourages
disadvantages and the inability of deterministic approaches researchers to provide more effective solutions to optimiza-
to solve complex optimization problems have prompted tion problems by designing newer metaheuristic algorithms.
researchers to develop stochastic approaches. Stochastic The innovation and novelty of this paper is the introduction
approaches, without the need for derivative information and design of a new metaheuristic algorithm called Red Panda
from the objective function and problem constraints, are Optimization (RPO) to solve optimization problems. The
able to provide suitable solutions for optimization problems main contributions of this paper are as follows:
based on the random search process in the problem-solving • The proposed RPO approach is based on the simulation
space [7]. Metaheuristic algorithms are one of the most of red panda behaviors in nature.
effective stochastic approaches in solving optimization prob- • The fundamental inspiration for RPO design is the for-
lems. Advantages such as simplicity of concepts, convenient aging strategy and tree climbing ability of red pandas.
implementation, no dependence on the type of problem, • The mathematical model of RPO is presented in two
no need for derivative information, efficient performance in phases of exploration and exploitation.
solving nonlinear, non-convex, high-dimensional, and NP- • The efficiency of RPO in optimization has been
hard problems, as well as desirable efficiency in nonlinear evaluated on fifty-two benchmark functions consist-
and unknown search spaces are the main reasons for the ing of unimodal, high-dimensional multimodal, and
popularity of metaheuristic algorithms [8]. fixed-dimensional multimodal types, as well as CEC
The nature of random search in metaheuristic algorithms 2017 test suite.
means that there is no guarantee of achieving the global • The performance of RPO is compared with twelve
optimal solution with these approaches. However, since the well-known metaheuristic algorithms.
solutions obtained by metaheuristic algorithms are close • The effectiveness of RPO in handling real-world
to the global optima, they are acceptable and known as applications is examined on four engineering design
quasi-optimal solutions [9]. In order to organize an effective problems.
search process, metaheuristic algorithms must be able to scan
the problem-solving space appropriately at both global and The rest of this article is organized in this way, first the
local levels. Global search with the concept of exploration literature review is presented in section II. Then, the proposed
leads to the ability of the algorithm to comprehensively Red Panda Optimization (RPO) algorithm is introduced and
search the problem-solving space with the aim of discov- mathematically modeled in section III. Simulation studies
ering the main optimal area and preventing the algorithm and results are presented in section IV. The performance of
from getting stuck in local optima. Local search with the the proposed RPO in solving real-world applications is eval-
concept of exploitation leads to the ability of the algorithm uated in section V. Finally, conclusions and several proposals
to achieve possible better solutions near the discovered solu- for future studies are provided in section VI.
tions [10]. In addition to exploration and exploitation abil-
ities, what leads to the success of metaheuristic algorithms II. LECTURE REVIEW
in the optimization process is the balancing of exploration Metaheuristic algorithms have been developed with inspira-
and exploitation during the search process [11]. The efforts of tion from various natural phenomena, animal life in nature,
researchers to achieve more effective solutions for optimiza- biological sciences, physical laws and phenomena, rules of
tion problems have led to the design of numerous metaheuris- games, human interactions, and other evolutionary processes.
tic algorithms [12]. These algorithms are employed in various Based on the idea used in the design, metaheuristic algo-
optimization applications in science, such as energy [13], rithms can be broadly classified into five groups: swarm-
[14], [15], [16], protection [17], energy carriers [18], [19], based, evolutionary-based, physics-based, human-based, and
and electrical engineering [20], [21], [22], [23], [24], [25]. game-based approaches [27].
The main research question in the study of metaheuristic Swarm-based metaheuristic algorithms are developed
algorithms is that considering various algorithms presented based on simulating the natural swarm behavior of ani-
so far, is there still a need to design newer metaheuristic mals, birds, aquatic animals, insects, plants, and other liv-
algorithms? In response to this question, No Free Lunch ing organisms in nature. Among the well-known algorithms
(NFL) theorem [26] explains that there is no specific meta- of this group, one can mention Particle Swarm Optimiza-
heuristic algorithm to be considered as the best optimizer for tion (PSO) [28], Ant Colony Optimization (ACO) [29],
all optimization problems. In fact, the optimal performance Artificial Bee Colony (ABC) [30], and Firefly Algorithm
of an algorithm in solving a set of optimization problems is (FA) [31]. ACO was proposed based on modeling the

57204 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

ability of ant swarm to identify the shortest communi- Algorithm (AOA) [58], Equilibrium Optimizer (EO) [59],
cation path between nests and food sources. ABC was Multi-Verse Optimizer (MVO) [60], and Electro-Magnetism
designed based on simulating interactions and natural behav- Optimization (EMO) [61].
iors of colony bees in obtaining food resources. FA was Human-based metaheuristic algorithms have been devel-
inspired by the communication feature of flashing light oped based on the simulation of human interactions, com-
in the firefly’s swarm. Finding food resources, migration, munication, thinking, and decision-making in social and
and chasing are common natural behaviors among ani- individual lives. Teaching-Learning Based Optimization
mals, whose simulation has inspired researchers to design (TLBO) [62] is one of the most widely used human-based
several swarm-based algorithms such as: Coati Optimiza- algorithms. The basic inspiration in its design was modelling
tion Algorithm (COA) [32], Golden Jackal Optimization the educational interactions of students and teachers in the
(GJO) [33], White Shark Optimizer (WSO) [34], Marine classroom. The economic activities of both the poor and the
Predator Algorithm (MPA) [35], African Vultures Optimiza- rich sections of the society, who are trying to improve their
tion Algorithm (AVOA) [36], Pelican Optimization Algo- economic conditions, have been a source of inspiration in
rithm (POA) [37], Tunicate Swarm Algorithm (TSA) [38], the design of Poor and Rich Optimization (PRO) [63]. Ther-
Honey Badger Algorithm (HBA) [39], Whale Optimiza- apeutic interactions between patients and physicians were
tion Algorithm (WOA) [40], Reptile Search Algorithm employed in the design of Doctor and Patients Optimiza-
(RSA) [41], Green Anaconda Optimization (GAO) [42], tion (DPO) [64]. The cooperation between the members of
Cuckoo Search Algorithm (CSA) [43], and Grey Wolf Opti- a team who are trying to achieve the team goal was the
mizer (GWO) [44]. basic idea in the design of Teamwork Optimization Algorithm
Evolutionary-based metaheuristic algorithms are intro- (TOA) [65]. Some other human-based metaheuristic algo-
duced based on the concepts of genetics, biology, natu- rithms are: Ali Baba and the Forty Thieves (AFT) [66], Skill
ral selection, and survival of the fittest. Genetic Algorithm Optimization Algorithm (SOA) [67], Language Education
(GA) [45] and Differential Evolution (DE) [46] are widely Optimization (LEO) [68], Coronavirus Herd Immunity Opti-
used approaches in this group. GA and DE were developed mizer (CHIO) [69], War Strategy Optimization (WSO) [70],
based on reproductive process modeling, biology concepts and Driving Training-Based Optimization (DTBO) [71].
and stochastic operators such as selection, crossover, and Game-based metaheuristic algorithms have been intro-
mutation. Artificial Immune Systems (AISs) is another evolu- duced based on modeling the rules of various individual and
tionary approach that has been introduced based on the ability group games, the strategy of players, coaches, referees, and
of the human body’s defense system against diseases and other influential persons of the games. Football Game Based
microbes [47]. Some other evolutionary-based metaheuristic Optimization (FGBO) [72] and Volleyball Premier League
algorithms are: Evolution Strategy (ES) [48], Genetic pro- (VPL) [73] are two game-based approaches, which were
gramming (GP) [49], and Cultural Algorithm (CA) [50]. designed based on the simulation of competitions between
Physics-based metaheuristic algorithms are designed clubs in soccer and volleyball leagues. The skill of the play-
based on simulating concepts, phenomena, and laws in ers in putting together the pieces of the puzzle has been a
physics. Simulated Annealing (SA) [51] is one of the most source of inspiration in the design of Puzzle Optimization
widely used physics-based approaches. SA was developed Algorithm (POA) [74]. The strategy of players in throwing
based on the modeling of metal annealing phenomenon darts and collecting points in the darts game was employed
in physics, where metals are melted under heat and then in the design of Darts Game Optimizer (DGO) [75]. Some
cooled in order to achieve ideal crystal. The modeling of other game-based metaheuristic algorithms are: Tug of War
physical forces has been the starting point for the introduc- Optimization (TWO) [76], Billiards Optimization Algorithm
tion of several physics-based algorithms, such as: Spring (BOA) [77], Dice Game Optimization (DGO) [78], Ring
Search Algorithm (SSA) [52], Momentum Search Algorithm Toss Game-Based Optimization (RTGBO) [79], Orienta-
(MSA) [53], and Gravitational Search Algorithm (GSA) [54]. tion Search Algorithm (OSA) [80], and Archery Algorithm
SSA was introduced based on the simulation of Hooke’s (AA) [81].
law, spring elastic force, and Newton’s laws of motion Based on the best knowledge obtained from the litera-
in a system consisting of weights connected by springs. ture review, no metaheuristic algorithm has been designed
MSA was proposed based on the modeling of the force based on simulating the natural behavior of the red panda.
resulting from the momentum between the bullets. GSA Meanwhile, the behavior of foraging and resting on trees
was designed based on simulating the gravitational force among red pandas are intelligent activities that has the
that masses exert on each other at different distances. The potential to design a metaheuristic algorithm. In order to
physical phenomenon of matter state transitions for water address this research gap in the studies of metaheuristic
was employed in design of the Water Cycle Algorithm algorithms, in this paper, a new metaheuristic algorithm is
(WCA) [55]. Some other physics-based metaheuristic algo- introduced based on the mathematical modeling of the natural
rithms are: Nuclear Reaction Optimization (NRO) [56], behavior of the red panda, which is discussed in the next
Lichtenberg Algorithm (LA) [57], Archimedes Optimization section.

VOLUME 11, 2023 57205

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

Algorithm 1 Pseudocode of RPO

Start RPO.
Input: The problem information (variables, objective function, and constraints).
Set RPO population size (N ) and the total number of iterations (T ).
Generate the initial population matrix at random using (1) and (2).
Evaluate the objective function by (3).
For t = 1 to T
For i = 1 to N
Phase 1: : The strategy of red pandas in foraging
Update food positions set for the ith RPO member using (4). PFSi ← {Xk |k ∈ {1, 2, . . . , N } ∩ Fk < Fi } ∪ {Xbest }
Determine the selected food by the ith red panda at random.
Calculate new position of the ith RPO member based on the 1st phase of RPO using (5). xi,j P1 ← x + r · SFS − I · x


( i,j i,j i,j
XiP1 , FiP1 < Fi
Update the ith RPO member using (6). Xi ←
Xi , else
Phase 2: Skill in climbing and resting on the tree

lbj +r· ubj −lbj
Calculate new position of the ith RPO member based on the 2nd phase of RPO using (7). xi,j
P2 ← x +
i,j t
(
P2 P2
Xi , Fi < Fi
Update the ith RPO member using (8). Xi ←
Xi , else
end
Save the best candidate solution found so far.
end
Output: The best solution obtained by RPO.
End RPO.

largely arboreal and solitary [83]. A picture of the red panda

is shown in Figure 1.
Red panda is largely herbivorous and eats mainly bamboo
leaves and shoots, as well as blooms and fruits. The red panda
has a good sense of sight, smell, and hearing and uses its long
white whiskers to search for food at night [84]. According
to observations, the red panda is a nocturnal animal. Due to
its high ability to climb, it sleeps and rests in high places,
especially trees during the day [85].
Among the natural behaviors of the red panda, its forag-
ing strategy based on its high ability of hearing, sight, and
smell, as well as the high skill of this animal in climbing
trees, is much more impressive. Mathematical modeling of
these natural behaviors of the red panda is the basis for the
FIGURE 1. Photo of a red panda; downloaded from free media Wikimedia design of the proposed RPO approach, which is explained
Commons.
below.

B. MATHEMATICAL MODELLING
In this subsection, first the initialization of the proposed RPO
III. RED PANDA OPTIMIZATION ALGORITHM
approach is described, then based on the simulation of the
In this section, the proposed Red Panda Optimization (RPO)
natural behaviors of the red panda, the mathematical model of
algorithm is introduced, then its mathematical modeling is
updating the candidate solutions in two phases of exploration
presented.
and exploitation is presented.

A. INSPIRATION OF RPO 1) INITIALIZATION

The red panda is a small endemic animal of the southern The proposed RPO approach is a population-based meta-
China and eastern Himalayas. It has dense reddish-brown heuristic algorithm, whose members consist of red pandas.
hair on its body and legs, a black belly and legs, white-lined In RPO design, each red panda is a candidate solution to
ears, a mainly white muzzle, and a ringed tail. It has a head- the problem, which suggests certain values for the problem
to-body length of 51-63.5 cm, a tail length of 28-48.5 cm, variables based on its position in the search space. There-
and weighs between 3.2 and 15 kg. Because of its flexible fore, from a mathematical point of view, each red panda
joints and curved semi-retractile claws, it is well adapted to (i.e., candidate solution) is modeled using a vector. Together,
climbing [82]. The red panda inhabits temperate broadleaf the red pandas of the algorithm population can be math-
and mixed forests as well as coniferous forests, favoring steep ematically modeled using a matrix according to (1). Each
slopes with dense bamboo cover close to water sources. It is row of this matrix represents a red panda (i.e., candidate

57206 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

FIGURE 2. Flowchart of RPO.

solution) and each column of this matrix represents the sug- problem. At the beginning of RPO execution, the position
gested values for the corresponding variable of the given of red pandas in the search space is randomly initialized

VOLUME 11, 2023 57207

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 1. Control parameter values for the competitor algorithms. TABLE 1. (Continued.) Control parameter values for the competitor
algorithms.

using (2).
 
X1
 .. 
 . 
 
X = Xi 

 .. 
 . 
XN N ×m
 
x1,1 ··· x1,j ··· x1,m
 .. .. .. . .. 
 .
 . . .. . 
=  xi,1
 ··· xi,j ··· xi,m  (1)
 .. .. .. .. 

.
 . .. . . . 
xN ,1 · · · xN ,j ··· xN ,m N ×m
xi,j = lbj + ri,j · ubj − lbj , i = 1, 2, . . . , N ,

j = 1, 2, . . . , m, (2)
where, X is the population matrix of red pandas’ locations,
Xi is the ith red panda (i.e., candidate solution), xi,j is its jth
dimension (problem variable), N is the number of red pandas,
m indicates the number of problem variables, ri,j are random
numbers in the interval [0, 1], lbj , and ubj are the lower bound
and upper bound of the jth problem variable, respectively.
Considering that the position of each red panda is a can-
didate solution for the problem, the objective function of the
problem corresponding to each of these candidate solutions
can be evaluated. The set of evaluated values for the objec-
tive function can be represented using a matrix according
to (3).
F (X1 )
   
F1
 ..   .. 
 .   . 
 F (Xi ) 
   
F = F
 i 
 = (3)
 ..   .. 
 
 .   . 
FN N ×1 F (XN ) N ×1
where F is the objective function values vector and Fi denotes
the value of the objective function obtained by the ith red
panda.
The evaluated values for the objective function of the prob-
lem are the main criterion in determining the quality of the
candidate solutions. In other words, the best obtained value

57208 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 2. Optimization results of the unimodal test functions.

for the objective function corresponds to the best candidate consists of two phases of exploration and exploitation, which
solution and similarly the worst value obtained for the objec- are described as follows.
tive function corresponds to the worst candidate solution.
Since the candidate solutions are updated in each iteration, 2) PHASE 1: THE STRATEGY OF RED PANDAS IN FORAGING
the best and worst candidate solutions must also be updated (EXPLORATION)
in each iteration. After the implementation of the algorithm, The position of red pandas in the first phase of RPO is
the best candidate solution obtained during the iterations of modeled based on their movement in order to forage in the
the algorithm is presented as a solution to the problem. The wild. Red pandas are highly skilled in identifying and moving
process of updating candidate solutions in the proposed RPO towards the location of food sources using their high abilities

VOLUME 11, 2023 57209

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 3. Optimization results of the high-dimensional multimodal test functions.

in smell, hearing, and vision. In RPO design, for each red Moving towards the food source leads to big changes in
panda, the location of other red pandas that lead to better the position of red pandas, which improves the capability
objective function values is considered as the location of food of the proposed algorithm in exploration and global search
resources. The set of proposed food resource positions for in the problem-solving space. In order to model the behav-
each red panda based on the comparison of the objective ior of red pandas during foraging, first a new position is
function values is modeled using (4). Among these proposed calculated for each red panda based on movement towards
positions, one position is randomly determined as the food the location of food source (the best candidate solution)
position selected by the corresponding red panda. using (5). Then, if the value of the objective function is
improved in the new position, the position of the red panda
PFSi = {Xk |k ∈ {1, 2, . . . , N } and Fk < Fi } ∪ {Xbest } ,
is updated to the position calculated in the exploration phase
(4) using (6).
where PFSi is the set of proposed food sources for ith red
panda and Xbest is the location of the red panda with best value
for the objective function (best candidate solution). XiP1 : xi,j
P1
= xi,j + r · (SFS i,j − I · xi,j ) (5)

57210 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 4. Optimization results of the fixed-dimensional multimodal test functions.

VOLUME 11, 2023 57211

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 4. (Continued.) Optimization results of the fixed-dimensional multimodal test functions.

FIGURE 3. Convergence curves of RPO and the competitor algorithms performances for F1 to F23 test functions.

57212 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

FIGURE 3. (Continued.) Convergence curves of RPO and the competitor algorithms performances for F1 to F23 test functions.

(
XiP1 , FiP1 < Fi ; C. REPETITIONS PROCESS, FLOWCHART, AND
Xi = (6)
Xi , else, PSEUDO-CODE OF RPO
The proposed RPO approach is an iteration-based meta-
where, XiP1 is the new position of the ith red panda based heuristic algorithm. After updating the position of all red pan-
on the first phase of RPO, xi,j P1 is its jth dimension, F P1 das based on the exploration and exploitation phases, the first
i
represents its objective function value, SFS i is the selected iteration of the RPO is completed. Then, based on the new
food source for ith red panda, SFS i,j denotes its jth dimension, values, the algorithm enters the next iteration and the pro-
r is a random number in the interval [0, 1], and I is a random cess of updating the position of the red pandas is repeated
number selected from the set {1, 2} randomly. using (4) to (8) until the last iteration of the algorithm.
After completion of RPO implementation, the position of
the best red panda, which results in the best value for the
3) PHASE 2: SKILL IN CLIMBING AND RESTING ON THE objective function, is presented as the solution of the problem.
TREE (EXPLOITATION) The implementation steps of RPO are presented in the form
The position of red pandas in the second phase of the RPO of a flowchart in Figure 2 and its pseudocode is given in
is modeled based on the skill of this animal in climbing Algorithm 1.
trees and resting on them. Red pandas spend most of their
time resting on trees. After foraging on the ground, this D. COMPUTATIONAL COMPLEXITY
animal climbs the nearby trees. Moving towards the tree
In this subsection, the computational complexity analysis of
and climbing it leads to small changes in the position of
the proposed RPO approach is discussed. RPO initialization
red pandas, which increases the capability of the proposed
has a computational complexity equal to O(Nm), where N is
RPO algorithm in exploitation and local search in promising
the number of red pandas and m denotes the number of prob-
areas. In order to mathematically model the natural behav-
lem variables. In each iteration of the algorithm, the position
ior of red pandas in climbing trees, first a new position is
of red pandas is updated in the two phases of exploration
calculated for each red panda using (7). Then, if the value
and exploitation. Therefore, the red pandas update process
of the objective function is improved, this new position
has a computational complexity equal to O(2NmT), where T
replaces the previous position of the corresponding red panda
is the maximum number of the algorithm iterations. There-
using (8).
fore, the total computational complexity of RPO is equal to
lbj + r · ubj − lbj

O(Nm(1 + 2T).
P2
xi,j = xi,j + , i = 1, 2, . . . , N ,
t
j = 1, 2, . . . , m, and t = 1, 2, . . . , T , (7) IV. SIMULATION STUDIES AND DISCUSSION
( In this section, simulation studies on the performance of the
Xi , Fi < Fi ;
P2 P2
proposed RPO in solving optimization problems are pre-
Xi = (8)
Xi , else, sented. For this purpose, fifty-two standard benchmark func-
tions consisting of unimodal, high-dimensional multimodal,
where XiP2 is the new position of the ith red panda based and fixed-dimensional multimodal types as well as CEC
on the second phase of RPO, xi,j P2 is its jth dimension,
2017 test suite [86] are employed. Also, in order to analyze
P2
Fi indicates its objective function value, r is a random the quality of RPO in providing appropriate solutions, the
number in the interval [0, 1], t represents the iteration results obtained from the proposed approach are compared
counter of the algorithm, and T is the maximum number of with the performance of twelve well-known metaheuristic
iterations. algorithms including: GA, PSO, GSA, TLBO, GWO, MVO,

VOLUME 11, 2023 57213

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 5. Optimization results of the CEC-2017 test suite.

57214 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 5. (Continued.) Optimization results of the CEC-2017 test suite.

VOLUME 11, 2023 57215

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 5. (Continued.) Optimization results of the CEC-2017 test suite.

57216 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 5. (Continued.) Optimization results of the CEC-2017 test suite.

WOA, TSA, MPA, RSA, WSO, and AVOA. The values of compared to functions F8 to F13, have a lower number of
the control parameters for these competitor algorithms are local optima. Functions F14 to F23 are suitable options for
specified in Table 1. Simulation results are reported using evaluating the ability of metaheuristic algorithms in balanc-
six statistical indicators: mean, best, worst, standard devia- ing exploration and exploitation features during the search
tion (std), median, and rank. The ranking criterion for meta- process. The results of using RPO and the competitor algo-
heuristic algorithms in solving each benchmark function is to rithms for optimizing the functions F14 to F23 are presented
provide a better value for the mean index. in Table 4. Based on the optimization results, RPO is the
first-best optimizer for the functions F14, F15, F21, F22, and
A. EVALUATION OF UNIMODAL TEST FUNCTIONS F23. In solving the functions F16, F17, F18, F19, and F20,
Seven benchmark functions F1 to F7 are selected from the RPO has the same conditions as some of the competing algo-
unimodal type. Because these functions have no local optima, rithms considering the mean index criterion. However, RPO
they are suitable options for evaluating the exploitation power has provided a more effective performance in handling these
of metaheuristic algorithms. The optimization results for the functions by providing better results from the std index view-
functions F1 to F7 using RPO and the competitor algorithms point. What is evident from the analysis of simulation results,
are presented in Table 2. Based on the optimization results, RPO has achieved better results in solving fixed-dimensional
RPO with high exploitation ability has converged to the multimodal functions with an appropriate ability to balance
global optima in solving functions F1, F2, F3, F4, F5, and exploration and exploitation, and compared to the competitor
F6. In solving the function F7, RPO is the first-best optimizer. algorithms, it has provided superior performance in optimiz-
The analysis of the simulation results shows that the proposed ing these functions.
RPO approach has provided better results and in total by win- The performance of RPO and the competitor algorithms in
ning the first rank, it has achieved a superior performance in solving benchmark functions F1 to F23 is illustrated in the
the optimization of unimodal benchmark functions compared form of convergence curves in Figure 3.
to the competitor algorithms.
D. EVALUATION OF CEC 2017 TEST SUITE
B. EVALUATION OF HIGH-DIMENSIONAL MULTIMODAL RPO’s performance in solving optimization problems is eval-
TEST FUNCTIONS uated on CEC 2017 test suite. This test suite has thirty
Six benchmark functions F8 to F13 are selected from benchmark functions consisting of three unimodal functions
high-dimensional multimodal type. In addition to the global of C17-F1 to C17-F3, seven multimodal functions of C17-F4
optima, these functions have a large number of local optima, to C17-F10, ten hybrid functions of C17-F11 to C17-F20, and
and for this reason, they are suitable options for evaluating the ten composition functions of C17-F21 to C17-F30. The C17-
exploration power of metaheuristic algorithms. The imple- F2 function has been excluded from the simulation studies
mentation results of RPO and the competitor algorithms for due to its unstable behavior. The full description of CEC
the functions F8 to F13 are reported in Table 3. The opti- 2017 test suite is provided in [86]. The implementation results
mization results show that RPO with high exploration ability of RPO and the competitor algorithms on CEC 2017 test suite
has converged to the global optima in the optimization of F9 are reported in Table 5. Based on the optimization result,
and F11 functions in addition to identifying the main optimal RPO is the first best optimizer for functions C17-F1, C17-
area in the search space. In solving the functions F8, F10, F4 to C17-F8, C17-F10 to C17-F21, C17-F23, C17-F24,
F12, and F13, RPO has provided suitable solutions with high and C17-F26 to C17-F30. The performance of RPO and the
exploration ability and is the first-best optimizer for these competitor algorithms in solving the CEC 2017 test suite
functions. The comparison of the simulation results indicates is drawn as boxplot diagrams in Figure 4. Analysis of the
that the proposed RPO approach, with a high exploration simulation results shows that RPO has provided better results
ability in the case of high-dimensional multimodal func- for most of the benchmark functions of CEC 2017 test suite
tions, has obtained superior performance over the competitor and overall, by winning the first rank, it has provided superior
algorithms. performance over the competitor algorithms in solving CEC
2017 test suite.
C. EVALUATION OF FIXED-DIMENSIONAL MULTIMODAL
TEST FUNCTIONS E. STATISTICAL ANALYSIS
Ten benchmark functions F14 to F23 have been selected from In this subsection, statistical analysis is presented on the per-
the fixed-dimensional multimodal type. These functions, formance of RPO and the competitor algorithms to determine

VOLUME 11, 2023 57217

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

FIGURE 4. Boxplot diagrams of RPO and the competitor algorithms performances for the CEC-2017 test suite.

57218 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

FIGURE 4. (Continued.) Boxplot diagrams of RPO and the competitor algorithms performances for the CEC-2017 test suite.

whether RPO has a significant statistical superiority or not. TABLE 6. p-values obtained from Wilcoxon rank sum test.
For this purpose, Wilcoxon rank sum test [87] is employed,
which is a non-parametric statistical test to determine the sig-
nificant difference between the average of two data samples.
In Wilcoxon rank sum test, an index called p-value is utilized
to evaluate whether RPO is significantly superior over any of
the competitor algorithms from a statistical point of view.
The results of the statistical analysis on the performance of
RPO and the competitor algorithms are reported in Table 6.
Based on the results, in cases where the p-value is less than
0.05, RPO has a significant statistical superiority over the
corresponding competitor algorithm.

V. RPO FOR REAL-WORLD APPLICATION

In this section, the effectiveness of the proposed RPO
approach for solving optimization problems in real-world mathematical model is as follows [88]:
applications is tested. In this regard, RPO is implemented on
four engineering design problems. Consider: X = [x1 , x2 , x3 , x4 ] = [Ts , Th , R, L] .
Minimize: f (x) = 0.6224x1 x3 x4 + 1.778x2 x32
A. PRESSURE VESSEL DESIGN PROBLEM
+ 3.1661x12 x4 + 19.84x12 x3 .
Pressure vessel design is an engineering minimization prob-
lem with the aim of reducing design cost. The schematic of Subject to: g1 (x) = −x1 + 0.0193x3 ≤ 0,
this design is presented in Figure 5. Pressure vessel design g2 (x) = −x2 + 0.00954x3 ≤ 0,

VOLUME 11, 2023 57219

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

FIGURE 5. Schematic view of pressure vessel design problem. FIGURE 7. Schematic view of speed reducer design problem.

 
× 3.3333x32 +14.9334x3 − 43.0934
 
− 1.508x1 x62 + x72 + 7.4777
 
× x63 + x73 + 0.7854(x4 x62 + x5 x72 ).
27
Subject to: g1 (x) = − 1 ≤ 0,
x1 x22 x3
397.5
g2 (x) = − 1 ≤ 0,
x1 x22 x3
1.93x43
g3 (x) = − 1 ≤ 0,
x2 x3 x64
FIGURE 6. Convergence analysis of the RPO for the pressure vessel
design optimization problem. 1.93x53
g4 (x) = − 1 ≤ 0,
x2 x3 x74
s
4 1

745x4 2

g3 (x) = −πx32 x4 − πx33 + 1296000 ≤ 0, g5 (x) = + 16.9 · 106 − 1
3 110x63 x2 x3
g4 (x) = x4 − 240 ≤ 0.
≤ 0,
s
with
745x5 2
 
1
g6 (x) = + 157.5 · 106 − 1
0 ≤ x1 , x2 ≤ 100 and 10 ≤ x 3 , x4 ≤ 200. 85x73 x2 x3
The results of implementing RPO and the competitor ≤ 0,
x2 xn3
algorithms for solving pressure vessel design problem are g7 (x) = − 1 ≤ 0,
reported in Tables 7 and 8. Based on the obtained results, 40
5x 2
RPO has provided the optimal solution of this design with the g8 (x) = − 1 ≤ 0,
design values equal to (0.778027, 0.384579, 40.31228, 200) x1
x1
and the corresponding objective function value is equal to g9 (x) = − 1 ≤ 0,
12x2
(5882.895). The convergence curve of RPO while achieving
1.5x 6 + 1.9
the solution for pressure vessel design is drawn in Figure 6. g10 (x) = − 1 ≤ 0,
Based on the simulation results, RPO has provided superior x4
1.1x 7 + 1.9
performance in pressure vessel design optimization compared g11 (x) = − 1 ≤ 0.
to the competitor algorithms. x5
with
B. SPEED REDUCER DESIGN PROBLEM
Speed reducer design is an engineering minimization prob- 2.6 ≤ x1 ≤ 3.6, 0.7 ≤ x2 ≤ 0.8, 17 ≤ x3 ≤ 28,
lem with the aim of reducing the weight of the speed reducer. 7.3 ≤ x4 ≤ 8.3, 7.8 ≤ x5 ≤ 8.3, 2.9 ≤ x6 ≤ 3.9,
The schematic of this design is illustrated in Figure 7. and 5 ≤ x7 ≤ 5.5.
The mathematical model of speed reducer design is as fol-
lows [89], [90]: Speed reducer design optimization results using RPO and
the competitor algorithms are reported in Tables 9 and 10.
Consider: X = [x1 , x2 , x3 , x4 , x5 , x 6 , x7 ]
Based on the obtained results, RPO has provided the optimal
= [b, M , p, l1 , l2 , d1 , d2 ] . solution of this design with the design values equal to (3.5,
Minimize: f (x) = 0.7854x1 x22 0.7, 17, 7.3, 7.8, 3.350215, 5.286683) and the corresponding

57220 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 7. Performance of optimization algorithms for pressure vessel design problem.

TABLE 8. Statistical results of optimization algorithms for pressure vessel design problem.

schematic of this design is shown in Figure 9. The mathemat-

ical model of the welded beam design is as follows [40]:

Consider: X = [x1 , x2 , x3 , x4 ] = [h, l, t, b] .

Minimize: f (x) = 1.10471x12 x2 + 0.04811x3 x4 (14.0 + x2 ).
Subject to: g1 (x) = τ (x) − 13600 ≤ 0,
g2 (x) = σ (x) − 30000 ≤ 0,
g3 (x) = x1 − x4 ≤ 0,
g4 (x) = 0.10471x12 + 0.04811x3 x4 (14 + x2 )
− 5.0 ≤ 0,
g5 (x) = 0.125 − x1 ≤ 0, g6 (x) = δ (x)
FIGURE 8. Convergence analysis of the RPO for the speed reducer design
optimization problem. − 0.25 ≤ 0,
g7 (x) = 6000 − pc (x) ≤ 0.
objective function value is equal to (2996.348). The RPO
convergence curve during solving the speed reducer design
where
is drawn in Figure 8. Analysis of the simulation results
shows that RPO is superior over the competitor algorithms by r
x2 6000
providing better results in the optimization of speed reducer τ (x) = (τ ′ )2 + (2τ τ ′ )
+ (τ ′′ )2 , τ ′ = √ ,
2R 2x1 x2
design.
MR  x2 
τ ′′ = , M = 6000 14 + ,
C. WELDED BEAM DESIGN PROBLEM sJ 2
x22 x1 + x3 2
 
Welded beam design is a real-world application with the aim
R= + ,
of minimizing the fabrication cost of the welded beam. The 4 2

VOLUME 11, 2023 57221

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 9. Performance of optimization algorithms for speed reducer design problem.

TABLE 10. Statistical results of optimization algorithms for speed reducer design problem.

!
√ x22 x1 +x3 2

504000
J = 2 2 x1 x2 + , σ (x) = ,
12 2 x4 x32
65856000
δ (x) = ,
30 · 106 x4 x33

 s 
4.013 30 · 106 x3 x43


x 30 · 106
3 .
pc (x) = 1 −
1176 28 4(12 · 106 )

with

0.1 ≤ x1 , x4 ≤ 2 and 0.1 ≤ x2 , x3 ≤ 10.

The results of employing RPO and the competitor algo-

rithms in handling the welded beam design problem are
presented in Tables 11 and 12. Based on the obtained FIGURE 9. Schematic view of the welded beam design problem.
results, RPO has provided the optimal solution of this
design with the design values equal to (0.20573, 3.470489, D. TENSION/COMPRESSION SPRING DESIGN PROBLEM
9.036624, 0.20573) and the corresponding objective func- Tension/compression spring design is a real-world applica-
tion value is equal to (1.72468). The convergence curve of tion aimed at minimizing the weight of tension/compression
RPO while reaching the solution for welded beam design spring. The schematic of this design is presented in
is drawn in Figure 10. What is evident from the simulation Figure 11. The mathematical model of tension/compression
results, RPO has a higher ability compared to the com- spring design is as follows [40]:
petitor algorithms in dealing with the welded beam design
problem. Consider: X = [x1 , x2 , x3 ] = [d, D, P] .

57222 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 11. Performance of optimization algorithms for welded beam design problem.

TABLE 12. Statistical results of optimization algorithms for welded beam design problem.

TABLE 13. Performance of optimization algorithms for tension/compression spring design problem.

x1 + x2
Minimize: f (x) = (x3 + 2) x2 x12 . g4 (x) = − 1 ≤ 0.
1.5
x23 x3
Subject to: g1 (x) = 1 − ≤ 0, with
71785x14
4x22 − x1 x2 1 0.05 ≤ x1 ≤ 2, 0.25 ≤ x 2 ≤ 1.3 and 2 ≤ x3 ≤ 15.
g2 (x) = + − 1 ≤ 0,
12566(x2 x13 ) 5108x12
The implementation results of RPO and the competitor
140.45x1
g3 (x) = 1− ≤ 0, algorithms for the tension/compression spring design prob-
x22 x3 lem are reported in Tables 13 and 14. Based on the obtained

VOLUME 11, 2023 57223

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

TABLE 14. Statistical results of optimization algorithms for tension/compression spring design problem.

is equal to (0.012602). The convergence curve of RPO

while achieving the optimal solution for tension/compression
spring design is drawn in Figure 12. What can be concluded
from the simulation results, RPO has provided a more effec-
tive performance compared to the competitor algorithms in
solving the tension/compression spring design problem.

VI. CONCLUSION AND FUTURE WORKS

In this paper, a new bio-inspired metaheuristic algorithm
called Red Panda Optimization (RPO) was introduced, which
can be applied to solve optimization problems. The fun-
damental inspiration of RPO is simulation of the behav-
FIGURE 10. Convergence analysis of the RPO for the welded beam design ior of red pandas when foraging and their ability to climb
optimization problem.
trees to rest. The implementation steps of RPO were
described and mathematically modeled in two phases of
exploration and exploitation. The effectiveness of RPO in
solving optimization problems was evaluated considering
fifty-two benchmark functions consisting of unimodal, high-
dimensional multimodal, fixed-dimensional multimodal, and
CEC 2017 test suite. The optimization results of unimodal
FIGURE 11. Schematic view of tension/compression spring design functions indicated the high ability of RPO in local search
problem.
and exploitation. The optimization results of multimodal
functions showed that RPO has a high ability in global
search and exploration. Also, the optimization results of
CEC 2017 test suite showed the high capability of the
proposed RPO approach in providing simultaneous explo-
ration and exploitation during the search process. The results
obtained from the implementation of RPO were compared
with the performance of twelve well-known metaheuristic
algorithms. The simulation results showed that the proposed
RPO approach by balancing exploration and exploitation
features during the search process, has provided superior
performance over the competitor algorithms. Based on the
simulation results, the proposed RPO approach provided bet-
ter results compared to the competitor algorithms in 100% of
FIGURE 12. Convergence analysis of the RPO for the
tension/compression spring design optimization problem.
unimodal functions, 100% of high-dimensional multimodal
functions, 100% of fixed-dimensional multimodal functions,
results, RPO has provided the optimal solution of this and 86.2% of CEC 2017 test suite benchmark functions.
design with the design values equal to (0.051689, 0.356718, In addition, the implementation of RPO on four engineering
11.28897) and the corresponding objective function value design problems showed that the proposed algorithm has

57224 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

a high ability to solve optimization problems in real-world [16] H. Rezk, A. Fathy, M. Aly, and M. N. Ibrahim, ‘‘Energy management
applications. control strategy for renewable energy system based on spotted hyena
optimizer,’’ Comput., Mater. Continua, vol. 67, no. 2, pp. 2271–2281,
Following the introduction of the proposed RPO approach, 2021.
several research paths are activated for further studies. The [17] A. Ehsanifar, M. Dehghani, and M. Allahbakhshi, ‘‘Calculating the leak-
development of binary and multi-objective versions of RPO is age inductance for transformer inter-turn fault detection using finite ele-
ment method,’’ in Proc. Iranian Conf. Electr. Eng. (ICEE), May 2017,
one of the most significant research potentials in this regard. pp. 1372–1377.
The use of RPO for solving optimization problems in various [18] M. Dehghani, Z. Montazeri, A. Ehsanifar, A. R. Seifi, M. J. Ebadi, and
fields of science as well as optimization tasks in real-world O. M. Grechko, ‘‘Planning of energy carriers based on final energy con-
applications is one of the other suggestions of this paper for sumption using dynamic programming and particle swarm optimization,’’
Electr. Eng. Electromech., no. 5, pp. 62–71, Oct. 2018.
future investigations. [19] Z. Montazeri and T. Niknam, ‘‘Energy carriers management based on
energy consumption,’’ in Proc. IEEE 4th Int. Conf. Knowl.-Based Eng.
ACKNOWLEDGMENT Innov. (KBEI), Dec. 2017, pp. 539–543.
[20] M. Dehghani, Z. Montazeri, and O. P. Malik, ‘‘Optimal sizing and place-
The authors thank Štěpán Hubálovský from the University of ment of capacitor banks and distributed generation in distribution systems
Hradec Králové for fruitful and informative discussions. using spring search algorithm,’’ Int. J. Emerg. Electr. Power Syst., vol. 21,
no. 1, pp. 1–9, Jan. 2020.
[21] M. Dehghani, Z. Montazeri, O. P. Malik, K. Al-Haddad, J. M. Guerrero,
REFERENCES and G. Dhiman, ‘‘A new methodology called dice game optimizer for
[1] H. Xian, C. Yang, H. Wang, and X. Yang, ‘‘A modified sine cosine capacitor placement in distribution systems,’’ Electr. Eng. Electromech.,
algorithm with teacher supervision learning for global optimization,’’ IEEE no. 1, pp. 61–64, Feb. 2020.
Access, vol. 9, pp. 17744–17766, 2021. [22] S. Dehbozorgi, A. Ehsanifar, Z. Montazeri, M. Dehghani, and A. Seifi,
[2] A. W. Mohamed, A. A. Hadi, and A. K. Mohamed, ‘‘Gaining-sharing ‘‘Line loss reduction and voltage profile improvement in radial distribution
knowledge based algorithm for solving optimization problems: A novel networks using battery energy storage system,’’ in Proc. IEEE 4th Int. Conf.
nature-inspired algorithm,’’ Int. J. Mach. Learn. Cybern., vol. 11, no. 7, Knowl.-Based Eng. Innov. (KBEI), Dec. 2017, pp. 215–219.
pp. 1501–1529, Jul. 2020. [23] Z. Montazeri and T. Niknam, ‘‘Optimal utilization of electrical energy from
[3] A. S. Assiri, A. G. Hussien, and M. Amin, ‘‘Ant lion optimization: Vari- power plants based on final energy consumption using gravitational search
ants, hybrids, and applications,’’ IEEE Access, vol. 8, pp. 77746–77764, algorithm,’’ Electr. Eng. Electromech., no. 4, pp. 70–73, Aug. 2018.
2020. [24] M. Dehghani, M. Mardaneh, Z. Montazeri, A. Ehsanifar, M. J. Ebadi, and
[4] P. Coufal, Š. Hubálovský, M. Hubálovská, and Z. Balogh, ‘‘Snow leopard O. M. Grechko, ‘‘Spring search algorithm for simultaneous placement of
optimization algorithm: A new nature-based optimization algorithm for distributed generation and capacitors,’’ Electr. Eng. Electromech., no. 6,
solving optimization problems,’’ Mathematics, vol. 9, no. 21, p. 2832, pp. 68–73, Dec. 2018.
Nov. 2021. [25] M. Premkumar, R. Sowmya, P. Jangir, K. S. Nisar, and M. Aldhaifallah,
[5] J. Xue and B. Shen, ‘‘A novel swarm intelligence optimization approach: ‘‘A new metaheuristic optimization algorithms for brushless direct current
Sparrow search algorithm,’’ Syst. Sci. Control Eng., vol. 8, no. 1, pp. 22–34, wheel motor design problem,’’ Comput., Mater. Continua, vol. 67, no. 2,
Jan. 2020. pp. 2227–2242, 2021.
[6] S. Mirjalili, A. H. Gandomi, S. Z. Mirjalili, S. Saremi, H. Faris, and [26] D. H. Wolpert and W. G. Macready, ‘‘No free lunch theorems for optimiza-
S. M. Mirjalili, ‘‘Salp swarm algorithm: A bio-inspired optimizer for tion,’’ IEEE Trans. Evol. Comput., vol. 1, no. 1, pp. 67–82, Apr. 1997.
engineering design problems,’’ Adv. Eng. Softw., vol. 114, pp. 163–191,
[27] M. Dehghani and P. Trojovský, ‘‘Serval optimization algorithm: A new bio-
Dec. 2017.
inspired approach for solving optimization problems,’’ Biomimetics, vol. 7,
[7] S. Mirjalili, ‘‘The ant lion optimizer,’’ Adv. Eng. Softw., vol. 83, pp. 80–98,
no. 4, p. 204, Nov. 2022.
May 2015.
[28] J. Kennedy and R. Eberhart, ‘‘Particle swarm optimization,’’ in Proc. Int.
[8] M. Cavazzuti, ‘‘Deterministic optimization,’’ Optimization Methods: From
Conf. Neural Netw., vol. 4, 1995, pp. 1942–1948.
Theory to Design Scientific and Technological Aspects in Mechanics.
Berlin, Germany: Springer, 2013, pp. 77–102. [29] M. Dorigo, V. Maniezzo, and A. Colorni, ‘‘Ant system: Optimization by
a colony of cooperating agents,’’ IEEE Trans. Syst., Man, Cybern. B,
[9] K. Iba, ‘‘Reactive power optimization by genetic algorithm,’’ IEEE Trans.
Cybern., vol. 26, no. 1, pp. 29–41, Feb. 1996.
Power Syst., vol. 9, no. 2, pp. 685–692, May 1994.
[10] S. S. Mohar, S. Goyal, and R. Kaur, ‘‘Localization of sensor nodes in [30] D. Karaboga and B. Basturk, ‘‘Artificial bee colony (ABC) optimization
wireless sensor networks using bat optimization algorithm with enhanced algorithm for solving constrained optimization problems,’’ in Proc. Int.
exploration and exploitation characteristics,’’ J. Supercomput., vol. 78, Fuzzy Syst. Assoc. World Congr., 2007, pp. 789–798.
no. 9, pp. 11975–12023, Jun. 2022. [31] X.-S. Yang, ‘‘Firefly algorithm, stochastic test functions and design opti-
[11] G. Brunetti, C. Stumpp, and J. Šimunek, ‘‘Balancing exploitation and misation,’’ 2010, arXiv:1003.1409.
exploration: A novel hybrid global-local optimization strategy for hydro- [32] M. Dehghani, Z. Montazeri, E. Trojovská, and P. Trojovský, ‘‘Coati
logical model calibration,’’ Environ. Model. Softw., vol. 150, Apr. 2022, optimization algorithm: A new bio-inspired metaheuristic algorithm for
Art. no. 105341. solving optimization problems,’’ Knowl.-Based Syst., vol. 259, Jan. 2023,
[12] M. Dehghani, Z. Montazeri, A. Dehghani, and O. P. Malik, ‘‘GO: Group Art. no. 110011.
optimization,’’ Gazi Univ. J. Sci., vol. 33, no. 2, pp. 381–392, Jun. 2020. [33] N. Chopra and M. M. Ansari, ‘‘Golden jackal optimization: A novel
[13] M. Dehghani, Z. Montazeri, and O. P. Malik, ‘‘Energy commitment: A nature-inspired optimizer for engineering applications,’’ Expert Syst. Appl.,
planning of energy carrier based on energy consumption,’’ Electr. Eng. vol. 198, Jul. 2022, Art. no. 116924.
Electromech., no. 4, pp. 69–72, Aug. 2019. [34] M. Braik, A. Hammouri, J. Atwan, M. A. Al-Betar, and M. A. Awadallah,
[14] M. Dehghani, M. Mardaneh, O. P. Malik, J. M. Guerrero, C. Sotelo, ‘‘White shark optimizer: A novel bio-inspired meta-heuristic algorithm for
D. Sotelo, M. Nazari-Heris, K. Al-Haddad, and R. A. Ramirez-Mendoza, global optimization problems,’’ Knowl.-Based Syst., vol. 243, May 2022,
‘‘Genetic algorithm for energy commitment in a power system supplied Art. no. 108457.
by multiple energy carriers,’’ Sustainability, vol. 12, no. 23, p. 10053, [35] A. Faramarzi, M. Heidarinejad, S. Mirjalili, and A. H. Gandomi, ‘‘Marine
Dec. 2020. predators algorithm: A nature-inspired metaheuristic,’’ Expert Syst. Appl.,
[15] M. Dehghani, M. Mardaneh, O. P. Malik, J. M. Guerrero, R. Morales- vol. 152, Aug. 2020, Art. no. 113377.
Menendez, R. A. Ramirez-Mendoza, J. Matas, and A. Abusorrah, ‘‘Energy [36] B. Abdollahzadeh, F. S. Gharehchopogh, and S. Mirjalili, ‘‘African vul-
commitment for a power system supplied by multiple energy carriers tures optimization algorithm: A new nature-inspired metaheuristic algo-
system using following optimization algorithm,’’ Appl. Sci., vol. 10, no. 17, rithm for global optimization problems,’’ Comput. Ind. Eng., vol. 158,
p. 5862, Aug. 2020. Aug. 2021, Art. no. 107408.

VOLUME 11, 2023 57225

 H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

[37] P. Trojovský and M. Dehghani, ‘‘Pelican optimization algorithm: A novel [62] R. V. Rao, V. J. Savsani, and D. P. Vakharia, ‘‘Teaching–learning-based
nature-inspired algorithm for engineering applications,’’ Sensors, vol. 22, optimization: A novel method for constrained mechanical design opti-
no. 3, p. 855, Jan. 2022. mization problems,’’ Comput.-Aided Des., vol. 43, no. 3, pp. 303–315,
[38] S. Kaur, L. K. Awasthi, A. L. Sangal, and G. Dhiman, ‘‘Tunicate swarm Mar. 2011.
algorithm: A new bio-inspired based metaheuristic paradigm for global [63] S. H. S. Moosavi and V. K. Bardsiri, ‘‘Poor and rich optimization algo-
optimization,’’ Eng. Appl. Artif. Intell., vol. 90, Apr. 2020, Art. no. 103541. rithm: A new human-based and multi populations algorithm,’’ Eng. Appl.
[39] F. A. Hashim, E. H. Houssein, K. Hussain, M. S. Mabrouk, and Artif. Intell., vol. 86, pp. 165–181, Nov. 2019.
W. Al-Atabany, ‘‘Honey badger algorithm: New metaheuristic algorithm [64] M. Dehghani, M. Mardaneh, J. M. Guerrero, O. P. Malik,
for solving optimization problems,’’ Math. Comput. Simul., vol. 192, R. A. Ramirez-Mendoza, J. Matas, J. C. Vasquez, and L. Parra-Arroyo,
pp. 84–110, Feb. 2022. ‘‘A new ‘doctor and patient’ optimization algorithm: An application
[40] S. Mirjalili and A. Lewis, ‘‘The whale optimization algorithm,’’ Adv. Eng. to energy commitment problem,’’ Appl. Sci., vol. 10, no. 17, p. 5791,
Softw., vol. 95, pp. 51–67, May 2016. Aug. 2020.
[65] M. Dehghani and P. Trojovský, ‘‘Teamwork optimization algorithm: A new
[41] L. Abualigah, M. A. Elaziz, P. Sumari, Z. W. Geem, and A. H. Gandomi,
optimization approach for function minimization/maximization,’’ Sensors,
‘‘Reptile search algorithm (RSA): A nature-inspired meta-heuristic opti-
vol. 21, no. 13, p. 4567, Jul. 2021.
mizer,’’ Expert Syst. Appl., vol. 191, Apr. 2022, Art. no. 116158.
[66] M. Braik, M. H. Ryalat, and H. Al-Zoubi, ‘‘A novel meta-heuristic algo-
[42] M. Dehghani, P. Trojovský, and O. P. Malik, ‘‘Green anaconda optimiza- rithm for solving numerical optimization problems: Ali baba and the forty
tion: A new bio-inspired metaheuristic algorithm for solving optimization thieves,’’ Neural Comput. Appl., vol. 34, no. 1, pp. 409–455, Jan. 2022.
problems,’’ Biomimetics, vol. 8, no. 1, p. 121, Mar. 2023. [67] H. Givi and M. Hubalovska, ‘‘Skill optimization algorithm: A new human-
[43] A. H. Gandomi, X.-S. Yang, and A. H. Alavi, ‘‘Cuckoo search algorithm: based metaheuristic technique,’’ Comput., Mater. Continua, vol. 74, no. 1,
A metaheuristic approach to solve structural optimization problems,’’ Eng. pp. 179–202, 2023.
with Comput., vol. 29, no. 1, pp. 17–35, Jan. 2013. [68] P. Trojovský, M. Dehghani, E. Trojovská, and E. Milkova, ‘‘Language
[44] S. Mirjalili, S. M. Mirjalili, and A. Lewis, ‘‘Grey wolf optimizer,’’ Adv. education optimization: A new human-based metaheuristic algorithm for
Eng. Softw., vol. 69, pp. 46–61, Mar. 2014. solving optimization problems,’’ Comput. Model. Eng. Sci., vol. 136, no. 2,
[45] D. E. Goldberg and J. H. Holland, ‘‘Genetic algorithms and machine pp. 1527–1573, 2023.
learning,’’ Mach. Learn., vol. 3, nos. 2–3, pp. 95–99, 1988. [69] M. A. Al-Betar, Z. A. A. Alyasseri, M. A. Awadallah, and I. A. Doush,
[46] R. Storn and K. Price, ‘‘Differential evolution-a simple and efficient heuris- ‘‘Coronavirus herd immunity optimizer (CHIO),’’ Neural Comput. Appl.,
tic for global optimization over continuous spaces,’’ J. Global Optim., vol. 33, no. 10, pp. 5011–5042, May 2021.
vol. 11, no. 4, pp. 341–359, Dec. 1997. [70] T. S. L. V. Ayyarao, N. S. S. Ramakrishna, R. M. Elavarasan,
[47] L. N. D. Castro and J. I. Timmis, ‘‘Artificial immune systems as a novel soft N. Polumahanthi, M. Rambabu, G. Saini, B. Khan, and B. Alatas, ‘‘War
computing paradigm,’’ Soft Comput., vol. 7, no. 8, pp. 526–544, Aug. 2003. strategy optimization algorithm: A new effective metaheuristic algorithm
[48] H.-G. Beyer and H.-P. Schwefel, ‘‘Evolution strategies—A comprehensive for global optimization,’’ IEEE Access, vol. 10, pp. 25073–25105, 2022.
introduction,’’ Natural Comput., vol. 1, no. 1, pp. 3–52, 2002. [71] M. Dehghani, E. Trojovská, and P. Trojovský, ‘‘A new human-based
metaheuristic algorithm for solving optimization problems on the base of
[49] J. R. Koza and J. R. Koza, Genetic Programming: On the Programming
simulation of driving training process,’’ Sci. Rep., vol. 12, no. 1, p. 9924,
of Computers by Means of Natural Selection. Cambridge, MA, USA:
Jun. 2022.
MIT Press, 1992.
[72] M. Dehghani, M. Mardaneh, J. Guerrero, O. Malik, and V. Kumar, ‘‘Foot-
[50] R. G. Reynolds, ‘‘An introduction to cultural algorithms,’’ in Proc. 3rd ball game based optimization: An application to solve energy commitment
Annu. Conf. On Evol. Program., 1994, pp. 131–139. problem,’’ Int. J. Intell. Eng. Syst., vol. 13, no. 5, pp. 514–523, Oct. 2020.
[51] S. Kirkpatrick, C. D. Gelatt, and M. P. Vecchi, ‘‘Optimization by simulated [73] R. Moghdani and K. Salimifard, ‘‘Volleyball premier league algorithm,’’
annealing,’’ Science, vol. 220, no. 4598, pp. 671–680, 1983. Appl. Soft Comput., vol. 64, pp. 161–185, Mar. 2018.
[52] M. Dehghani, Z. Montazeri, A. Dehghani, and A. Seifi, ‘‘Spring search [74] F. A. Zeidabadi and M. Dehghani, ‘‘POA: Puzzle optimization algorithm,’’
algorithm: A new meta-heuristic optimization algorithm inspired by Int. J. Intell. Eng. Syst., vol. 15, no. 1, pp. 273–281, 2022.
Hooke’s law,’’ in Proc. IEEE 4th Int. Conf. Knowl.-Based Eng. Innov. [75] M. Dehghani, Z. Montazeri, H. Givi, J. Guerrero, and G. Dhiman, ‘‘Darts
(KBEI), Dec. 2017, pp. 210–214. game optimizer: A new optimization technique based on darts game,’’ Int.
[53] M. Dehghani and H. Samet, ‘‘Momentum search algorithm: A new meta- J. Intell. Eng. Syst., vol. 13, no. 5, pp. 286–294, Oct. 2020.
heuristic optimization algorithm inspired by momentum conservation [76] A. Kaveh and A. Zolghadr, ‘‘A novel meta-heuristic algorithm: Tug of war
law,’’ Social Netw. Appl. Sci., vol. 2, no. 10, pp. 1–15, Oct. 2020. optimization,’’ Iran Univ. Sci. Technol., vol. 6, no. 4, pp. 469–492, 2016.
[54] E. Rashedi, H. Nezamabadi-Pour, and S. Saryazdi, ‘‘GSA: A gravitational [77] H. Givi and M. Hubáovská, ‘‘Billiards optimization algorithm: A new
search algorithm,’’ Inf. Sci., vol. 179, no. 13, pp. 2232–2248, Jun. 2009. game-based metaheuristic approach,’’ Comput., Mater. Continua, vol. 74,
[55] H. Eskandar, A. Sadollah, A. Bahreininejad, and M. Hamdi, ‘‘Water no. 3, pp. 5283–5300, 2023.
cycle algorithm—A novel metaheuristic optimization method for solv- [78] M. Dehghani, Z. Montazeri, and O. P. Malik, ‘‘DGO: Dice game opti-
ing constrained engineering optimization problems,’’ Comput. Struct., mizer,’’ Gazi Univ. J. Sci., vol. 32, no. 3, pp. 871–882, Sep. 2019.
vols. 110–111, pp. 151–166, Nov. 2012. [79] S. Doumari, H. Givi, M. Dehghani, and O. Malik, ‘‘Ring toss game-based
optimization algorithm for solving various optimization problems,’’ Int. J.
[56] Z. Wei, C. Huang, X. Wang, T. Han, and Y. Li, ‘‘Nuclear reaction optimiza-
Intell. Eng. Syst., vol. 14, no. 3, pp. 545–554, Jun. 2021.
tion: A novel and powerful physics-based algorithm for global optimiza-
[80] M. Dehghani, Z. Montazeri, O. P. Malik, A. Ehsanifar, and A. Dehghani,
tion,’’ IEEE Access, vol. 7, pp. 66084–66109, 2019.
‘‘OSA: Orientation search algorithm,’’ Int. J. Ind. Electron., Control
[57] J. L. J. Pereira, M. B. Francisco, C. A. Diniz, G. A. Oliver, S. S. Cunha, and
Optim., vol. 2, no. 2, pp. 99–112, 2019.
G. F. Gomes, ‘‘Lichtenberg algorithm: A novel hybrid physics-based meta- [81] F. A. Zeidabadi, M. Dehghani, P. Trojovský, Š. Hubálovský, V. Leiva,
heuristic for global optimization,’’ Expert Syst. Appl., vol. 170, May 2021, and G. Dhiman, ‘‘Archery algorithm: A novel stochastic optimization
Art. no. 114522. algorithm for solving optimization problems,’’ Comput., Mater. Continua,
[58] F. A. Hashim, K. Hussain, E. H. Houssein, M. S. Mabrouk, and vol. 72, no. 1, pp. 399–416, 2022.
W. Al-Atabany, ‘‘Archimedes optimization algorithm: A new metaheuris- [82] A. R. Glatston, Red Panda: Biology and Conservation of the First Panda,
tic algorithm for solving optimization problems,’’ Appl. Intell., vol. 51, 2nd ed. London, U.K.: Academic, 2021, pp. 81–93.
no. 3, pp. 1531–1551, Mar. 2021. [83] F. Wei, Z. Feng, Z. Wang, and J. Hu, ‘‘Habitat use and separation
[59] A. Faramarzi, M. Heidarinejad, B. Stephens, and S. Mirjalili, ‘‘Equilibrium between the giant panda and the red panda,’’ J. Mammal., vol. 81, no. 2,
optimizer: A novel optimization algorithm,’’ Knowl.-Based Syst., vol. 191, pp. 448–455, May 2000.
Mar. 2020, Art. no. 105190. [84] J. Nijboer and E. S. Dierenfeld, ‘‘Red panda nutrition: How to feed a veg-
[60] S. Mirjalili, S. M. Mirjalili, and A. Hatamlou, ‘‘Multi-verse optimizer: A etarian carnivore,’’ in Red Panda. Amsterdam, The Netherlands: Elsevier,
nature-inspired algorithm for global optimization,’’ Neural Comput. Appl., 2022, pp. 225–238.
vol. 27, no. 2, pp. 495–513, Feb. 2016. [85] M. S. Roberts and J. L. Gittleman, ‘‘Ailurus fulgens,’’ Mammalian
[61] E. Cuevas, D. Oliva, D. Zaldivar, M. Pérez-Cisneros, and H. Sossa, ‘‘Circle Species, no. 222, pp. 1–8, Nov. 1984. [Online]. Available: https://academic.
detection using electro-magnetism optimization,’’ Inf. Sci., vol. 182, no. 1, oup.com/mspecies/article/doi/10.2307/3503840/2600576, doi: 10.2307/
pp. 40–55, Jan. 2012. 3503840.

57226 VOLUME 11, 2023

H. Givi et al.: RPO Algorithm: An Effective Bio-Inspired Metaheuristic Algorithm

[86] N. Awad, M. Ali, J. Liang, B. Qu, P. Suganthan, and P. Definitions, MOHAMMAD DEHGHANI received the B.S.
‘‘Evaluation criteria for the CEC 2017 special session and competition on degree in electrical engineering from Shahid
single objective real-parameter numerical optimization,’’ Nanyang Tech- Bahonar University, Kerman, Iran, in 2012, the
nol. Univ., Singapore, Tech. Rep., Nov. 2016. M.S. degree in electrical engineering from Shiraz
[87] F. Wilcoxon, ‘‘Individual comparisons by ranking methods,’’ in Break- University, Shiraz, Iran, in 2016, and the Ph.D.
throughs in Statistics. New York, NY, USA: Springer, 1992, pp. 196–202. degree from the Shiraz University of Technol-
[Online]. Available: https://link.springer.com/chapter/10.1007/978-1-
ogy, Shiraz, in 2020. His current research interests
4612-4380-9_16
include optimization, metaheuristic algorithms,
[88] B. K. Kannan and S. N. Kramer, ‘‘An augmented Lagrange multiplier
based method for mixed integer discrete continuous optimization and power systems, and energy commitment.
its applications to mechanical design,’’ J. Mech. Des., vol. 116, no. 2,
pp. 405–411, Jun. 1994.
[89] A. H. Gandomi and X.-S. Yang, ‘‘Benchmark problems in structural
optimization,’’ in Computational Optimization, Methods and Algorithms.
Berlin, Germany: Springer, 2011, pp. 259–281. [Online]. Available:
https://link.springer.com/chapter/10.1007/978-3-642-20859-1_12
[90] E. Mezura-Montes and C. A. C. Coello, ‘‘Useful infeasible solutions in
engineering optimization with evolutionary algorithms,’’ in Proc. Mexican
Int. Conf. Artif. Intell., 2005, pp. 652–662.

HADI GIVI (Member, IEEE) was born in Isfahan,

Iran, in 1988. He received the B.Sc. degree in ŠTĚPÁN HUBÁLOVSKÝ received the M.Sc. and
electrical power engineering from the Isfahan Uni- Ph.D. degrees from the Faculty of Mathemat-
versity of Technology, in 2010, the M.Sc. degree in ics and Physics, Charles University in Prague,
electrical power engineering from the Amirkabir Czech Republic, in 1995 and 1998, respectively.
University of Technology, in 2012, and the Ph.D. He became an Associate Professor with the Fac-
degree in electrical power engineering from Shiraz ulty of Informatics and Management, University of
University, in 2018. Since 2018, he has been an Hradec Králové, Czech Republic, in 2012. He is
Assistant Professor with the Department of Elec- currently the Vice-Dean of the Faculty of Sci-
trical Engineering, Shahreza Campus, University ence, University of Hradec Králové. His research
of Isfahan. His main research interests include condition monitoring of power interests include technical cybernetics, computer
electronic converters, optimization algorithms as well as their applications in simulation and optimization, and big data processing.
the power systems, and power quality analysis.

VOLUME 11, 2023 57227