You are on page 1of 15

Knowledge-Based Systems 256 (2022) 109711

Contents lists available at ScienceDirect

Knowledge-Based Systems
journal homepage: www.elsevier.com/locate/knosys

Discrete Artificial Algae Algorithm for solving Job-Shop Scheduling


Problems

Mehmet Akif Şahman a , , Sedat Korkmaz b
a
Department of Electrical and Electronics Engineering, Faculty of Technology, Selçuk University, Konya, Turkey
b
Department of Computer Engineering, Faculty of Engineering and Natural Sciences, Konya Technical University, Konya, Turkey

article info a b s t r a c t

Article history: The Job-Shop Scheduling Problem (JSSP) is an NP-hard problem and can be solved with both exact
Received 2 July 2021 methods and heuristic algorithms. When the dimensionality is increased, exact methods cannot
Received in revised form 28 June 2022 produce proper solutions, but heuristic algorithms can produce optimal or near-optimal results for
Accepted 13 August 2022
high-dimensional JSSPs in a reasonable time. In this work, novel versions of the Artificial Algae
Available online 19 August 2022
Algorithm (AAA) have been proposed to solve discrete optimization problems. Three encoding schemes
Keywords: (Random-Key (RK), Smallest Position Value (SPV), and Ranked-Over Value (ROV) Encoding Schemes)
Discrete optimization were integrated with AAA to solve JSSPs. In addition, the comparison of these three encoding schemes
Job Shop Scheduling Problem was carried out for the first time in this study. In the experiments, 48 JSSP problems that have 36 to
Metaheuristic algorithms 300 dimensions were solved with 24 different approaches obtained by integrating 3 different coding
Encoding schemes schemes into 8 state-of-the-art algorithms. As a result of the comparative and detailed analysis, the
best results in terms of makespan value were obtained by integrating the SPV coding scheme into the
AAA method.
© 2022 Elsevier B.V. All rights reserved.

1. Introduction machine at the same time and an uninterrupted processing time


of a given length to be executed on m machines [2].
Increasing global competition and changing market condi- This intractable problem, JSSP, is not solved effectively with
tions force companies to benefit from advanced manufacturing mathematical models [3]. Therefore, researchers try to solve JSSP
processes in which new technologies are used. However, new with heuristic algorithms. Heuristic algorithms do not guarantee
production tactics and methods need to be developed rapidly. the optimal solutions, but in a reasonable time, they produce
This is possible with effective planning and accurate scheduling, near-optimal solutions for JSSP. The minimum makespan of a
which is an important step in the manufacture of products and is product equals the minimum cost of a product and the maxi-
frequently used in modeling production processes. mum profit for the company. So that, the main aim is effective
Due to the manufacturing industry having a primary role in scheduling in JSSP.
human life and the job-shop scheduling problem (JSSP) being a Metaheuristic algorithms are enhanced versions of heuristic
challenging task in this area, increasing the solution quality in algorithms. Particle Swarm Optimization(PSO) is a population-
JSSP is important for companies in the production environment based evolutionary computation method that is proposed by
to be able to compete in the market. Kennedy and Eberhart [4]. PSO has been inspired by the social
In the JSSP, which is a type of scheduling problem known behaviors of the birds’ flock in order to search for food. It is
as one of the combinatorial NP-hard optimization problems, the a popular algorithm used by many researchers because it is
purpose is to reach optimum scheduling solutions by allocating easy to apply to optimization problems [5–9]. The Tree-Seed
limited resources to various tasks [1]. Classic n by m JSSP can Algorithm (TSA) [10] is a metaheuristic algorithm that is pro-
be defined as a generalization of the flow shop problem. Unlike posed for solving continuous optimization problems by Kiran. The
the flow shop problem, each job has a unique route or order of main inspiration source of TSA is the relationship between trees
visiting specific machines. The aim of JSSP is to create a schedule and their seeds. TSA is a widely used metaheuristic algorithm
that will minimize the completion time (makespan) for n jobs [11–18]. Karaboğa proposed an Artificial Bee Colony Algorithm
that have several operations that can be operated by only one (ABC) firstly in 2005 [19]. Foraging behaviors of the three types
of bees (workers, onlookers, and scout bees) have been modeled.
∗ Correspondence to: Selçuk University, Faculty of Technology, Department Due to the effectiveness of the ABC, it has been utilized for
of Electrical and Electronics Engineering, 42075, Konya, Turkey. different kinds of optimization problems by lots of researchers
E-mail address: asahman@selcuk.edu.tr (M.A. Şahman). [20–22]. Seasonal immigration processes of the superorganisms

https://doi.org/10.1016/j.knosys.2022.109711
0950-7051/© 2022 Elsevier B.V. All rights reserved.
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

are modeled by a Differential Search Algorithm (DSA) [23]. DSA new candidate solutions. Unlike previous studies, the maximum
has been implemented for many areas, such as power systems FEs number varies depending on the size of the problem.
[24], motor designs [25], and some real-life engineering problems The approaches used to discretize the search space are very
[26]. The Grey Wolf Optimizer (GWO) mimics hunting mecha- important for the performance of metaheuristic algorithms de-
nisms (searching for prey, encircling prey, and attacking prey) pending on the problem and the quality of the solution to be ob-
and the leadership hierarchy (alpha, beta, delta, and omega) of tained. In this study, 3 proposed different encoding schemes (ran-
the grey wolves in nature [27]. GWO has been employed on dom key, smallest position value, and ranked-over value encoding
electrical conductivity [28], data clustering [29], PID controller schemes) are used in experimental analyses for each metaheuris-
[30], hydropower reservoir operation [31], and so forth. Whale tic algorithm. In addition, the comparison of these three encoding
Optimization Algorithm (WOA) has been proposed by Mirjalili schemes was carried out for the first time in this study. Exper-
and Lewis [32]. The inspiration of the WOA is the bubble-net
imental analyses are presented to investigate the performance
hunting strategy of humpback whales. Recently, WOA has been
of 3 proposed different encoding schemes for each metaheuristic
applied to some optimization problems such as the soil wind
algorithm.
erodability [33], green vehicle routing [34], radial distribution
The remainder of the paper is organized as follows. Section 2
system [35]. Lévy Flight Distribution Algorithm (LFD) is inspired
by Lévy flight random walk [36]. The Artificial Algae Algorithm presented the description and the mathematical model of JSSP.
(AAA) is a bio-inspired metaheuristic algorithm that mimics the The encoding schemes and details of AAA were given in Sec-
living behaviors of the microalgae [37]. In many areas, due to tions 3 and 4, respectively. The experimental setup and experi-
the effectiveness and robustness of the AAA algorithm have been mental results and discussions were presented in Sections 5 and
applied [38–44]. 6, respectively. Finally, the work is concluded in Section 7.
Scheduling is a discrete optimization problem. Most meta-
heuristic algorithms are firstly proposed for continuous opti- 2. Job-shop scheduling problem
mization problems. In the literature, some of such emerging
metaheuristic algorithms or hybrid versions of them that are
given above, thanks to the discretization of these algorithms, have In the n × m JSSP, there are n jobs J = {J1 , J2 , J3 , . . . , Jn }
been used to address the JSSP. In the first studies, well-known to be completed and m machines M = {M1 , M2 , M3 , . . . , Mm }
metaheuristic algorithms were applied to the JSSP benchmark in which operations are carried out. Each job has predefined m
problems, such as simulated annealing (SA) [45–47], Taboo Search operations Oij = {Oi1 , Oi2 , Oi3 , . . . , Oim } (i = 1, 2, . . . , n; j =
(TS) [48–50], Genetic Algorithm (GA) [51–53], and PSO [54]. In 1, 2, . . . , m) that are strictly ordered and completed in a fixed
order to enhance the local search ability, metaheuristic algo- time without an interruption in a given machine. Each machine
rithms have benefited from other heuristic algorithms or their can merely operate an operation at the time. It is assumed that
specific mechanisms and proposed as a hybrid. Lian et al. [55] consecutive operations of the same job are processed on different
proposed A Similar Particle Swarm Optimization Algorithm (SP- machines. The total number of the operations is stated as O =
SOA) to solve JSSP by using GA operators. Sha and Hsu [56] 0, 1, 2, . . . , m × n, m × n + 1, where O and m × n + 1 are first
presented a Hybrid Particle Swarm Optimization (HPSO) that used and last operation that have no duration time. The makespan of
Giffler and Thompson’s heuristic [57] and also applied tabu search operations can be expressed as Cij = {Ci1 , Ci2 , Ci3 , . . . , Cim } (i =
to improve solution quality. Tasgetiren et al. [58] investigated 1, 2, . . . , n; j = 1, 2, . . . , m) and total makespan is max(Cim ) or
on JSSP solution quality of PSO and Differential Evolution(DE) Cmax . Explanations of the notations are given as follows;
on JSSP combined with Variable Neighborhood Search (VNS). A
hybrid algorithm of SA and PSO (HPSO) was introduced by Xia and n : The number of the jobs,
Wu [59]. The neighborhood structure and cooling schedule of SA m : The number of the machines,
have been employed for avoiding individuals being trapped to the Ci : The completion time of operation
local optimum. Lin et al. [60] combined a Multi-type Individual i, (i = 0, 1, 2, . . . , m × n + 1),
Enhancement scheme (MIE), Random-Key (RK) encoding scheme, Ti : Processing time of operation i on given machine,
and PSO called MPSO. pi : All predecessor operations of operation i,
It has been observed that some of the state-of-the-art meta- A(t) : The set of operations processed at time t,
heuristic approaches proposed recently are also used for JSSP Rij : The token of operation j requiring on machine m
such as Biogeography-Based Optimization (BBO) [61], Bat Algo- be processed
rithm (BA) [62,63], Firefly Algorithm (FA) [64,65], Cuckoo search
(COA) [66], Teaching–learning-based Optimization (TLBO) [67,
68], Grey Wolf Optimizer (GWO) [69], and Whale Optimization The mathematical model of JSSP is given as follows:
Algorithm (WOA) [26]. minimize Cmax (1)
This work differs from the existing ones at least in three
aspects: Subject to:
In the literature, as far as we know, there is no study about Cs ≤ Ci − Ti , i = 0, 1, 2, . . . , m × n + 1 ; s ∈ pi (2)
solving JSSP with TSA, DSA, LFD, and AAA metaheuristic algo- ∑
rithms. The performance of these algorithms is carried out for Rij ≤ 1 , m ∈ M ; t ≥ 0 (3)
j∈A(t)
JSSP for the first time. So in this study, comprehensive compar-
isons have been conducted in order to investigate the metaheuris- Ci ≥ 0, i = 0, 1, 2, . . . , m × n + 1 (4)
tics on solving the JSSPs.
When the literature is examined, it can be seen that the algo- In the mathematical model of JSSP, Eq. (1) represents the objec-
rithms proposed for JSSP use iteration number as the stopping cri- tive function of JSSP that is minimized total makespan. Eq. (2)
terion. In this study, the stopping criteria are determined as max- is the constraint that represents the relations of the operations.
imum function evaluation numbers (FEs) for a fair comparison. Eq. (3) is the constraint that indicates one operation can solely
Especially, it should be used FEs for algorithms such as TSA and be processed on one machine at a time. Eq. (4) provides the
AAA, which call many times objective functions while generating completion time to be non-negative.
2
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

Table 1 Table 2
RK processing from the real values to operation sequence for 2 × 3 JSSP. SPV processing from the real values to operation sequence for 2 × 3 JSSP.
Real values 0.25 0.97 0.86 0.50 0.70 0.15 Real values 0.25 0.97 0.86 0.50 0.70 0.15
Integer series (ϕ ) 2 6 5 3 4 1 Integer series (ϕ ) 2 6 5 3 4 1
Job indexes 1 1 2 2 1 2 Job indexes 1 2 2 1 2 1
Operation sequence O1,1 O1,2 O2,1 O2,2 O1,3 O2,3 Operation sequence O1,1 O2,1 O2,2 O1,2 O2,3 O1,3

Table 3
3. Encoding scheme approaches from continuous to discrete ROV processing from the real values to operation sequence for 2 × 3 JSSP.
Integer positions 1 2 3 4 5 6
Job indexes pattern 1 2 2 1 2 1
In the literature, many metaheuristic algorithms have been Operation sequence pattern O1,1 O2,1 O2,2 O1,2 O2,3 O1,3
proposed in order to solve continuous optimization problems.
Real values 0.25 0.97 0.86 0.50 0.70 0.15
There are many different optimization problems that exist in Integer series (ϕ ) 2 6 5 3 4 1
the literature, such as binary optimization, integer optimization, Job indexes 2 1 2 2 1 1
discrete optimization, etc. By researchers, according to the opti- Operation sequence O2,1 O1,1 O2,2 O2,3 O1,2 O1,3
mization problem kinds, the algorithms that are proposed to solve
continuous optimization problems either are changed as struc-
tural for solving different types of optimization problems. Addi-
tionally, encoding schemes are used without structural chancing.
ϕk − 1
( )
To solve the JSSPs without modification on the metaheuristic ⃗ = floor
O + 1, k = 1, 2, . . . , n × m (6)
algorithms, three different ranked-based encoding schemes have m
been proposed for mapping from continuous to discrete. ⃗ is operation sequence, ϕk is the kth value of the integer
where, O
For n jobs and m machines JSSP, the dimension length of
series, and m is the machine number of the JSSP. Two phases of
artificial agents of each metaheuristic algorithm is n×m. Positions
the encoding scheme are the same with RK. The integer series
of the artificial agents are represented with real numbers. These
are determined as {2,6,5,3,4,1} and Eq. (6) is utilized to obtain
n × m real numbers are transformed into the integer series
job indexes as given in Table 2. According to calculations, first,
to imitate the permutation sequence of JSSP by using encoding
fourth, and sixth integer values are determined as the operations
scheme approaches. Converted integer series are represented as
O1,1 , O1,2 , O1,3 of the first job, and similarly, the rest of the integer
operation sequences.
values are determined as the operations O2,1 , O2,2 , O2,3 of the
Transforming real value series to operation sequence of the
second job.
JSSP is carried out in three-phase. In the first phase, real values
are sorted in ascending order. In the second phase, the ranking of
3.3. Ranked-over value encoding scheme
the real values is expressed in ascending order of integer series.
The process steps in the first two stages are the same, but the
As a different encoding scheme, a ranked-over value encoding
last phase, where the operation sequence of the JSSP is found,
scheme (ROV) is proposed by Liu et al. [26]. This method uses a
is different for RK, SPV, and ROV encoding scheme approaches.
job index pattern in order to determine the operation sequence,
Details of the third phase of these approaches are given below.
unlike other coding schemes. Two phases of the encoding scheme
are the same with RK and SPV. In the third phase, this encoding
3.1. Random key encoding scheme
scheme is benefited from the job index pattern that is determined
in the initialization step and is constant, i.e., not changed accord-
A random-key encoding scheme (RK) that can be utilized to
ing to metaheuristic algorithms’ processes. Let us suppose, in the
convert continuous search space to discrete search space in JSSP
initialization step, job indexes pattern is decided as {1,2,2,1,2,1}
is proposed by Bean [70]. In the third phase of the RK, integer
with respect to ascending ordered integer positions {1,2,3,4,5,6}.
series are assigned to job indexes via Eq. (5) given below. When the position of the artificial agent value is updated in
⃗ = (ϕk mod n) + 1, k = 1, 2, . . . , n × m
O (5) the process of the metaheuristic algorithm, these real-valued
positions are ordered and founded integer series. And then, in
where, O ⃗ is operation sequence, ϕk is the kth value of the integer accordance with the founded integer series are paired through
series, and n is the job number of the JSSP. Let us suppose, we the job indexes pattern, i.e. {2,6,5,3,4,1} integer series given in
have 2 jobs, and each job has 3 operations (Table 1), and n × m Table 3 is matched with {2,1,2,2,1,1} operation sequence. In this
real values are {0.25, 0.97, 0.86, 0.50, 0.70, 0.15} respectively. To manner, each different integer series can be encoded properly.
obtain the integer series, real values are sorted in ascending order,
i.e., 0.15 is the smallest value, it is ranked to 1, and so on. The 4. Artificial algae algorithm
integer series is determined as {2,6,5,3,4,1}. The first value in the
integer series is 2, and when Eq. (5) is used for the first integer The artificial algae algorithm(AAA) that simulates the real
value, the job index is calculated as (2 mod 2) + 1 = 1. Calculated algae is a bio-inspired algorithm proposed by Uymaz et al. [37].
similarly, first, second and fifth integer values are determined as Algae can conform to their environment and can be dominant
the operations (O1,1 , O1,2 , O1,3 ) of the first job and the rest of the species in the appropriate environment. Since algae need a light
integer series are determined as the operations (O2,1 , O2,2 , O2,3 ) of source to photosynthesize, algae move towards to light source by
the second job. By RK, the operation sequence is mapped properly helical swimming movement. Algae reproduce by mitosis. When
and guaranteed the feasibility. the mitotic division is occurring, both of the algae live together,
and these algae split in half again, and so on. The algae colony is
3.2. Smallest position value encoding scheme composed thanks to much mitotic division. The algae colony lives
like a single cell. Under unsuitable environmental conditions, part
Tasgetiren et al. [58] are proposed the smallest position value of the algae in the colony can die, or due to external impacts, the
encoding scheme (SPV). In the third phase of the SPV, the op- algae colony can be divided. These divided parts can survive their
eration sequence of JSSP is calculated with Eq. (6) given below. lives and create a new colony. In order to stay alive, ensure their
3
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

development, and live as a community, three main processes, the The algal colonies perform a helical motion with their flagella.
evolution process, adaptation process, and helical motion process, However, two effects prevent the forward movement of algae
of the real algae, are imitated by AAA. colonies. These are friction surface and gravity.
The size of algae colonies affects the helical velocity of move-
4.1. Evolutionary process ment in the water. Larger algae colonies move slowly. On the
contrary, smaller algal colonies move faster. Thus, local search
Under environmental conditions where adequate nutrients capability (exploitation) is provided. Algae colonies move with
and light exist, an artificial algal colony like real algae can be their available energy, which is calculated according to their
grown and reproduce itself by mitosis. Otherwise, it can be live a size. Each movement will decrease somewhat in energy levels
short time, and then it will die. A growth kinetic value is utilized depending on the energy loss parameter (ψ ). As a result of the
so as to determine how much artificial algae would reproduce. helical motion of the algae colony, if the colony is displaced to
The growth kinetic is calculated by Eqs. (7) and (8). a better place in the solution space, its energy will be reduced
by half the value of the energy loss parameter (ψ/2). Otherwise,
µmax .f t (⃗x)
µ
⃗t = ( t ) (7) its energy will be reduced as much as the energy loss parameter

G
+ f t (⃗x) (ψ ). The algae colony will continue the helical movement until
2
its energy is zero and will be displaced in the solution space.
G ⃗t + µ
⃗ t +1 = G ⃗t (8) Another important factor affecting the movement of algae
colonies is the friction surface. The friction surface is considered
where, µ is the specific growth rate vector of the algal colony, to be spherical and is calculated with Eq. (15).
and µmax is the maximum growth rate. f is the objective function √
value of the algal colony, x is the algae cell vector of the algal 3 3.G

colony, and G is the size of the algal colony. In the evolutionary τ⃗ = 2.π.( )2 (15)
process, one cell of the smallest algal colony dies, and the dy- 4.π
ing cell is replaced with a cell of the largest algal colony (Eqs. where, τ is the friction surface of the artificial algal colony.
(9)–(11)). Helical motion occurs in three dimensions. Therefore, three
( ) sizes in the algae colony are chosen randomly. The first move-
ρ t = max G
⃗t (9) ment is linear. The second and third movement is angular. The
( ) friction surface, together with the distance from the light source
σ t = min G
⃗t (10) of the algal colony, will affect the distance of movement. Three-
dimensional movement is calculated with Eqs. (16)–(18).
σ t +1
m = ρ , m = 1, 2, . . . , D
t
m (11)
xti,+m1 = xti,m + xtj,m − xti,m . (∆ − τi ) .p
( )
(16)
where, ρ represents the biggest algal colony and σ the smallest
xti,+k 1 xti,k xtj,k xti,k . (∆ − τi ) . cos α
( )
algal colony at the time t. D is the dimension of the algal colony = + − (17)
(dimension of the problem), and m is the randomly selected algal
xti,+l 1 xti,l xtj,l xti,l . (∆ − τi ) . sin β
( )
= + − (18)
cell in the colony.
The following Eq. (12) is used to calculate the energies of algae where, , ,xti,m xti,k xti,k
are randomly selected ith algae cells which
colonies (E) by normalizing their size. represent x, y, and z coordinates. j indicates different algal colony
that is determined as the light source. ∆ is shear force, τi is
E⃗ t = norm((rank(G
⃗ t ))2 ) (12)
friction surface of the ith algal colony, α, β ∈ [0 − 2π ], p is a
random number in the range [0–1].
4.2. Adaptation process
The pseudo-code of AAA for JSSP is presented in Algorithm 1.
The algal colonies should adapt to the environment in the best
5. Experimental setup
way. In the adaptation process, the algal colony that is survived
but not much growth is attempted to emulate the biggest algal In order to compare eight algorithms and three encoding
colony. The algal colony that is unable to grow is determined in schemes, 48 well-known JSSP benchmarks are used in experi-
accordance with the starvation level. In the initialization phase, mental studies. Benchmarks consist of four classes. These classes
the starvation level of the whole algal colony is considered as are cited from Fisher and Thompson (ft06-ft20) [71], Adams et al.
zero. After the helical movement of the algal colony, the light (abz5-abz9) [72], Lawrence (la01-la30) [73], Applegate and Cook
quality may increase or decrease. The starvation level of growing (orb01-orb10) [74]. Name, size (n × m), dimension(D), and the
algae colonies does not change. However, in case of insufficient best-known solution (BKS) of the JSSP benchmarks are given in
light exposure of algae cells, starvation increase in the algal Table 4.
colony. The artificial algae colony with the highest starvation In this study, eight metaheuristic algorithms (Particle Swarm
value is determined by Eq. (13) and according to the adaptation Optimization (PSO), Tree-Seed Algorithm (TSA), Artificial Bee
parameter (Ap ) that is selected in the range of [0,1], the adaptation Colony (ABC), Differential Search Algorithm (DSA), Grey Wolf Op-
process is applied by Eq. (14). timizer (GWO), Whale Optimization Algorithm (WOA), Lévy flight
distribution algorithm (LFD), Artificial Algae Algorithm (AAA)) are
( )
ϑ t = max A⃗ t (13)
used to solve JSSP by using different encoding schemes (RK, SPV,
ϑ t +1 = ϑ t + ρ t − ϑ t . rand and ROV). The peculiar parameter settings of all algorithms which
( )
(14)
are carried out to solve this JSSP are given in Table 5.
where, A is the starvation values vector of the algal colonies, and
ϑ is the highest starvation value of the algal colonies at time t. 6. Experimental results and discussions

4.3. Helical movement process In this study, basic versions of eight state-of-the-art meta-
heuristic algorithms are addressed on JSSP. Basic versions of
The algal colonies need to swim with the help of flagella these metaheuristic algorithms have been proposed for continu-
towards the water surface to survive and be closer to light. ous problems. However, JSSPs are discrete optimization problems.
4
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

Table 4
Properties of the benchmark problems.
Name Size (n × m) Dimensions (D) BKS Name Size (n × m) Dimensions (D) BKS
ft06 6 × 6 36 55 la17 10 × 10 100 784
ft10 10 × 10 100 930 la18 10 × 10 100 848
ft20 20 × 5 100 1165 la19 10 × 10 100 842
abz5 10 × 10 100 1234 la20 10 × 10 100 902
abz6 10 × 10 100 943 la21 15 × 10 150 1046
abz7 20 × 15 300 656 la22 15 × 10 150 927
abz8 20 × 15 300 665 la23 15 × 10 150 1032
abz9 20 × 15 300 679 la24 15 × 10 150 935
la01 10 × 5 50 666 la25 15 × 10 150 977
la02 10 × 5 50 655 la26 20 × 10 200 1218
la03 10 × 5 50 597 la27 20 × 10 200 1235
la04 10 × 5 50 590 la28 20 × 10 200 1216
la05 10 × 5 50 593 la29 20 × 10 200 1152
la06 15 × 5 75 926 la30 20 × 10 200 1355
la07 15 × 5 75 890 orb01 10 × 10 100 1059
la08 15 × 5 75 863 orb02 10 × 10 100 888
la09 15 × 5 75 951 orb03 10 × 10 100 1005
la10 15 × 5 75 958 orb04 10 × 10 100 1005
la11 20 × 5 100 1222 orb05 10 × 10 100 887
la12 20 × 5 100 1039 orb06 10 × 10 100 1010
la13 20 × 5 100 1150 orb07 10 × 10 100 397
la14 20 × 5 100 1292 orb08 10 × 10 100 899
la15 20 × 5 100 1207 orb09 10 × 10 100 934
la16 10 × 10 100 945 orb10 10 × 10 100 944

Therefore, encoding schemes are needed to transform continu- Tables 6–8.


ous problems into discrete problems. In this experimental study,
[ ]
EBSA − BKS
three different encoding schemes (RK, SPV, ROV) are applied for RPD = 100 ∗ ( ) (19)
BKS
each metaheuristic algorithm and tested on 48 JSSP benchmark
In other studies in the literature, the RPD value was calculated
problems. While testing algorithms, they were run 30 times for
by using each algorithm’s best solution (EBS). In this case, it is
each different situation in a notebook PC with an Intel Core i7
known that if the algorithm obtains a BKS value even only once,
machine, CPU 2.6 GHz, 16 GB of RAM, Windows 10 (64) system, it will increase its RPD value. For this reason, it is thought that
®
and MATLAB software. After 30 runs, for each algorithm tested it will be more appropriate to calculate the RPD value using the
on JSSP benchmarks, respectively, the best solutions’ average EBSA values obtained after 30 runs in this study. At the bottom of
(EBSA), standard deviation, hit numbers, and relative percentage Tables 6–8, the total number of hits, the total number of optimally
deviation (RPD, which is calculated by Eq. (19)) are reported in solved problems at least once (NIS), and average RPD values are
5
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

Table 5
Parameter settings of algorithms.
Parameters Algorithms
PSO TSA ABC DSA GWO WOA LFD AAA
Population size (N) 40 40 40 40 40 40 40 40
Dimension size (D) n × m n × m n × m n × m n × m n × m n × m n × m
dmin, dmax −5, 5 −5, 5 −5, 5 −5, 5 −5, 5 −5, 5 −5, 5 −5, 5
Max_FEs D*1000 D*1000 D*1000 D*1000 D*1000 D*1000 D*1000 D*1000
C1 2 NA NA NA NA NA NA NA
C2 2 NA NA NA NA NA NA NA
maxLimit NA NA 40 NA NA NA NA NA
L NA N*0.1 NA NA NA NA NA NA
U NA N*0.25 NA NA NA NA NA NA
ST NA 0.2 NA NA NA NA NA NA
b NA NA NA NA NA 1 NA NA

a NA NA NA NA 2 to 0 2 to 0 NA NA
CSV NA NA NA NA NA NA 0.5 NA
K NA NA NA NA NA NA NA 2
Le NA NA NA NA NA NA NA 0.3
Ap NA NA NA NA NA NA NA 0.5

given. In addition, the best solutions obtained are indicated in Wilcoxon signed-rank test [76] were used. It is possible to com-
bold. pare the column effects in a two-way layout with Friedman’s test,
When the RK transformation scheme was used (Table 6), the which is a nonparametric test.
AAA method achieved the best result with an average RPD of 8.66, When the Friedman rank results in Table 10 are examined,
while the TSA method showed the worst performance with an the AAA algorithm using SPV encoding scheme showed the best
RPD of 24.43. While the AAA method obtained BKS values in 318 performance with an average rank result of 2.61. The best perfor-
out of 1440 runs, the TSA method could only reach 74 Sum Hit mance is followed by the AAA algorithm using the ROV encoding
values. When the SPV transformation scheme was used (Table 7), scheme and the RK encoding scheme, with average rank values of
the AAA method achieved the best result with an average RPD of 3.35 and 3.39, respectively. The worst performance was obtained
7.33, while the PSO method showed the worst performance with with the PSO algorithm using the SPV encoding scheme with the
an RPD of 34.68. While the AAA method obtained BKS values in average rank value of 23.18.
339 out of 1440 runs, the PSO method could only reach 5 Sum The overall Friedman’s test results of all algorithms and en-
Hit values. coding schemes are given in Fig. 1. As can be seen from Fig. 1,
When the ROV transformation scheme was used (Table 8), the it is clearly seen that the AAA algorithm obtained the best result
AAA method achieved the best result with an average RPD of 8.49, for all encoding schemes. When the encoding schemes for each
while the PSO method showed the worst performance with an method are examined separately, it is seen that the use of PSO
RPD of 25.17. While the AAA method obtained BKS values in 309 and WOA methods with the RK encoding scheme and the use
out of 1440 runs, the PSO method could only reach 67 Sum Hit of SPV encoding scheme with all other methods produce better
values. results. The results obtained by using the ROV encoding scheme
When Tables 6–8 are examined in detail, it is seen that all
seem to have poor performance for all algorithms.
algorithms successfully solved the ft06, la05, la06, la10, and La14
The Friedman rank test was performed for each method and
problems, and they have difficulties in solving the abz7, abz8,
encoding scheme (a total of 24 combinations). According to the
abz9, and la29 problems. Generally, it is seen that AAA achieves
Friedman rank test results, the best encoding scheme was deter-
better quality results than other algorithms under the same con-
mined (PSO-RK, TSA-SPV, ABC-SPV, DSA-SPV, GWO-SPV, WOA-
ditions for all encoding schemes.
RK, LFD-SPV, and AAA-SPV) for each algorithm. Due to the sig-
In order to evaluate the effect of used encoding schemes on the
nificant performance of the AAA method using the SPV encoding
metaheuristic algorithms, the results in Tables 6–8 are compiled
scheme in the Friedman rank test, the Wilcoxon signed-rank test,
and given in Table 12. When Table 12 is examined in terms of
which is another nonparametric test, was performed separately
Avg. RPD values, it is seen that the RK encoding scheme is more
compatible with PSO, DSA, and WOA algorithms, while the SPV for each algorithm used determined encoding scheme by compar-
encoding scheme is more compatible with TSA, ABC, GWO, LFD, ing with the AAA-SPV method. Wilcoxon signed-rank test results
and AAA algorithms. When all algorithms and encoding schemes applied with a significance level value of 0.05 are presented in
are examined together, the best result was obtained with a 7.33 Table 11. Column S in Table 11 is expressed as ‘‘1’’ if the AAA-SPV
Avg. RPD value when the SPV encoding scheme was used in the method is successful, ‘‘−1’’ if the other method is successful, and
AAA algorithm. ‘‘0’’ if the results of the two methods are statistically equal. The
Elapsed times of algorithms in encoding schemes are listed total count of winners, losers, and equals is given in the last row
in Table 9, and the best solutions for each encoding scheme are of Table 11. When Table 11 is examined, it is seen that there is no
shown in bold. According to the elapsed time analysis made ac- significant difference with the better results of the other methods
cording to the algorithms and encoding schemes used in Table 9, in comparisons with the AAA-SPV method, and therefore there is
AAA’s elapsed time is the least in RK and ROV encoding schemes, no ‘‘−1’’ value in Table 11. It is clearly seen that the AAA-SPV
while the elapsed time of the ABC algorithm is the least in the method obtains meaningful results compared to each algorithm
SPV encoding scheme. The longest elapsed time in all encoding used determined encoding scheme.
schemes is obtained with the LFD algorithm. In accordance with Since it is very difficult to show the convergence graph of all
these results, it can be said that the AAA algorithm is superior JSSPs that are used for experimental studies, only the conver-
to other algorithms in terms of elapsed time as well as solution gence graphs of the largest dimensional ones in the four classes of
quality. benchmarks (Ft20 (20 × 5), Orb10 (10 × 10), La30 (20 × 10),
In order to statistically evaluate obtained results for all al- Abz9 (20 × 15)) are given separately for each encoding scheme
gorithms and encoding schemes, the Friedman’s test [75] and (Figs. 2–4). Convergence graphs are drawn by taking the average
6
M.A. Şahman and S. Korkmaz
Table 6
Comparison of the 8 algorithms used RK mapping method on JSSP benchmarks.
Problem PSO TSA ABC DSA GWO WOA LFD AAA

Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD

ft06 55.43 0.77 22 0.79 57.27 0.98 1 4.12 57.00 1.36 8 3.64 55.20 0.55 26 0.36 57.70 0.84 0 4.91 55.93 1.44 20 1.70 57.07 1.28 6 3.76 55.03 0.18 29 0.06
ft10 1192.07 22.78 0 28.18 1215.50 20.91 0 30.70 1172.43 23.84 0 26.07 1155.60 15.75 0 24.26 1207.23 47.47 0 29.81 1148.20 23.58 0 23.46 1202.73 24.29 0 29.33 1044.30 24.37 0 12.29
ft20 1618.63 19.27 0 38.94 1551.07 21.50 0 33.14 1489.70 29.30 0 27.87 1482.37 21.69 0 27.24 1461.57 83.06 0 25.46 1478.40 45.23 0 26.90 1533.70 22.27 0 31.65 1317.70 32.09 0 13.11
abz5 1370.30 31.94 0 11.05 1473.13 20.37 0 19.38 1440.87 24.09 0 16.76 1346.97 22.85 0 9.15 1467.37 31.26 0 18.91 1340.70 24.01 0 8.65 1448.40 35.04 0 17.37 1313.33 27.08 0 6.43
abz6 1042.37 19.60 0 10.54 1121.17 22.63 0 18.89 1097.40 17.95 0 16.37 1033.77 14.54 0 9.63 1113.83 31.10 0 18.12 1027.67 19.31 0 8.98 1105.57 20.71 0 17.24 998.60 17.66 0 5.90
abz7 853.97 19.06 0 30.18 986.07 14.08 0 50.32 954.70 19.75 0 45.53 819.87 16.86 0 24.98 985.10 17.60 0 50.17 812.10 11.01 0 23.80 925.17 43.34 0 41.03 787.67 12.33 0 20.07
abz8 898.13 24.01 0 35.06 1018.70 12.14 0 53.19 992.07 16.18 0 49.18 846.83 22.42 0 27.34 1020.13 10.31 0 53.40 830.30 10.64 0 24.86 964.93 43.43 0 45.10 804.07 14.23 0 20.91
abz9 956.40 18.98 0 40.85 1031.20 13.96 0 51.87 1003.50 17.76 0 47.79 901.37 21.75 0 32.75 1029.63 11.68 0 51.64 891.07 17.41 0 31.23 1000.03 24.89 0 47.28 828.87 13.21 0 22.07
la01 687.77 11.94 1 3.27 712.50 13.48 0 6.98 698.47 12.18 0 4.87 682.63 8.34 2 2.50 698.03 20.91 4 4.81 685.33 14.25 2 2.90 705.87 10.88 0 5.99 667.07 2.89 24 0.16
la02 749.70 14.46 0 14.46 751.47 13.55 0 14.73 737.57 15.43 0 12.61 733.47 12.43 0 11.98 741.97 19.12 0 13.28 744.30 22.64 0 13.63 747.97 13.08 0 14.19 703.67 16.16 0 7.43
la03 665.10 12.60 0 11.41 691.57 11.04 0 15.84 673.10 11.70 0 12.75 666.17 9.89 0 11.59 678.57 17.80 0 13.66 667.97 16.58 0 11.89 679.43 13.02 0 13.81 640.23 8.82 0 7.24
la04 631.23 5.88 0 6.99 675.10 11.96 0 14.42 653.43 13.44 0 10.75 629.57 5.39 0 6.71 648.63 18.98 0 9.94 638.87 18.67 0 8.28 652.07 15.81 0 10.52 620.60 7.54 0 5.19
la05 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.10 0.55 29 0.02 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.00 0.00 30 0.00
la06 933.13 10.43 17 0.77 956.63 10.72 0 3.31 940.30 10.31 4 1.54 926.73 2.33 26 0.08 951.60 14.93 3 2.76 928.67 7.70 21 0.29 946.70 13.48 2 2.24 926.00 0.00 30 0.00
la07 988.87 21.36 0 11.11 1001.87 13.14 0 12.57 973.37 16.60 0 9.37 961.97 16.82 0 8.09 981.03 24.27 0 10.23 970.27 17.17 0 9.02 992.50 16.12 0 11.52 903.90 12.38 4 1.56
la08 917.60 19.05 0 6.33 951.27 16.57 0 10.23 926.70 16.12 0 7.38 900.93 17.53 0 4.40 935.50 30.43 2 8.40 901.10 12.29 0 4.41 932.67 17.28 0 8.07 867.37 9.11 22 0.51
la09 958.37 8.88 13 0.77 988.53 14.65 1 3.95 971.53 15.66 3 2.16 951.00 0.00 30 0.00 981.43 18.63 5 3.20 951.30 1.64 29 0.03 983.80 15.87 1 3.45 951.00 0.00 30 0.00
la10 958.00 0.00 30 0.00 962.20 5.07 13 0.44 958.00 0.00 30 0.00 958.00 0.00 30 0.00 960.07 4.19 20 0.22 958.00 0.00 30 0.00 959.00 2.17 24 0.10 958.00 0.00 30 0.00
la11 1241.03 18.01 5 1.56 1288.10 16.04 0 5.41 1258.87 17.06 1 3.02 1228.00 11.57 21 0.49 1282.30 18.02 0 4.93 1222.60 2.93 28 0.05 1277.97 23.65 1 4.58 1222.07 0.37 29 0.01
la12 1075.20 20.47 1 3.48 1097.83 14.47 0 5.66 1064.30 16.38 1 2.44 1051.27 11.98 8 1.18 1084.47 28.38 4 4.38 1053.80 17.73 12 1.42 1090.30 19.64 1 4.94 1039.00 0.00 30 0.00
la13 1171.83 16.70 2 1.90 1219.97 17.21 0 6.08 1191.10 18.66 2 3.57 1155.33 10.17 21 0.46 1202.40 26.62 4 4.56 1151.17 4.49 28 0.10 1211.70 21.86 0 5.37 1150.00 0.00 30 0.00
la14 1292.00 0.00 30 0.00 1292.13 0.73 29 0.01 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.20 0.81 28 0.02 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00
la15 1401.73 22.33 0 16.13 1413.93 18.34 0 17.14 1376.40 19.06 0 14.03 1363.37 19.23 0 12.95 1388.77 42.09 0 15.06 1359.37 27.38 0 12.62 1396.20 35.17 0 15.68 1250.70 20.41 0 3.62
7

la16 1119.00 14.60 0 18.41 1119.50 21.11 0 18.47 1093.17 21.41 0 15.68 1086.63 17.27 0 14.99 1111.03 28.60 0 17.57 1072.10 27.45 0 13.45 1115.40 20.27 0 18.03 1005.63 17.42 0 6.42
la17 865.93 14.60 0 10.45 937.37 17.28 0 19.56 910.23 22.19 0 16.10 861.17 11.53 0 9.84 924.83 24.25 0 17.96 847.20 8.75 0 8.06 911.73 17.10 0 16.29 821.77 12.42 0 4.82
la18 952.70 18.46 0 12.35 1021.73 14.78 0 20.49 998.53 16.42 0 17.75 934.23 14.64 0 10.17 1015.27 16.58 0 19.72 926.30 12.56 0 9.23 1005.20 20.44 0 18.54 896.53 14.38 0 5.72
la19 1024.43 23.14 0 21.67 1042.20 18.37 0 23.78 1017.73 15.64 0 20.87 1006.67 13.66 0 19.56 1042.23 27.83 0 23.78 972.03 21.17 0 15.44 1039.50 25.60 0 23.46 914.40 25.05 0 8.60
la20 1050.60 26.05 0 16.47 1089.50 22.56 0 20.79 1045.07 21.55 0 15.86 1036.03 22.67 0 14.86 1079.37 37.00 0 19.66 1001.07 31.75 0 10.98 1091.20 21.06 0 20.98 951.43 22.50 0 5.48
la21 1323.13 37.04 0 26.49 1423.43 21.58 0 36.08 1370.27 22.04 0 31.00 1237.10 21.40 0 18.27 1418.97 28.53 0 35.66 1261.43 34.01 0 20.60 1384.60 47.21 0 32.37 1189.10 22.99 0 13.68
la22 1188.43 23.06 0 28.20 1284.17 24.66 0 38.53 1240.57 21.43 0 33.83 1125.13 22.78 0 21.37 1273.70 24.59 0 37.40 1134.43 30.75 0 22.38 1241.47 33.60 0 33.92 1054.60 23.69 0 13.76
la23 1233.93 26.92 0 19.57 1353.40 26.51 0 31.14 1308.80 22.78 0 26.82 1169.73 19.76 0 13.35 1345.63 34.99 0 30.39 1171.47 8.86 0 13.51 1339.23 37.27 0 29.77 1113.30 23.49 0 7.88
la24 1196.90 23.96 0 28.01 1284.27 20.83 0 37.35 1243.80 22.88 0 33.03 1135.63 21.04 0 21.46 1287.00 17.98 0 37.65 1128.67 35.79 0 20.71 1265.83 37.14 0 35.38 1066.73 17.72 0 14.09
la25 1226.90 27.52 0 25.58 1323.37 15.68 0 35.45 1285.03 21.42 0 31.53 1176.07 17.24 0 20.38 1309.07 36.12 0 33.99 1179.07 23.20 0 20.68 1305.03 34.69 0 33.58 1107.57 20.10 0 13.36
la26 1560.67 26.08 0 28.13 1703.30 26.51 0 39.84 1652.93 27.99 0 35.71 1450.57 27.44 0 19.09 1699.20 42.01 0 39.51 1468.90 27.69 0 20.60 1646.37 49.80 0 35.17 1373.57 26.10 0 12.77
la27 1609.43 43.29 0 30.32 1754.07 30.32 0 42.03 1706.40 26.36 0 38.17 1514.27 30.70 0 22.61 1752.57 30.57 0 41.91 1529.47 32.71 0 23.84 1705.93 45.16 0 38.13 1430.70 20.72 0 15.85
la28 1559.53 37.62 0 28.25 1708.60 25.03 0 40.51 1644.00 30.92 0 35.20 1451.50 38.19 0 19.37 1696.90 22.78 0 39.55 1463.70 24.37 0 20.37 1654.27 57.23 0 36.04 1388.27 21.11 0 14.17
la29 1518.23 28.84 0 31.79 1680.50 23.83 0 45.88 1625.20 33.05 0 41.08 1428.00 30.61 0 23.96 1679.97 18.47 0 45.83 1451.00 22.47 0 25.95 1589.00 39.50 0 37.93 1372.63 23.91 0 19.15
la30 1663.07 35.48 0 22.74 1799.03 25.73 0 32.77 1735.60 24.60 0 28.09 1594.20 30.68 0 17.65 1799.97 29.57 0 32.84 1591.13 22.07 0 17.43 1739.87 47.24 0 28.40 1493.63 21.82 0 10.23
orb01 1358.60 25.93 0 28.29 1403.43 17.10 0 32.52 1360.73 33.76 0 28.49 1322.77 17.80 0 24.91 1397.07 27.48 0 31.92 1317.57 26.49 0 24.42 1385.50 26.90 0 30.83 1221.17 25.15 0 15.31
orb02 1057.93 19.69 0 19.14 1120.93 18.34 0 26.23 1078.13 21.12 0 21.41 1020.83 19.08 0 14.96 1100.70 32.24 0 23.95 1025.40 21.39 0 15.47 1090.77 23.96 0 22.83 971.30 21.32 0 9.38
orb03 1413.80 28.45 0 40.68 1391.73 26.08 0 38.48 1339.67 33.46 0 33.30 1321.03 28.55 0 31.45 1370.43 57.09 0 36.36 1312.63 52.09 0 30.61 1374.53 36.82 0 36.77 1177.97 27.66 0 17.21
orb04 1203.03 26.97 0 19.70 1252.10 20.45 0 24.59 1213.33 25.69 0 20.73 1186.70 26.70 0 18.08 1247.80 25.61 0 24.16 1162.17 25.25 0 15.64 1244.50 28.36 0 23.83 1089.53 17.38 0 8.41
orb05 1182.73 20.14 0 33.34 1174.53 21.02 0 32.42 1127.23 22.40 0 27.08 1114.00 22.61 0 25.59 1143.80 63.66 0 28.95 1110.90 41.25 0 25.24 1154.83 32.69 0 30.20 996.40 20.73 0 12.33
orb06 1296.70 28.52 0 28.39 1351.23 21.85 0 33.79 1310.47 34.02 0 29.75 1260.33 23.37 0 24.79 1344.00 38.42 0 33.07 1268.07 29.68 0 25.55 1331.13 30.52 0 31.80 1151.67 27.62 0 14.03

Knowledge-Based Systems 256 (2022) 109711


orb07 484.27 10.73 0 21.98 507.97 7.77 0 27.95 491.07 10.33 0 23.69 474.23 8.29 0 19.45 503.27 10.87 0 26.77 466.80 11.79 0 17.58 499.47 15.49 0 25.81 436.57 10.26 0 9.97
orb08 1255.73 19.93 0 39.68 1225.83 21.22 0 36.36 1187.73 23.09 0 32.12 1149.90 21.06 0 27.91 1217.73 38.60 0 35.45 1161.73 36.42 0 29.23 1210.10 30.25 0 34.61 1040.10 29.33 0 15.70
orb09 1136.37 18.37 0 21.67 1173.47 21.69 0 25.64 1132.63 21.05 0 21.27 1119.80 18.59 0 19.89 1150.10 48.45 0 23.14 1103.07 27.28 0 18.10 1154.30 24.76 0 23.59 1016.50 18.66 0 8.83
orb10 1220.87 30.31 0 29.33 1260.90 24.58 0 33.57 1201.43 25.10 0 27.27 1171.27 22.43 0 24.07 1232.30 48.24 0 30.54 1167.60 35.28 0 23.69 1229.90 30.95 0 30.29 1059.17 25.37 0 12.20

Sum Hit 151 74 109 224 99 230 95 318


NIS 10 5 9 10 9 10 8 12
Avg. RPD 18.84 24.43 20.91 14.88 23.24 14.65 22.12 8.66
M.A. Şahman and S. Korkmaz
Table 7
Comparison of the 8 algorithms used SPV mapping method on JSSP benchmarks.
Problem PSO TSA ABC DSA GWO WOA LFD AAA

Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD

ft06 58.60 1.10 1 6.55 57.20 1.16 3 4.00 57.67 1.27 3 4.85 56.40 1.19 11 2.55 57.60 1.54 6 4.73 58.63 1.50 2 6.61 57.17 1.34 7 3.94 55.07 0.37 29 0.12
ft10 1286.27 23.45 0 38.31 1197.40 23.47 0 28.75 1174.87 32.35 0 26.33 1113.70 27.66 0 19.75 1063.83 38.00 0 14.39 1146.80 38.27 0 23.31 1175.63 31.32 0 26.41 1025.13 17.09 0 10.23
ft20 1519.17 38.34 0 30.40 1531.90 20.02 0 31.49 1482.10 25.05 0 27.22 1373.43 25.38 0 17.89 1381.53 69.10 0 18.59 1423.33 36.25 0 22.17 1466.60 32.73 0 25.89 1315.43 25.87 0 12.91
abz5 1588.93 42.63 0 28.76 1474.40 21.21 0 19.48 1433.87 31.73 0 16.20 1374.57 21.19 0 11.39 1344.37 35.83 0 8.94 1420.37 52.26 0 15.10 1417.63 32.52 0 14.88 1278.90 13.49 0 3.64
abz6 1238.03 39.39 0 31.29 1116.87 15.58 0 18.44 1090.97 19.88 0 15.69 1041.37 19.17 0 10.43 1031.07 36.72 0 9.34 1079.43 34.04 0 14.47 1091.23 33.99 0 15.72 986.17 15.82 0 4.58
abz7 1124.07 24.77 0 71.35 978.50 15.19 0 49.16 939.60 15.42 0 43.23 890.43 13.68 0 35.74 856.20 46.31 0 30.52 919.20 28.65 0 40.12 953.17 26.03 0 45.30 776.30 14.73 0 18.34
abz8 1170.10 28.08 0 75.95 1015.63 9.30 0 52.73 968.77 17.51 0 45.68 918.63 11.98 0 38.14 902.33 60.17 0 35.69 945.73 36.17 0 42.22 984.23 19.50 0 48.01 797.67 16.28 0 19.95
abz9 1181.73 25.06 0 74.04 1021.77 14.06 0 50.48 983.90 26.51 0 44.90 943.30 14.33 0 38.92 910.17 63.49 0 34.05 954.03 28.53 0 40.51 1001.13 28.80 0 47.44 814.60 13.43 0 19.97
la01 755.27 17.52 0 13.40 703.00 11.78 0 5.56 699.40 14.77 1 5.02 679.03 10.45 4 1.96 676.43 13.68 12 1.57 699.00 21.78 1 4.95 702.97 12.87 0 5.55 666.23 1.28 29 0.04
la02 782.97 14.58 0 19.54 748.50 10.04 0 14.27 737.73 18.50 0 12.63 719.17 11.83 0 9.80 710.53 21.03 0 8.48 741.77 21.79 0 13.25 749.47 9.47 0 14.42 685.07 12.31 1 4.59
la03 710.57 15.69 0 19.02 680.23 9.91 0 13.94 677.73 10.41 0 13.52 656.37 10.61 0 9.94 653.57 19.56 0 9.48 676.50 21.95 0 13.32 679.50 14.07 0 13.82 629.87 6.23 0 5.51
la04 718.30 21.04 0 21.75 655.43 14.16 0 11.09 653.10 17.06 0 10.69 632.47 8.36 0 7.20 623.77 8.59 0 5.72 659.33 28.19 0 11.75 659.10 17.37 0 11.71 608.27 6.33 0 3.10
la05 607.60 10.53 4 2.46 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.00 0.00 30 0.00
la06 1005.97 23.00 0 8.64 953.07 9.70 0 2.92 935.13 10.13 8 0.99 926.10 0.55 29 0.01 932.47 12.59 23 0.70 931.57 8.74 18 0.60 936.03 10.33 7 1.08 926.00 0.00 30 0.00
la07 1035.67 20.77 0 16.37 989.23 12.93 0 11.15 970.10 17.29 0 9.00 941.07 12.78 0 5.74 926.27 31.04 1 4.07 953.90 23.92 0 7.18 978.93 18.93 0 9.99 897.50 8.10 9 0.84
la08 1033.83 25.97 0 19.80 948.80 13.88 0 9.94 921.80 19.90 0 6.81 896.47 13.96 1 3.88 887.90 31.26 9 2.89 912.90 24.06 1 5.78 926.20 19.61 0 7.32 863.00 0.00 30 0.00
la09 1069.17 22.84 0 12.43 985.13 10.50 0 3.59 964.23 13.13 8 1.39 954.33 6.56 20 0.35 953.07 6.21 26 0.22 965.43 22.57 12 1.52 974.13 17.20 4 2.43 951.00 0.00 30 0.00
la10 1018.73 27.79 0 6.34 958.90 1.95 22 0.09 958.00 0.00 30 0.00 958.00 0.00 30 0.00 958.00 0.00 30 0.00 959.60 4.40 26 0.17 959.80 6.00 27 0.19 958.00 0.00 30 0.00
la11 1402.83 30.68 0 14.80 1291.93 16.28 0 5.72 1252.53 15.32 1 2.50 1235.73 11.57 6 1.12 1238.40 28.64 20 1.34 1238.17 17.25 8 1.32 1261.77 21.96 2 3.25 1222.00 0.00 30 0.00
la12 1213.00 28.97 0 16.75 1092.93 15.78 0 5.19 1058.73 16.74 6 1.90 1044.57 6.83 13 0.54 1048.40 18.52 21 0.90 1053.57 18.86 10 1.40 1070.30 21.88 2 3.01 1039.00 0.00 30 0.00
la13 1344.67 33.94 0 16.93 1215.23 19.24 0 5.67 1185.33 17.40 1 3.07 1159.27 8.58 8 0.81 1168.63 27.73 15 1.62 1172.60 27.89 7 1.97 1193.03 21.04 1 3.74 1150.00 0.00 30 0.00
la14 1374.97 32.46 0 6.42 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00
la15 1521.90 39.19 0 26.09 1403.77 19.88 0 16.30 1368.23 23.55 0 13.36 1339.37 16.29 0 10.97 1320.13 65.49 0 9.37 1355.07 30.01 0 12.27 1384.27 31.61 0 14.69 1238.90 20.09 1 2.64
8

la16 1190.97 26.77 0 26.03 1114.70 17.65 0 17.96 1093.70 21.22 0 15.74 1038.00 21.27 0 9.84 1034.77 26.70 0 9.50 1078.63 35.94 0 14.14 1085.67 28.56 0 14.89 989.50 9.57 0 4.71
la17 1017.60 35.14 0 29.80 923.40 11.85 0 17.78 909.17 20.47 0 15.97 868.13 13.16 0 10.73 846.67 34.43 0 7.99 902.90 36.37 0 15.17 901.10 21.77 0 14.94 806.47 10.68 0 2.87
la18 1120.20 33.27 0 32.10 1018.13 14.20 0 20.06 990.50 18.82 0 16.80 952.73 16.83 0 12.35 925.93 26.54 0 9.19 985.00 30.57 0 16.16 989.27 20.27 0 16.66 886.20 14.58 0 4.50
la19 1129.70 35.36 0 34.17 1031.13 17.40 0 22.46 1011.03 24.53 0 20.08 961.20 24.97 0 14.16 949.87 34.52 0 12.81 1001.93 40.33 0 18.99 1009.63 30.27 0 19.91 899.00 15.77 0 6.77
la20 1168.03 27.07 0 29.49 1074.40 16.18 0 19.11 1062.40 20.49 0 17.78 1009.40 18.50 0 11.91 971.73 29.46 0 7.73 1059.33 35.96 0 17.44 1049.27 34.04 0 16.33 934.83 14.14 0 3.64
la21 1593.23 41.63 0 52.32 1411.83 24.65 0 34.97 1352.77 28.51 0 29.33 1310.33 22.71 0 25.27 1278.17 82.63 0 22.20 1339.43 46.96 0 28.05 1352.80 35.43 0 29.33 1167.80 14.91 0 11.64
la22 1430.27 35.07 0 54.29 1270.27 24.65 0 37.03 1232.17 26.97 0 32.92 1159.20 22.36 0 25.05 1125.13 66.03 0 21.37 1202.10 44.86 0 29.68 1209.90 44.72 0 30.52 1052.27 24.15 0 13.51
la23 1538.43 42.46 0 49.07 1346.73 21.92 0 30.50 1289.50 32.86 0 24.95 1226.40 26.52 0 18.84 1180.37 68.46 0 14.38 1265.43 52.75 0 22.62 1287.57 42.48 0 24.76 1097.43 17.50 0 6.34
la24 1458.40 31.03 0 55.98 1275.90 17.77 0 36.46 1227.03 28.38 0 31.23 1173.03 16.81 0 25.46 1120.30 56.50 0 19.82 1195.43 34.90 0 27.85 1227.27 35.19 0 31.26 1044.93 16.08 0 11.76
la25 1510.00 45.90 0 54.55 1314.13 23.62 0 34.51 1252.53 27.30 0 28.20 1210.43 20.71 0 23.89 1187.07 64.30 0 21.50 1254.23 36.58 0 28.38 1270.57 39.59 0 30.05 1096.40 23.51 0 12.22
la26 1941.63 43.09 0 59.41 1693.97 22.62 0 39.08 1617.47 37.23 0 32.80 1547.17 21.44 0 27.03 1525.97 109.21 0 25.28 1594.07 48.55 0 30.88 1637.37 48.23 0 34.43 1356.23 32.46 0 11.35
la27 1997.53 67.07 0 61.74 1754.17 24.95 0 42.04 1672.87 37.72 0 35.45 1616.00 18.01 0 30.85 1589.97 104.23 0 28.74 1637.57 44.67 0 32.60 1700.67 39.50 0 37.71 1411.10 24.38 0 14.26
la28 1939.33 52.11 0 59.48 1697.70 25.39 0 39.61 1613.53 43.28 0 32.69 1552.40 30.62 0 27.66 1502.03 89.75 0 23.52 1574.50 48.72 0 29.48 1620.53 53.60 0 33.27 1371.07 25.46 0 12.75
la29 1902.70 46.24 0 65.16 1666.93 22.83 0 44.70 1590.60 31.53 0 38.07 1538.33 22.72 0 33.54 1504.27 100.16 0 30.58 1584.27 57.88 0 37.52 1608.43 40.30 0 39.62 1361.20 24.04 0 18.16
la30 2051.97 57.42 0 51.44 1782.80 27.03 0 31.57 1715.47 38.29 0 26.60 1659.87 19.23 0 22.50 1621.70 95.95 0 19.68 1684.83 36.69 0 24.34 1737.07 41.49 0 28.20 1476.33 28.03 0 8.95
orb01 1456.73 28.80 0 37.56 1383.30 18.74 0 30.62 1348.83 27.11 0 27.37 1289.27 26.62 0 21.74 1252.33 47.29 0 18.26 1336.70 45.28 0 26.22 1349.33 40.05 0 27.42 1182.17 24.41 0 11.63
orb02 1230.33 27.82 0 38.55 1096.00 17.80 0 23.42 1081.23 21.43 0 21.76 1030.27 11.93 0 16.02 999.20 35.44 0 12.52 1052.73 32.43 0 18.55 1085.63 29.35 0 22.26 952.77 16.06 0 7.29
orb03 1381.90 32.44 0 37.50 1373.53 26.18 0 36.67 1344.40 33.91 0 33.77 1274.93 30.63 0 26.86 1225.33 47.22 0 21.92 1304.53 39.78 0 29.80 1331.60 35.23 0 32.50 1150.37 22.29 0 14.46
orb04 1321.80 32.60 0 31.52 1244.03 19.89 0 23.78 1217.83 24.81 0 21.18 1160.33 21.21 0 15.46 1118.87 26.54 0 11.33 1203.90 38.47 0 19.79 1215.93 31.34 0 20.99 1077.60 19.45 0 7.22
orb05 1210.80 35.60 0 36.51 1151.30 18.76 0 29.80 1128.70 28.21 0 27.25 1042.43 25.25 0 17.52 1023.23 31.25 0 15.36 1093.40 48.18 0 23.27 1123.63 29.81 0 26.68 974.10 23.79 0 9.82
orb06 1429.00 32.90 0 41.49 1324.77 20.35 0 31.17 1299.53 30.07 0 28.67 1229.83 31.19 0 21.77 1186.33 52.00 0 17.46 1298.43 44.15 0 28.56 1298.43 39.25 0 28.56 1135.00 26.53 0 12.38

Knowledge-Based Systems 256 (2022) 109711


orb07 557.70 13.58 0 40.48 498.33 8.46 0 25.52 491.17 12.47 0 23.72 464.93 11.16 0 17.11 447.00 16.23 0 12.59 482.27 13.03 0 21.48 493.03 15.78 0 24.19 428.20 8.72 0 7.86
orb08 1158.17 33.00 0 28.83 1200.17 21.13 0 33.50 1175.80 30.40 0 30.79 1095.47 21.55 0 21.85 1069.43 37.01 0 18.96 1143.90 41.64 0 27.24 1166.70 31.19 0 29.78 1021.83 20.13 0 13.66
orb09 1261.87 39.96 0 35.10 1151.03 19.31 0 23.24 1137.40 25.77 0 21.78 1070.93 21.63 0 14.66 1044.10 37.00 0 11.79 1111.13 31.62 0 18.97 1121.50 28.80 0 20.07 1003.63 17.01 0 7.46
orb10 1365.13 29.32 0 44.61 1217.83 28.31 0 29.01 1199.77 32.63 0 27.09 1126.17 30.51 0 19.30 1069.33 39.24 0 13.28 1175.57 53.19 0 24.53 1191.33 41.82 0 26.20 1039.80 24.89 0 10.15

Sum Hit 5 85 118 182 223 145 110 339


NIS 2 4 10 11 12 11 9 14
Avg. RPD 34.68 23.22 20.15 15.39 13.13 18.58 20.40 7.33
M.A. Şahman and S. Korkmaz
Table 8
Comparison of the 8 algorithms used ROV mapping method on JSSP benchmarks.
Problem PSO TSA ABC DSA GWO WOA LFD AAA

Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD Avg. Std. Hit RPD

ft06 57.90 1.37 3 5.27 57.37 1.03 3 4.30 57.10 0.99 3 3.82 57.10 1.12 4 3.82 57.20 1.45 7 4.00 57.77 1.25 3 5.03 57.47 0.90 2 4.48 55.30 0.70 25 0.55
ft10 1221.03 27.29 0 31.29 1215.73 24.97 0 30.72 1167.50 22.68 0 25.54 1157.13 13.88 0 24.42 1212.63 23.54 0 30.39 1152.37 35.06 0 23.91 1196.43 30.55 0 28.65 1049.13 23.36 0 12.81
ft20 1559.47 39.30 0 33.86 1537.30 25.14 0 31.96 1484.70 27.42 0 27.44 1467.50 22.68 0 25.97 1492.00 63.45 0 28.07 1468.37 55.15 0 26.04 1515.27 40.15 0 30.07 1310.57 29.51 0 12.49
abz5 1491.07 41.16 0 20.83 1479.20 29.43 0 19.87 1439.43 24.44 0 16.65 1408.53 18.88 0 14.14 1476.70 27.14 0 19.67 1410.07 42.31 0 14.27 1463.20 31.49 0 18.57 1299.73 22.38 0 5.33
abz6 1124.37 23.42 0 19.23 1115.50 18.12 0 18.29 1093.27 22.03 0 15.93 1075.27 16.91 0 14.03 1114.63 26.24 0 18.20 1076.83 25.75 0 14.19 1111.87 21.99 0 17.91 992.37 19.26 0 5.24
abz7 993.73 19.63 0 51.48 986.60 11.62 0 50.40 957.80 15.26 0 46.01 931.20 13.51 0 41.95 988.43 10.00 0 50.68 925.53 28.63 0 41.09 970.63 17.90 0 47.96 785.63 14.44 0 19.76
abz8 1029.57 18.28 0 54.82 1016.13 17.55 0 52.80 989.93 14.82 0 48.86 953.80 18.51 0 43.43 1015.77 17.01 0 52.75 941.53 21.38 0 41.58 996.43 18.66 0 49.84 805.67 12.97 0 21.15
abz9 1043.80 23.40 0 53.73 1032.57 16.80 0 52.07 1001.37 18.98 0 47.48 968.70 13.87 0 42.67 1031.17 15.22 0 51.87 963.70 27.85 0 41.93 1017.63 19.21 0 49.87 821.93 13.96 0 21.05
la01 710.77 19.12 0 6.72 713.37 9.54 0 7.11 694.03 11.51 0 4.21 690.13 14.27 2 3.62 692.07 19.36 5 3.91 695.30 17.20 2 4.40 703.63 14.77 1 5.65 669.63 6.35 20 0.55
la02 752.00 20.79 0 14.81 749.43 15.00 0 14.42 737.57 11.27 0 12.61 730.80 12.50 0 11.57 740.30 15.35 0 13.02 739.40 28.99 0 12.89 748.70 16.36 0 14.31 696.07 13.21 0 6.27
la03 683.00 18.73 0 14.41 688.93 9.90 0 15.40 672.20 11.17 0 12.60 669.23 7.19 0 12.10 675.73 17.82 0 13.19 671.67 16.73 0 12.51 675.47 11.59 0 13.14 640.73 8.57 0 7.33
la04 676.93 20.21 0 14.73 673.50 15.00 0 14.15 649.60 15.55 0 10.10 646.33 11.75 0 9.55 657.70 23.87 0 11.47 649.27 25.88 0 10.05 653.57 17.35 0 10.77 617.07 7.83 0 4.59
la05 593.40 1.85 28 0.07 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.00 0.00 30 0.00 593.67 2.66 28 0.11 593.00 0.00 30 0.00 593.00 0.00 30 0.00
la06 962.47 21.94 1 3.94 954.67 12.76 1 3.10 935.03 9.73 9 0.98 932.87 8.28 13 0.74 955.00 14.65 1 3.13 936.43 14.01 15 1.13 948.80 15.32 2 2.46 926.00 0.00 30 0.00
la07 1005.23 24.35 0 12.95 1002.70 14.27 0 12.66 973.37 15.91 0 9.37 964.47 9.87 0 8.37 976.87 29.94 0 9.76 963.50 29.25 0 8.26 986.53 25.32 0 10.85 904.53 12.64 6 1.63
la08 954.30 27.84 0 10.58 951.60 16.24 0 10.27 925.10 17.33 0 7.20 912.87 15.77 1 5.78 932.10 27.69 0 8.01 918.20 27.36 0 6.40 939.53 18.49 0 8.87 868.37 9.23 18 0.62
la09 995.17 20.80 1 4.64 987.73 12.89 0 3.86 965.93 13.42 6 1.57 960.50 8.45 6 1.00 980.13 21.72 6 3.06 965.93 15.92 10 1.57 981.77 16.24 0 3.24 951.00 0.00 30 0.00
la10 967.03 11.20 13 0.94 961.33 4.75 17 0.35 958.00 0.00 30 0.00 958.00 0.00 30 0.00 960.07 4.96 23 0.22 959.73 5.82 27 0.18 960.23 4.61 22 0.23 958.00 0.00 30 0.00
la11 1290.20 23.76 0 5.58 1290.37 13.03 0 5.59 1253.83 16.79 0 2.61 1248.40 13.90 2 2.16 1285.20 23.06 2 5.17 1246.60 21.32 8 2.01 1285.23 23.00 0 5.17 1222.00 0.00 30 0.00
la12 1112.23 31.70 0 7.05 1099.20 13.46 0 5.79 1068.13 13.92 0 2.80 1057.60 8.70 1 1.79 1087.77 20.79 2 4.69 1053.83 20.01 15 1.43 1090.50 18.17 1 4.96 1039.00 0.00 30 0.00
la13 1224.87 30.91 0 6.51 1219.60 19.94 0 6.05 1186.13 16.89 0 3.14 1169.00 11.30 1 1.65 1209.80 24.45 0 5.20 1166.53 18.67 10 1.44 1205.83 21.99 0 4.86 1150.00 0.00 30 0.00
la14 1297.10 10.97 21 0.39 1292.23 1.10 28 0.02 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00 1292.00 0.00 30 0.00
la15 1424.10 32.04 0 17.99 1413.27 20.13 0 17.09 1371.80 25.47 0 13.65 1353.77 23.55 0 12.16 1402.10 27.29 0 16.16 1349.47 29.33 0 11.80 1394.77 28.90 0 15.56 1251.03 17.41 0 3.65
9

la16 1114.10 27.76 0 17.89 1124.03 15.44 0 18.95 1087.57 23.28 0 15.09 1067.80 17.70 0 12.99 1107.60 36.48 0 17.21 1064.57 28.59 0 12.65 1115.87 20.70 0 18.08 996.57 17.29 0 5.46
la17 943.27 36.22 0 20.31 938.10 14.66 0 19.66 905.67 22.34 0 15.52 891.97 14.97 0 13.77 929.10 22.81 0 18.51 894.30 22.97 0 14.07 928.57 21.46 0 18.44 819.57 15.93 0 4.54
la18 1023.30 26.10 0 20.67 1023.43 19.73 0 20.69 993.33 15.86 0 17.14 973.57 15.36 0 14.81 1018.00 23.37 0 20.05 980.40 22.61 0 15.61 1014.80 17.86 0 19.67 901.87 21.61 0 6.35
la19 1051.37 22.57 0 24.87 1047.10 17.95 0 24.36 1008.73 18.34 0 19.80 994.47 16.56 0 18.11 1035.30 24.57 0 22.96 996.50 23.96 0 18.35 1037.93 21.45 0 23.27 913.67 17.95 0 8.51
la20 1103.50 33.25 0 22.34 1090.17 20.26 0 20.86 1056.67 17.60 0 17.15 1036.73 25.04 0 14.94 1082.37 27.96 0 20.00 1038.90 38.49 0 15.18 1085.43 22.58 0 20.34 947.60 23.57 0 5.06
la21 1431.27 30.61 0 36.83 1427.10 18.82 0 36.43 1368.13 21.60 0 30.80 1342.50 24.45 0 28.35 1416.97 33.48 0 35.47 1338.67 44.53 0 27.98 1395.30 40.73 0 33.39 1187.40 20.36 0 13.52
la22 1284.97 29.32 0 38.62 1280.47 23.19 0 38.13 1228.60 25.38 0 32.54 1207.47 20.02 0 30.26 1284.47 18.71 0 38.56 1215.83 46.96 0 31.16 1264.17 35.35 0 36.37 1046.67 22.86 0 12.91
la23 1379.87 36.35 0 33.71 1357.50 23.98 0 31.54 1299.93 31.33 0 25.96 1269.17 22.40 0 22.98 1350.03 45.60 0 30.82 1270.87 46.91 0 23.15 1338.80 36.48 0 29.73 1111.57 25.56 0 7.71
la24 1309.07 31.08 0 40.01 1284.60 26.08 0 37.39 1240.33 28.35 0 32.66 1211.00 19.01 0 29.52 1280.27 29.31 0 36.93 1190.33 45.74 0 27.31 1262.00 36.57 0 34.97 1061.40 20.12 0 13.52
la25 1334.60 30.67 0 36.60 1323.53 22.56 0 35.47 1281.07 18.07 0 31.12 1243.80 25.16 0 27.31 1311.50 26.23 0 34.24 1240.87 36.55 0 27.01 1294.07 24.97 0 32.45 1101.53 14.75 0 12.75
la26 1710.80 48.09 0 40.46 1696.90 25.08 0 39.32 1627.70 28.31 0 33.64 1594.30 26.35 0 30.89 1705.70 27.19 0 40.04 1592.73 42.65 0 30.77 1672.60 37.45 0 37.32 1364.30 24.26 0 12.01
la27 1775.33 49.78 0 43.75 1754.23 28.62 0 42.04 1693.93 33.84 0 37.16 1649.60 27.85 0 33.57 1752.00 27.94 0 41.86 1634.17 52.26 0 32.32 1726.37 44.57 0 39.79 1424.47 19.98 0 15.34
la28 1721.27 40.37 0 41.55 1702.27 19.17 0 39.99 1651.00 27.53 0 35.77 1602.13 18.96 0 31.75 1700.20 25.27 0 39.82 1583.57 50.54 0 30.23 1676.90 31.75 0 37.90 1396.57 25.35 0 14.85
la29 1687.80 48.54 0 46.51 1679.50 24.61 0 45.79 1619.93 34.72 0 40.62 1576.43 23.47 0 36.84 1665.10 42.93 0 44.54 1573.70 50.23 0 36.61 1655.40 27.50 0 43.70 1366.40 25.77 0 18.61
la30 1803.67 29.49 0 33.11 1799.87 21.55 0 32.83 1726.60 33.81 0 27.42 1698.23 25.23 0 25.33 1785.90 30.22 0 31.80 1677.20 39.20 0 23.78 1762.10 42.56 0 30.04 1491.03 23.33 0 10.04
orb01 1405.90 27.43 0 32.76 1392.77 27.87 0 31.52 1353.47 23.88 0 27.81 1329.80 26.85 0 25.57 1391.07 22.58 0 31.36 1347.87 33.29 0 27.28 1383.20 32.30 0 30.61 1215.93 26.04 0 14.82
orb02 1126.43 28.04 0 26.85 1113.67 17.25 0 25.41 1073.33 20.39 0 20.87 1057.03 20.19 0 19.04 1103.57 22.91 0 24.28 1067.53 33.25 0 20.22 1106.67 26.59 0 24.62 978.53 21.47 0 10.20
orb03 1408.73 33.56 0 40.17 1393.30 19.80 0 38.64 1328.47 31.52 0 32.19 1304.53 27.91 0 29.80 1376.83 53.49 0 37.00 1317.70 58.72 0 31.11 1367.33 34.78 0 36.05 1171.10 30.77 0 16.53
orb04 1263.00 35.67 0 25.67 1253.97 19.68 0 24.77 1210.63 32.44 0 20.46 1195.70 21.05 0 18.98 1246.03 36.66 0 23.98 1204.23 31.04 0 19.82 1247.60 23.24 0 24.14 1094.23 19.08 0 8.88
orb05 1188.80 28.63 0 34.02 1178.70 23.77 0 32.89 1122.93 30.31 0 26.60 1102.10 29.34 0 24.25 1147.13 45.79 0 29.33 1094.67 41.81 0 23.41 1148.10 37.98 0 29.44 996.23 22.08 0 12.31
orb06 1360.80 41.52 0 34.73 1352.07 22.52 0 33.87 1302.07 28.19 0 28.92 1283.67 21.11 0 27.10 1316.80 46.35 0 30.38 1292.63 38.21 0 27.98 1341.17 27.17 0 32.79 1146.93 29.90 0 13.56

Knowledge-Based Systems 256 (2022) 109711


orb07 507.87 15.13 0 27.93 505.73 8.54 0 27.39 486.33 11.22 0 22.50 481.40 9.14 0 21.26 497.53 17.04 0 25.32 478.63 14.85 0 20.56 497.40 12.12 0 25.29 438.60 9.90 0 10.48
orb08 1231.47 31.32 0 36.98 1219.33 22.00 0 35.63 1183.33 19.38 0 31.63 1151.43 19.83 0 28.08 1202.77 40.56 0 33.79 1162.67 36.16 0 29.33 1197.80 33.10 0 33.24 1041.00 30.39 0 15.80
orb09 1178.27 34.67 0 26.15 1165.47 17.50 0 24.78 1127.03 20.18 0 20.67 1107.60 19.64 0 18.59 1147.90 33.62 0 22.90 1108.63 39.62 0 18.70 1146.20 31.11 0 22.72 1017.90 17.02 0 8.98
orb10 1264.57 31.58 0 33.96 1251.37 26.48 0 32.56 1197.37 33.92 0 26.84 1187.07 21.08 0 25.75 1224.73 62.84 0 29.74 1173.47 32.87 0 24.31 1230.97 40.03 0 30.40 1056.90 23.53 0 11.96

Sum Hit 67 79 108 120 106 148 88 309


NIS 6 5 6 11 9 10 7 12
Avg. RPD 25.17 24.32 20.49 18.56 23.20 18.56 22.75 8.49
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

Table 9
Comparison of elapsed times of the 8 algorithms used RK, SPV, ROV encoding scheme approaches for JSSP benchmarks.
Problem RK ElapsedTime (s) SPV ElapsedTime (s) ROV ElapsedTime (s)
PSO TSA ABC DSA GWO WOA LFD AAA PSO TSA ABC DSA GWO WOA LFD AAA PSO TSA ABC DSA GWO WOA LFD AAA
ft06 2.35 2.09 2.35 1.50 3.21 2.36 4.15 1.56 1.25 1.72 2.74 2.82 3.05 2.19 3.42 3.00 1.76 2.11 2.11 1.85 3.16 1.49 4.01 3.11
ft10 7.59 8.13 7.59 9.24 11.82 7.29 15.43 8.88 9.47 8.79 8.01 8.60 11.96 7.56 14.61 9.75 8.66 7.53 6.88 8.76 10.85 6.42 14.96 8.08
ft20 7.64 8.22 7.64 6.92 10.82 6.90 15.21 6.57 9.40 8.55 7.81 8.25 12.23 7.48 14.95 9.82 8.74 7.68 6.95 7.40 10.06 6.40 14.89 5.70
abz5 7.85 8.44 7.85 6.71 11.23 6.79 14.22 6.01 9.21 8.65 7.89 8.60 12.74 7.59 14.76 10.20 9.00 7.85 7.18 6.78 10.49 6.21 13.86 5.60
abz6 7.72 8.33 7.72 6.55 13.19 7.14 12.57 5.93 9.75 8.94 8.24 8.94 12.65 7.91 14.27 10.28 8.49 7.73 7.01 6.56 12.01 6.38 12.12 5.44
abz7 36.48 49.69 36.48 40.16 80.73 42.03 85.02 22.77 57.63 50.84 35.83 42.34 79.23 41.17 89.80 38.74 52.94 43.56 30.23 37.74 72.22 36.77 78.20 19.23
abz8 36.35 49.67 36.35 45.18 70.28 44.41 82.55 21.88 56.43 49.67 35.17 42.26 79.28 40.35 91.37 38.65 53.24 43.42 30.17 40.36 77.51 38.54 76.22 18.79
abz9 36.58 51.28 36.58 46.97 59.78 45.59 82.91 22.43 56.29 49.70 34.69 41.22 80.33 39.94 82.21 37.84 54.57 44.66 30.10 41.90 67.55 38.91 77.39 18.89
la01 3.31 3.26 3.31 3.28 3.09 3.52 4.36 2.52 3.63 3.36 3.53 3.39 4.38 3.19 4.43 4.52 3.29 3.04 3.06 3.72 3.61 3.21 4.25 2.31
la02 3.31 3.28 3.31 3.76 3.10 3.60 4.35 2.59 3.58 3.39 3.41 3.43 4.45 3.18 4.42 4.54 3.26 3.07 3.07 3.51 3.43 3.40 4.24 2.25
la03 3.33 3.27 3.33 3.97 2.99 3.53 4.34 2.56 3.54 3.33 3.57 3.43 4.56 3.14 4.40 4.57 3.23 3.10 3.03 3.47 3.38 3.25 4.27 2.24
la04 3.32 3.31 3.32 3.91 3.04 3.58 4.37 2.57 3.59 3.44 3.70 3.35 4.55 3.15 4.32 4.79 3.24 3.08 3.05 3.20 3.34 3.26 4.27 2.44
la05 3.32 3.30 3.32 3.91 2.99 3.27 4.34 2.55 3.51 3.35 3.56 3.65 4.47 3.15 4.37 4.90 3.25 3.05 3.04 3.52 3.36 3.26 4.28 2.54
la06 5.32 5.67 5.32 6.12 5.46 5.92 7.76 3.99 6.22 5.84 5.64 6.04 8.11 5.28 7.80 7.42 5.67 5.26 4.83 6.07 5.96 5.12 7.68 3.80
la07 5.46 5.75 5.46 6.48 5.46 5.96 7.80 3.92 6.15 5.97 5.55 6.01 7.97 5.58 8.04 7.53 5.70 5.35 4.97 6.06 6.41 5.62 7.75 3.82
la08 5.40 5.72 5.40 6.56 5.60 5.97 7.90 3.93 6.16 5.79 5.61 6.01 8.18 5.59 7.80 7.40 5.78 5.32 4.98 5.69 5.94 5.57 7.74 3.84
la09 5.45 5.80 5.45 6.37 5.40 5.91 7.74 3.98 6.15 5.98 5.70 5.85 8.14 5.50 7.84 7.48 5.74 5.37 5.01 6.00 5.96 5.33 7.68 3.84
la10 5.49 5.86 5.49 6.36 5.42 5.62 7.90 3.94 6.14 5.75 5.98 5.98 8.06 5.58 7.75 7.44 5.74 5.37 5.03 6.07 7.47 5.53 7.72 3.81
la11 7.95 8.91 7.95 7.90 8.95 6.91 12.37 5.39 9.61 9.00 8.43 9.12 12.41 7.86 12.06 10.20 9.40 8.11 7.08 9.12 12.35 8.08 12.12 5.17
la12 8.00 8.79 8.00 7.60 9.68 6.84 12.31 5.41 9.79 9.07 8.36 9.06 12.69 7.99 12.20 10.19 9.52 8.11 7.13 8.23 11.77 8.10 12.00 5.17
la13 7.97 8.79 7.97 6.43 11.91 6.17 12.11 5.47 9.80 8.95 8.32 8.80 12.51 7.88 12.34 10.24 9.48 8.22 7.14 7.20 11.06 7.50 12.08 5.19
la14 8.02 8.81 8.02 6.72 10.25 6.15 12.39 5.48 9.66 8.82 8.24 9.03 12.18 7.93 12.00 9.85 9.24 8.24 7.28 6.65 9.58 6.49 12.21 5.09
la15 8.10 8.81 8.10 6.93 9.88 6.47 12.24 5.46 9.57 9.02 8.26 8.99 11.64 7.76 12.38 10.05 9.17 8.16 7.35 6.07 9.92 6.34 11.97 5.18
la16 8.21 8.77 8.21 7.15 9.48 6.39 12.16 5.40 9.83 8.66 8.15 8.97 11.65 7.88 12.17 9.97 9.13 8.15 7.32 6.53 9.62 5.69 12.05 5.24
la17 8.19 8.68 8.19 8.10 8.93 6.84 12.44 5.46 10.00 8.80 8.26 8.65 12.90 7.82 12.10 9.91 9.31 8.14 7.36 6.39 9.99 5.71 11.96 5.16
la18 8.08 8.75 8.08 7.61 8.63 7.28 12.18 5.44 9.55 8.72 8.12 8.86 12.49 7.79 12.45 10.07 9.45 8.11 7.33 6.43 9.93 6.08 12.00 5.14
la19 8.10 8.63 8.10 6.83 8.96 6.49 12.18 5.44 9.95 8.74 8.12 8.92 12.42 7.81 12.16 10.17 9.45 8.13 7.47 7.36 9.81 5.75 11.86 5.15
la20 8.06 8.71 8.06 6.80 8.84 6.37 12.15 5.36 9.93 8.90 8.05 8.46 12.28 7.73 12.29 10.30 9.29 8.00 7.47 7.20 10.46 6.18 12.03 5.11
la21 13.93 16.12 13.93 12.11 17.25 11.40 24.22 8.69 18.03 15.87 13.52 15.56 23.79 13.37 23.68 16.30 17.80 14.78 12.38 10.65 19.46 11.63 23.07 8.19
la22 13.80 16.08 13.80 11.67 18.51 11.37 24.46 8.71 18.44 15.87 13.42 15.38 22.09 13.55 23.35 16.26 17.42 14.66 12.38 10.66 19.18 10.21 23.04 8.12
la23 13.69 16.19 13.69 17.05 20.68 12.98 23.70 8.74 17.99 15.86 13.50 15.62 14.87 13.56 23.52 16.51 17.76 14.75 12.27 10.83 17.23 10.06 23.12 8.01
la24 13.71 16.01 13.71 17.05 18.80 14.94 23.58 9.36 16.74 16.05 13.58 15.46 14.22 13.71 23.61 16.52 17.70 14.60 12.28 12.23 15.55 10.43 23.43 8.05
la25 13.60 16.17 13.60 13.97 16.81 14.47 23.79 8.91 17.78 15.93 13.54 15.48 12.94 13.65 23.33 15.86 17.50 14.69 12.16 16.10 12.76 10.09 23.09 8.11
la26 20.39 25.85 20.39 19.05 22.16 18.54 39.67 13.25 28.22 23.41 20.00 23.18 18.77 20.95 37.81 21.26 28.35 23.18 17.78 21.43 19.08 21.13 37.83 12.22
la27 20.54 25.77 20.54 18.60 19.61 18.49 42.27 13.17 28.46 25.21 20.09 20.85 16.28 21.11 37.90 23.13 24.25 23.20 17.89 17.11 17.51 19.32 37.15 11.66
la28 20.52 22.52 20.52 19.31 19.37 18.87 46.93 12.13 26.31 24.98 20.17 22.34 15.34 20.45 38.09 22.89 23.80 23.07 17.62 17.24 18.24 15.99 41.70 12.37
la29 20.53 21.44 20.53 20.01 18.88 19.11 47.19 12.37 15.55 25.20 18.72 22.60 15.51 19.18 37.67 23.07 20.04 23.07 17.75 16.64 17.04 16.39 44.71 11.02
la30 20.45 20.39 20.45 18.67 18.18 18.99 46.91 12.21 14.70 24.05 18.85 22.71 15.75 21.33 37.80 23.79 18.02 19.68 17.76 18.36 16.94 16.90 43.40 11.05
orb01 7.91 6.04 7.91 7.12 6.03 6.63 14.59 5.28 4.38 5.48 8.31 8.77 4.80 7.62 12.04 7.61 5.73 6.58 7.23 7.27 5.69 6.14 14.31 4.58
orb02 7.96 6.04 7.96 6.43 6.06 6.46 14.79 4.93 4.33 5.14 8.21 8.93 4.77 7.75 12.28 6.44 5.74 6.62 7.21 6.45 5.68 6.12 14.24 4.54
orb03 7.00 6.05 7.00 6.10 6.10 6.60 14.57 4.96 4.32 5.18 8.04 8.65 4.82 7.78 12.23 6.42 5.79 6.62 7.20 6.44 5.65 6.45 14.23 4.55
orb04 6.69 5.62 6.69 6.04 6.12 6.07 14.45 4.98 4.20 5.12 8.19 7.85 4.83 7.79 12.11 6.36 5.69 6.70 7.14 6.84 5.67 6.73 14.57 4.54
orb05 6.66 5.58 6.66 6.13 6.17 5.65 14.57 4.95 4.08 5.10 7.85 5.01 4.81 7.80 12.36 6.11 5.76 6.70 7.20 6.53 5.79 5.87 14.32 4.96
orb06 6.68 5.58 6.68 5.40 6.22 5.63 14.59 4.03 3.59 4.68 8.19 4.64 4.79 7.81 12.07 5.36 5.72 5.75 7.12 6.88 5.79 6.34 14.25 4.56
orb07 6.70 5.54 6.70 5.05 6.23 5.63 14.30 3.99 3.27 4.34 8.29 4.63 4.77 7.71 12.19 5.37 4.90 5.57 7.16 5.63 5.78 5.77 14.26 4.61
orb08 6.74 5.52 6.74 4.55 6.23 5.32 13.21 3.98 3.28 4.37 8.20 4.61 4.73 7.60 12.41 5.37 5.00 5.62 7.18 5.58 5.80 6.05 14.45 4.62
orb09 6.68 5.58 6.68 4.17 6.18 4.64 13.36 4.08 3.26 4.37 8.06 4.56 4.77 7.81 12.35 5.14 4.91 5.57 7.22 5.54 5.72 5.98 14.46 4.55
orb10 6.74 5.53 6.74 4.81 5.74 4.68 13.42 3.84 3.27 4.37 7.96 4.57 4.74 5.48 11.96 4.99 4.78 5.59 7.24 5.46 5.76 6.09 14.54 4.60
Average time (s) 10.58 11.92 10.58 10.61 13.55 10.25 20.42 6.99 12.42 11.92 10.74 11.47 14.48 10.87 19.53 11.97 12.34 11.14 9.50 10.08 13.49 9.46 19.50 6.50

Table 10
Friedman’s test of all results.
Problems RK SPV ROV
PSO TSA ABC DSA GWO WOA LFD AAA PSO TSA ABC DSA GWO WOA LFD AAA PSO TSA ABC DSA GWO WOA LFD AAA
ft06 5.95 13.75 12.73 4.97 15.98 8.22 13.10 4.42 20.22 13.85 15.95 10.02 15.83 19.48 13.42 4.38 17.05 14.52 13.07 13.00 13.57 16.47 14.57 5.50
ft10 15.93 19.28 12.63 9.77 18.38 9.12 17.45 2.47 23.93 16.53 13.07 5.83 3.78 9.47 12.88 1.62 19.50 19.13 11.55 10.00 18.57 9.92 16.27 2.92
ft20 23.93 20.10 12.93 12.10 12.07 12.03 18.40 2.42 17.05 18.45 12.32 4.90 5.90 7.43 10.82 2.20 20.90 19.03 12.43 10.38 14.18 11.52 16.28 2.22
abz5 7.27 18.68 14.83 5.65 17.97 5.27 15.77 3.43 23.90 18.73 13.58 8.00 5.57 12.60 11.98 1.37 20.13 19.33 14.40 11.08 19.00 11.50 17.48 2.47
abz6 7.23 18.88 14.95 6.42 17.92 5.47 16.17 2.72 24.00 18.33 13.60 7.30 6.52 12.07 14.05 1.75 19.65 18.05 14.15 11.20 17.78 11.72 17.65 2.43
abz7 6.68 19.83 14.53 4.70 20.02 4.37 12.12 2.47 24.00 18.45 12.20 8.50 7.23 10.82 14.72 1.57 20.73 19.83 14.90 11.35 20.20 11.52 17.02 2.25
abz8 6.98 19.57 15.37 4.98 20.08 3.97 13.20 2.10 24.00 19.15 12.45 8.23 8.47 10.52 14.43 1.93 21.25 19.42 15.38 10.95 19.32 9.88 16.02 2.35
abz9 9.07 19.60 15.15 5.47 19.43 4.98 14.75 2.45 24.00 17.87 12.32 7.98 7.02 9.47 14.57 1.48 20.78 19.88 14.30 10.38 19.47 10.07 17.35 2.17
la01 10.38 18.82 14.23 8.62 13.65 9.78 16.60 2.90 23.70 16.03 14.87 7.20 6.37 13.75 15.97 2.42 17.62 19.37 12.42 11.05 11.85 12.63 15.88 3.90
la02 16.05 17.07 12.45 10.65 13.63 14.27 15.53 4.22 23.05 16.20 12.37 6.67 5.68 13.25 16.27 1.78 17.05 16.22 11.90 10.08 13.27 13.02 16.45 2.88
la03 10.02 19.58 13.17 10.12 15.47 11.58 15.32 3.13 22.95 15.85 15.17 6.75 6.85 14.22 15.37 1.50 16.42 19.02 12.72 11.25 14.03 12.62 13.82 3.10
la04 7.60 19.75 14.50 7.13 12.87 9.78 14.13 4.00 23.77 15.22 14.77 8.05 5.17 15.43 15.72 1.42 19.48 19.48 13.40 12.87 14.93 12.60 14.65 3.28
la05 11.98 11.98 11.98 11.98 12.37 11.98 11.98 11.98 22.43 11.98 11.98 11.98 11.98 11.98 11.98 11.98 12.75 11.98 11.98 11.98 11.98 12.78 11.98 11.98
la06 9.63 19.50 13.90 6.22 17.13 7.43 16.07 5.33 23.68 18.35 11.55 5.57 8.68 9.00 12.10 5.33 19.28 18.57 11.60 10.03 18.47 10.73 16.50 5.33
la07 16.42 19.62 12.87 9.83 15.03 11.98 17.32 2.45 23.43 16.78 11.93 6.23 5.67 8.98 14.15 1.88 19.65 19.58 12.53 9.83 14.38 10.60 16.33 2.50
la08 11.58 19.30 13.58 8.38 16.35 7.88 15.20 2.57 24.00 18.93 12.60 7.23 6.45 10.67 13.75 1.90 18.90 19.45 13.48 10.88 15.28 11.83 17.03 2.75
la09 9.25 18.68 14.28 5.02 16.30 5.25 17.60 5.02 24.00 18.30 11.55 7.25 6.12 10.73 14.98 5.02 19.28 18.23 12.38 10.98 15.60 11.93 17.22 5.02
la10 10.40 16.92 10.40 10.40 14.12 10.40 12.52 10.40 23.83 13.30 10.40 10.40 10.40 11.98 11.62 10.40 17.23 15.45 10.40 10.40 13.07 11.73 13.43 10.40
la11 9.75 19.00 12.97 5.45 17.47 4.07 16.95 3.83 24.00 19.42 12.10 8.23 7.58 8.80 13.72 3.73 18.97 19.67 12.35 11.07 18.38 10.53 18.23 3.73
la12 14.33 19.25 12.10 8.20 15.85 8.60 17.60 3.63 24.00 18.28 10.18 6.43 6.32 8.80 13.28 3.63 20.15 19.72 13.07 10.32 16.72 8.42 17.48 3.63
la13 10.47 19.13 14.20 5.57 15.90 4.08 17.83 3.67 24.00 18.78 13.37 7.33 8.82 9.88 14.48 3.67 19.47 19.18 13.53 10.17 17.38 8.90 16.52 3.67
la14 11.77 12.15 11.77 11.77 12.52 11.77 11.77 11.77 24.00 11.77 11.77 11.77 11.77 11.77 11.77 11.77 15.23 12.53 11.77 11.77 11.77 11.77 11.77 11.77
la15 16.63 19.37 12.50 10.10 15.43 10.13 16.35 2.50 23.97 17.53 11.10 7.15 8.15 9.78 14.05 1.63 19.43 19.12 11.45 9.02 17.27 9.03 15.98 2.32
la16 18.50 18.32 13.05 11.85 17.20 9.58 17.17 2.98 23.67 17.60 13.18 5.35 5.43 10.40 12.13 1.58 16.77 19.65 11.88 8.70 16.50 8.55 17.52 2.43
la17 7.47 19.72 14.88 6.88 17.25 4.77 14.92 2.57 23.67 17.13 14.13 7.97 5.80 13.83 13.20 1.62 19.37 19.68 13.45 11.22 17.93 11.62 18.35 2.58
la18 7.93 19.42 15.08 5.90 18.27 4.95 16.20 2.35 24.00 19.20 13.33 7.72 5.15 12.73 12.87 1.55 19.28 19.52 13.48 10.78 18.30 11.63 17.52 2.83
la19 14.48 18.45 13.67 11.53 18.43 6.55 17.98 2.67 23.80 16.27 12.57 5.58 5.05 11.17 12.75 1.75 19.55 19.23 11.80 9.25 17.37 10.02 17.65 2.43
la20 12.12 18.82 11.18 10.00 17.10 6.22 19.12 2.70 23.72 15.93 14.02 6.60 4.37 13.90 12.10 1.63 20.00 18.98 12.62 10.00 17.43 10.98 17.92 2.55
la21 9.43 19.53 13.55 4.65 19.20 5.52 15.48 2.43 24.00 18.17 12.53 8.38 7.88 11.13 11.97 1.37 19.98 20.22 14.00 11.00 19.05 11.15 16.82 2.55
la22 9.45 19.75 14.95 5.33 18.98 5.83 14.73 2.20 24.00 17.87 13.62 6.97 6.15 10.58 11.92 2.13 19.55 19.53 13.32 10.98 20.20 12.53 17.38 2.03
la23 8.42 19.17 14.35 5.08 18.28 5.35 17.52 2.38 24.00 18.30 12.52 8.20 6.10 11.12 12.67 1.63 20.70 19.60 13.58 10.70 19.07 11.25 17.77 2.25
la24 9.98 19.15 14.40 5.70 19.60 5.43 16.82 2.65 24.00 18.60 12.65 7.78 5.37 9.90 13.07 1.38 21.17 19.02 14.03 11.05 19.23 9.78 16.80 2.43
la25 8.90 20.07 14.93 5.25 18.12 5.42 17.77 2.43 24.00 18.53 11.27 7.48 6.77 11.53 14.00 1.70 20.53 19.70 14.03 10.58 18.50 10.23 16.12 2.13
la26 8.27 19.92 14.98 4.60 19.57 5.05 14.85 2.47 24.00 18.85 12.32 7.60 8.65 10.62 14.07 1.50 20.07 19.23 12.90 10.45 20.28 10.48 17.08 2.20
la27 7.88 19.38 14.77 4.80 19.28 5.13 15.13 2.52 24.00 19.62 12.12 8.13 8.77 9.80 14.55 1.48 20.27 19.47 13.98 10.52 19.20 10.02 17.12 2.07
la28 9.13 20.52 14.30 4.67 19.20 5.08 15.28 2.33 24.00 19.37 12.17 8.20 7.10 9.53 12.62 1.42 20.58 19.52 14.87 11.10 19.13 10.18 17.20 2.50
la29 7.60 20.07 14.75 4.37 19.80 5.20 12.15 2.27 24.00 18.67 11.73 8.40 8.57 11.78 13.55 1.87 20.23 20.58 14.23 10.98 18.60 10.98 17.62 2.00
la30 8.53 19.87 14.18 5.27 19.90 5.17 14.92 2.33 24.00 18.78 12.33 7.95 7.93 10.13 14.43 1.60 20.50 20.27 13.47 11.05 18.97 9.77 16.58 2.07
orb01 13.35 19.98 13.67 8.60 19.27 8.18 17.32 2.62 23.67 16.88 11.77 5.80 4.47 10.48 12.32 1.47 20.03 18.48 12.68 9.63 18.12 11.92 16.72 2.58
orb02 10.70 20.43 13.52 5.68 17.40 6.52 15.55 2.55 24.00 16.48 14.17 6.90 4.75 10.18 14.93 1.43 20.50 19.45 12.88 10.37 17.62 12.23 18.60 3.15
orb03 21.62 19.18 12.47 10.42 17.38 10.02 17.08 2.60 17.73 16.78 13.48 6.28 4.68 8.85 11.78 1.50 20.73 19.72 11.20 8.65 18.03 11.18 16.12 2.50
orb04 11.50 19.02 12.87 9.45 18.03 6.67 17.67 2.45 23.80 17.62 13.40 6.53 3.77 10.77 13.40 1.63 19.77 19.15 12.38 10.35 17.55 11.78 17.83 2.62
orb05 20.48 19.15 12.80 10.93 15.93 11.05 16.50 2.75 22.17 15.95 12.73 5.27 4.10 9.60 12.38 1.58 20.60 19.82 12.23 10.07 15.98 9.32 15.98 2.62
orb06 11.67 19.65 14.30 7.87 18.60 8.55 17.05 2.37 23.83 15.83 12.40 5.88 4.28 12.80 12.50 1.97 19.92 19.47 12.85 10.02 15.42 12.15 18.08 2.55
orb07 11.45 20.12 14.07 8.45 18.15 7.02 17.25 2.48 24.00 16.67 14.17 6.58 4.03 11.05 14.98 1.60 19.22 19.60 12.63 10.47 16.65 10.03 16.48 2.85
orb08 23.12 20.10 14.63 9.22 19.13 10.88 18.03 2.47 10.92 16.70 12.77 5.20 3.83 9.72 11.83 1.78 20.18 19.22 13.63 9.70 17.17 11.15 15.85 2.77
orb09 14.52 20.13 13.73 11.40 16.95 9.50 17.28 2.47 23.70 16.53 14.30 5.73 4.55 10.35 11.97 1.63 19.82 19.02 12.58 9.58 16.12 9.87 15.73 2.53
orb10 15.95 20.58 13.07 9.25 17.45 9.45 16.48 2.67 24.00 15.27 12.77 6.00 3.33 10.72 12.15 1.87 20.20 19.40 12.75 11.25 16.18 9.83 16.58 2.80
Avg. rank 11.70 18.96 13.67 7.72 17.18 7.61 15.92 3.39 23.18 17.20 12.78 7.32 6.61 11.12 13.34 2.61 19.38 18.84 12.97 10.59 16.98 11.05 16.53 3.35

10
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

Table 11
Wilcoxon sign test results of algorithms according to the determined coding scheme.
Problems PSO-RK/AAA-SPV TSA-SPV/AAA-SPV ABC-SPV/AAA-SPV DSA-SPV/AAA-SPV GWO-SPV/AAA-SPV WOA-RK/AAA-SPV LFD-SPV/AAA-SPV
p_value S p_value S p_value S p_value S p_value S p_value S p_value S
ft06 5.47E−02 0 4.20E−06 1 4.26E−06 1 2.09E−04 1 1.44E−05 1 1.95E−03 1 2.08E−05 1
ft10 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 2.59E−05 1 1.73E−06 1 1.73E−06 1
ft20 1.73E−06 1 1.73E−06 1 1.73E−06 1 8.45E−06 1 1.13E−05 1 1.73E−06 1 1.73E−06 1
abz5 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1
abz6 1.73E−06 1 1.72E−06 1 1.73E−06 1 1.73E−06 1 3.17E−06 1 2.72E−06 1 1.92E−06 1
abz7 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.72E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1
abz8 1.73E−06 1 1.72E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 3.50E−06 1 1.72E−06 1
abz9 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1
la01 2.51E−06 1 1.71E−06 1 2.54E−06 1 8.07E−06 1 1.89E−04 1 3.66E−06 1 1.72E−06 1
la02 1.71E−06 1 1.72E−06 1 1.73E−06 1 2.34E−06 1 9.74E−06 1 1.92E−06 1 1.72E−06 1
la03 1.73E−06 1 1.72E−06 1 1.72E−06 1 1.90E−06 1 1.25E−05 1 1.72E−06 1 1.70E−06 1
la04 1.72E−06 1 1.73E−06 1 1.73E−06 1 1.72E−06 1 2.32E−06 1 1.73E−06 1 1.72E−06 1
la05 1.00E+00 0 1.00E+00 0 1.00E+00 0 1.00E+00 0 1.00E+00 0 1.00E+00 0 1.00E+00 0
la06 2.44E−04 1 1.72E−06 1 3.97E−05 1 1.00E+00 0 1.56E−02 1 3.91E−03 1 2.69E−05 1
la07 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.72E−06 1 1.87E−05 1 1.73E−06 1 1.72E−06 1
la08 1.72E−06 1 1.72E−06 1 1.73E−06 1 2.55E−06 1 5.95E−05 1 1.63E−06 1 1.73E−06 1
la09 2.91E−04 1 1.73E−06 1 3.98E−05 1 1.95E−03 1 1.25E−01 0 1.00E+00 0 8.25E−06 1
la10 1.00E+00 0 7.81E−03 1 1.00E+00 0 1.00E+00 0 1.00E+00 0 1.00E+00 0 2.50E−01 0
la11 1.22E−05 1 1.73E−06 1 2.53E−06 1 1.81E−05 1 1.95E−03 1 5.00E−01 0 3.78E−06 1
la12 2.55E−06 1 1.72E−06 1 1.81E−05 1 2.89E−04 1 3.91E−03 1 1.96E−04 1 3.77E−06 1
la13 3.78E−06 1 1.73E−06 1 2.53E−06 1 3.95E−05 1 6.10E−05 1 5.00E−01 0 2.55E−06 1
la14 1.00E+00 0 1.00E+00 0 1.00E+00 0 1.00E+00 0 1.00E+00 0 1.00E+00 0 1.00E+00 0
la15 1.73E−06 1 1.73E−06 1 1.72E−06 1 1.72E−06 1 3.17E−06 1 1.73E−06 1 1.73E−06 1
la16 1.73E−06 1 1.72E−06 1 1.72E−06 1 2.54E−06 1 2.87E−06 1 1.73E−06 1 1.73E−06 1
la17 1.73E−06 1 1.72E−06 1 1.72E−06 1 1.73E−06 1 4.10E−06 1 1.72E−06 1 1.73E−06 1
la18 1.73E−06 1 1.72E−06 1 1.73E−06 1 1.72E−06 1 2.55E−06 1 2.56E−06 1 1.73E−06 1
la19 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.91E−06 1 4.72E−06 1 1.73E−06 1 1.73E−06 1
la20 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 2.37E−05 1 1.73E−06 1 1.73E−06 1
la21 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1
la22 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 4.28E−06 1 1.73E−06 1 1.73E−06 1
la23 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1
la24 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1
la25 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 5.47E−06 1 1.73E−06 1 1.73E−06 1
la26 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1
la27 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1
la28 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 2.56E−06 1 1.73E−06 1
la29 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1
la30 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1
orb01 1.73E−06 1 1.72E−06 1 1.73E−06 1 1.73E−06 1 6.64E−06 1 1.73E−06 1 1.73E−06 1
orb02 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.72E−06 1 6.31E−06 1 1.73E−06 1 1.73E−06 1
orb03 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 3.18E−06 1 1.73E−06 1 1.73E−06 1
orb04 1.72E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 8.86E−06 1 1.73E−06 1 1.73E−06 1
orb05 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 7.68E−06 1 1.73E−06 1 1.73E−06 1
orb06 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 8.53E−05 1 1.73E−06 1 1.73E−06 1
orb07 1.73E−06 1 1.72E−06 1 1.72E−06 1 1.73E−06 1 3.77E−05 1 1.72E−06 1 1.73E−06 1
orb08 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 3.55E−05 1 1.73E−06 1 1.73E−06 1
orb09 1.73E−06 1 1.72E−06 1 1.73E−06 1 1.92E−06 1 7.18E−05 1 1.73E−06 1 1.73E−06 1
orb10 1.73E−06 1 1.73E−06 1 1.73E−06 1 1.73E−06 1 3.72E−03 1 1.73E−06 1 1.73E−06 1
win/equal/lose 44/4/0 46/2/0 45/3/0 44/4/0 44/4/0 42/6/0 45/3/0

Table 12
Comparison of all algorithms and used encoding scheme approaches for JSSP benchmarks.
Algorithms RK SPV ROV
Sum Hit NIS Avg. RPD Sum Hit NIS Avg. RPD Sum Hit NIS Avg. RPD
PSO 151 10 18.84 5 2 34.68 67 6 25.17
TSA 74 5 24.43 85 4 23.22 79 5 24.32
ABC 109 9 20.91 118 10 20.15 108 6 20.49
DSA 224 10 14.88 182 11 15.39 120 11 18.56
GWO 99 9 23.24 223 12 13.13 106 9 23.20
WOA 230 10 14.65 145 11 18.58 148 10 18.56
LFD 95 8 22.12 110 9 20.40 88 7 22.75
AAA 318 12 8.66 339 14 7.33 309 12 8.49

of 30 runs of all algorithms for the relevant JSSP benchmark achieve similar results, it can be said that it achieves better results
problems. As can be seen from the convergence graphs, the AAA when using the SPV encoding scheme.
algorithm converges slowly, but it continues to converge until
the determined Max_FEs are consumed. Unlike AAA, the PSO 7. Conclusions
algorithm converges very fast and gets stuck in local solutions.
Overall, WOA achieves the best convergence after AAA. The effi- JSSP is an integer-type discrete optimization problem. JSSP can
ciency of AAA can be seen better, especially in JSSP benchmarks be solved by the exact mathematical solvers and the heuristic
with large sizes. While AAA, in RK, and ROV encoding schemes approaches. However, exact mathematical solvers cannot produce
11
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

Fig. 1. The overall Friedman’s test results of all methods.

Fig. 2. Convergence graphs of the algorithms used RK encoding scheme.

useful solutions for NP-hard optimization problems like JSSPs problems such as JSSP by these algorithms. When the literature
in a reasonable time. For this reason, metaheuristic algorithms is examined, three different Encoding schemes that are widely
are frequently preferred in the solution of NP-hard optimization used have been determined. A total of 24 different combinations
problems in recent decades. were obtained by applying the determined encoding schemes
In this study, eight different state-of-the-art algorithms (AAA, to each algorithm. Experimental studies were carried out with
PSO, TSA, ABC, DSA, GWO, WOA, and LFD) with different char- these combinations on 48 benchmark problems that are widely
acteristics were selected for the solution of JSSPs. Metaheuristic used in the literature. As a result of this comparison, the perfor-
algorithms are often proposed for solving continuous optimiza- mance of both algorithms and encoding schemes has been exami-
tion problems. Encoding schemes are used to solve discrete ned in detail. When the results for each encoding scheme were
12
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

Fig. 3. Convergence graphs of the algorithms used SPV encoding scheme.

Fig. 4. Convergence graphs of the algorithms used ROV encoding scheme.

13
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

examined, the AAA algorithm obtained the best result. As a result [9] M.A. Şahman, A.A. Altun, A.O. Dündar, A new MILP model proposal in
of the evaluation of all combinations according to the Friedman feed formulation and using a hybrid-linear binary PSO (H-LBP) approach
for alternative solutions, Neural Comput. Appl. 29 (2018) 537–552.
rank test, it was seen that AAA-SPV was the most successful com-
[10] M.S. Kiran, TSA: Tree-seed algorithm for continuous optimization, Expert
bination. In addition, when the best performing coding scheme Syst. Appl. 42 (2015) 6686–6698.
of each algorithm is examined, it has been determined that SPV [11] A. Babalik, A.C. Cinar, M.S. Kiran, A modification of tree-seed algorithm
encoding scheme performs better with 6 algorithms and RK en- using Deb’s rules for constrained optimization, Appl. Soft Comput. 63
(2018) 289–305.
coding scheme performs better with 2 algorithms even though
[12] A. Cinar, M. Kiran, The performance of penalty methods on tree-seed
ROV coding scheme has not been the best performing coding algorithm for numerical constrained optimization problems, Int. Arab J.
scheme in any algorithm. In the next step, the best performing Inf. Technol. 17 (2020) 799–807.
coding scheme of each algorithm according to the Friedman rank [13] A.C. Cinar, Training feed-forward multi-layer perceptron artificial neural
test was subjected to the Wilcoxon sign test with the most networks with a tree-seed algorithm, Arab. J. Sci. Eng. 45 (2020)
10915–10938.
successful combination, AAA-SPV. As a result of this test, the [14] A.C. Cinar, M.S. Kiran, A parallel implementation of tree-seed algorithm
WOA_RK combination obtained the closest success to the AAA- on CUDA-supported graphical processing unit, J. Faculty Eng. Archit.
SPV combination in terms of the significance level. Despite the Gazi Univ. 33 (2018) 1397–1409.
fact this, AAA-SPV achieved better results in 42 of 48 problems [15] A.C. Cinar, S. Korkmaz, M.S. Kiran, A discrete tree-seed algorithm for
solving symmetric traveling salesman problem, Eng. Sci. Technol. Int. J.
than WOA-RK, and for the remaining 6 problems, they performed
23 (2020) 879–890.
equally in terms of the significance level. As a result, it has been [16] I. Gungor, B.G. Emiroglu, A.C. Cinar, M.S. Kiran, Integration search
seen that AAA in algorithms and SPV in coding schemes provide strategies in tree seed algorithm for high dimensional function
superior performance compared to others. optimization, Int. J. Mach. Learn. Cybern. 11 (2020) 249–267.
[17] M. Sahman, A. Cinar, I. Saritas, A. Yasar, Tree-seed algorithm in solving
real-life optimization problems, in: IOP Conference Series: Materials
CRediT authorship contribution statement Science and Engineering, 2019, 012030, Publishing.
[18] M.A. Sahman, A.C. Cinar, Binary tree-seed algorithms with S-shaped
Mehmet Akif Şahman: Investigation, Conceptualization, and V-shaped transfer functions, Int. J. Intell. Syst. Appl. Eng. 7 (2019)
111–117.
Methodology, Software, Data curation, Writing – original draft,
[19] D. Karaboga, An idea based on honey bee swarm for numerical
Visualization, Validation, Writing – review & editing, Super- optimization, 2005, Publishing.
vision. Sedat Korkmaz: Software, Data curation, Visualization, [20] M.S. Kiran, The continuous artificial bee colony algorithm for binary
Validation, Writing – review & editing. optimization, Appl. Soft Comput. 33 (2015) 15–23.
[21] M.S. Kiran, H. Hakli, M. Gunduz, H. Uguz, Artificial bee colony algorithm
with variable search strategy for continuous optimization, Inform. Sci.
Declaration of competing interest 300 (2015) 140–157.
[22] M. Zhang, V. Palade, Y. Wang, Z. Ji, Attention-based word embeddings
The authors declare that they have no known competing using artificial bee colony algorithm for aspect-level sentiment
financial interests or personal relationships that could have classification, Inform. Sci. 545 (2021) 713–738.
[23] P. Civicioglu, Transforming geocentric cartesian coordinates to geodetic
appeared to influence the work reported in this paper. coordinates by using differential search algorithm, Comput. Geosci. 46
(2012) 229–247.
Acknowledgments [24] K. Abaci, V. Yamacli, Differential search algorithm for solving multi-
objective optimal power flow problem, Int. J. Electr. Power Energy Syst.
79 (2016) 1–10.
The authors wish to thank the Scientific Research Projects [25] M. Mutluer, M.A. Şahman, M. Çunkaş, Heuristic optimization based on
Coordinatorship at Selçuk University, Scientific Research penalty approach for surface permanent magnet synchronous machines,
Projects Coordinatorship at Konya Technical University, and Arab. J. Sci. Eng. 45 (2020) 6751–6767.
The Scientific and Technological Research Council of Turkey for [26] M. Liu, X. Yao, Y. Li, Hybrid whale optimization algorithm enhanced
with Lévy flight and differential evolution for job shop scheduling
their institutional supports. This research did not receive any
problems, Appl. Soft Comput. 87 (2020) 105954.
specific grant from funding agencies in the public, commercial, [27] S. Mirjalili, S.M. Mirjalili, A. Lewis, Grey wolf optimizer, Adv. Eng. Softw.
or not-for-profit sectors. 69 (2014) 46–61.
[28] A. Mosavi, S. Samadianfard, S. Darbandi, N. Nabipour, S.N. Qasem,
E. Salwana, .S.S. Band, Predicting soil electrical conductivity using
References
multi-layer perceptron integrated with grey wolf optimizer, J. Geochem.
Explor. 220 (2021) 106639.
[1] J. Chen, S.-y Zhang, Z. Gao, L.-x. Yang, Feature-based initial population [29] R. Ahmadi, G. Ekbatanifard, P. Bayat, A modified grey wolf optimizer
generation for the optimization of job shop problems, J. Zhejiang Univ. based data clustering algorithm, Appl. Artif. Intell. 35 (2021) 63–79.
Sci. C 11 (2010) 767–777. [30] P. Dutta, S.K. Nayak, Grey wolf optimizer based PID controller for speed
[2] L. Sun, X. Cheng, Y. Liang, Solving job shop scheduling problem using control of BLDC motor, J. Electr. Eng. Technol. 1–7.
genetic algorithm with penalty function, Int. J. Intell. Inform. Process. 1 [31] W.-j. Niu, Z.-k. Feng, S. Liu, Y.-b. Chen, Y.-s. Xu, J. Zhang, Multiple
(2010) 65–77. hydropower reservoirs operation by hyperbolic grey wolf optimizer
[3] M.R. Garey, D.S. Johnson, R. Sethi, The complexity of flowshop and based on elitism selection and adaptive mutation, Water Resour.
jobshop scheduling, Math. Oper. Res. 1 (1976) 117–129. Manage. 1–19.
[4] J. Kennedy, R. Eberhart, Particle swarm optimization, in: Proceedings [32] S. Mirjalili, A. Lewis, The whale optimization algorithm, Adv. Eng. Softw.
of ICNN’95-International Conference on Neural Networks, 1995, pp. 95 (2016) 51–67.
1942–1948. Publishing. [33] B. Raei, A. Ahmadi, M.R. Neyshaburi, M.A. Ghorbani, F. Asadzadeh,
[5] A.A. Altun, M.A. Şahman, Cost optimization of mixed feeds with the Comparative evaluation of the whale optimization algorithm and
particle swarm optimization method, Neural Comput. Appl. 22 (2013) backpropagation for training neural networks to model soil wind
383–390. erodibility, Arab. J. Geosci. 14 (2021) 1–19.
[6] N. Karasekreter, M.A. Şahman, F. Başçiftçi, U. Fidan, PSO-based [34] S.K. Dewi, D.M. Utama, A new hybrid whale optimization algorithm for
clustering for the optimization of energy consumption in wireless green vehicle routing problem, Syst. Sci. Control Eng. 9 (2021) 61–72.
sensor network, Emerg. Mater. Res. 9 (2020) 776–783. [35] M.R. Babu, C.V. Kumar, S. Anitha, Simultaneous reconfiguration
[7] M.S. Kıran, M. Gündüz, Ö.K. Baykan, A novel hybrid algorithm based and optimal capacitor placement using adaptive whale optimization
on particle swarm and ant colony optimization for finding the global algorithm for radial distribution system, J. Electr. Eng. Technol. 16
minimum, Appl. Math. Comput. 219 (2012) 1515–1521. (2021) 181–190.
[8] M.S. Kıran, E. Özceylan, M. Gündüz, T. Paksoy, A novel hybrid approach [36] E.H. Houssein, M.R. Saad, F.A. Hashim, H. Shaban, M. Hassaballah,
based on particle swarm optimization and ant colony algorithm to Lévy flight distribution: A new metaheuristic algorithm for solving
forecast energy demand of Turkey, Energy Convers. Manage. 53 (2012) engineering optimization problems, Eng. Appl. Artif. Intell. 94 (2020)
75–83. 103731.

14
M.A. Şahman and S. Korkmaz Knowledge-Based Systems 256 (2022) 109711

[37] S.A. Uymaz, G. Tezel, E. Yel, Artificial algae algorithm (AAA) for [59] W.-j Xia, Z.-m Wu, A hybrid particle swarm optimization approach for
nonlinear global optimization, Appl. Soft Comput. 31 (2015) 153–171. the job-shop scheduling problem, Int. J. Adv. Manuf. Technol. 29 (2006)
[38] A. Babalik, A. Ozkis, S.A. Uymaz, M.S. Kiran, A multi-objective artificial 360–366.
algae algorithm, Appl. Soft Comput. 68 (2018) 377–395. [60] T.-L. Lin, S.-J. Horng, T.-W. Kao, Y.-H. Chen, R.-S. Run, R.-J. Chen, J.-L.
[39] E. Kaya, S.A. Uymaz, B. Kocer, Boosting galactic swarm optimization Lai, I.-H. Kuo, An efficient job-shop scheduling algorithm based on
with ABC, Int. J. Mach. Learn. Cybern. 10 (2019) 2401–2419. particle swarm optimization, Expert Syst. Appl. 37 (2010) 2629–2636.
[40] H.G. Kocer, S.A. Uymaz, A modified artificial algae algorithm for large [61] X. Wang, H. Duan, A hybrid biogeography-based optimization algorithm
scale global optimization problems, Int. J. Intell. Syst. Appl. Eng. 6
for job shop scheduling problem, Comput. Ind. Eng. 73 (2014) 96–114.
(2018) 306–310.
[62] T.-K. Dao, T.-S. Pan, J.-S. Pan, Parallel bat algorithm for optimizing
[41] S. Korkmaz, A. Babalik, M.S. Kiran, An artificial algae algorithm for
makespan in job shop scheduling problems, J. Intell. Manuf. 29 (2018)
solving binary optimization problems, Int. J. Mach. Learn. Cybern. 9
451–462.
(2018) 1233–1247.
[42] S. Korkmaz, M.S. Kiran, An artificial algae algorithm with stigmergic [63] Y. Lu, T. Jiang, Bi-population based discrete bat algorithm for the
behavior for binary optimization, Appl. Soft Comput. 64 (2018) low-carbon job shop scheduling problem, IEEE Access 7 (2019)
627–640. 14513–14522.
[43] B. Turkoglu, E. Kaya, Training multi-layer perceptron with artificial algae [64] A. Khadwilard, S. Chansombat, T. Thepphakorn, P. Thapatsuwan, W.
algorithm, Eng. Sci. Technol. Int. J. 23 (2020) 1342–1350. Chainate, P. Pongcharoen, Investigation of firefly algorithm parameter
[44] S.A. Uymaz, G. Tezel, E. Yel, Artificial algae algorithm with multi-light setting for solving job shop scheduling problems, Oper. Res. Netw.
source for numerical optimization and applications, Biosystems 138 Thailand (2011) 89–97.
(2015) 25–38. [65] M. Rohaninejad, A.S. Kheirkhah, B. Vahedi Nouri, P. Fattahi, Two hybrid
[45] K. Steinhöfel, A. Albrecht, C.-K. Wong, Two simulated annealing-based tabu search–firefly algorithms for the capacitated job shop scheduling
heuristics for the job shop scheduling problem, European J. Oper. Res. problem with sequence-dependent setup cost, Int. J. Comput. Integr.
118 (1999) 524–548. Manuf. 28 (2015) 470–487.
[46] R. Suresh, K. Mohanasundaram, Pareto archived simulated annealing [66] R. Babukartik, P. Dhavachelvan, Hybrid algorithm using the advantage of
for job shop scheduling with multiple objectives, Int. J. Adv. Manuf. ACO and cuckoo search for job scheduling, Int. J. Inf. Technol. Converg.
Technol. 29 (2006) 184–196.
Serv. 2 (25) (2012).
[47] P.J. Van Laarhoven, E.H. Aarts, J.K. Lenstra, Job shop scheduling by
[67] A. Baykasoğlu, A. Hamzadayi, S.Y. Köse, Testing the performance of
simulated annealing, Oper. Res. 40 (1992) 113–125.
teaching–learning based optimization (TLBO) algorithm on combinatorial
[48] E. Nowicki, C. Smutnicki, An advanced tabu search algorithm for the job
problems: Flow shop and job shop scheduling cases, Inform. Sci. 276
shop problem, J. Sched. 8 (2005) 145–159.
[49] S. Ponnambalam, P. Aravindan, S. Rajesh, A tabu search algorithm for (2014) 204–218.
job shop scheduling, Int. J. Adv. Manuf. Technol. 16 (2000) 765–771. [68] K. Yu, X. Wang, Z. Wang, An improved teaching-learning-based
[50] E.D. Taillard, Parallel taboo search techniques for the job shop optimization algorithm for numerical and engineering optimization
scheduling problem, ORSA J. Comput. 6 (1994) 108–117. problems, J. Intell. Manuf. 27 (2016) 831–843.
[51] J.F. Gonçalves, J.J. de Magalhães Mendes, M.c.G. Resende, A hybrid [69] C. Lu, S. Xiao, X. Li, L. Gao, An effective multi-objective discrete
genetic algorithm for the job shop scheduling problem, European J. grey wolf optimizer for a real-world scheduling problem in welding
Oper. Res. 167 (2005) 77–95. production, Adv. Eng. Softw. 99 (2016) 161–176.
[52] B.J. Park, H.R. Choi, H.S. Kim, A hybrid genetic algorithm for the job [70] J.C. Bean, Genetic algorithms and random keys for sequencing and
shop scheduling problems, Comput. Ind. Eng. 45 (2003) 597–613. optimization, ORSA J. Comput. 6 (1994) 154–160.
[53] L. Wang, D.-Z. Zheng, An effective hybrid optimization strategy for [71] H. Fisher, G.L. Thompson, Probabilistic Learning Combinations of Local
job-shop scheduling problems, Comput. Oper. Res. 28 (2001) 585–596. Job-Shop Scheduling Rules, Prentice-Hall, Englewood Cliffs, 1963, pp.
[54] H. Ge, Y. Liang, Y. Zhou, X. Guo, A particle swarm optimization-based 225–251.
algorithm for job-shop scheduling problems, Int. J. Comput. Methods 2 [72] J. Adams, E. Balas, D. Zawack, The shifting bottleneck procedure for job
(2005) 419–430. shop scheduling, Manage. Sci. 34 (1988) 391–401.
[55] Z. Lian, B. Jiao, X. Gu, A similar particle swarm optimization algorithm
[73] S. Lawrence, Resouce Constrained Project Scheduling: An Experimental
for job-shop scheduling to minimize makespan, Appl. Math. Comput.
Investigation of Heuristic Scheduling Techniques (Supplement), Graduate
183 (2006) 1008–1017.
School of Industrial Administration, Carnegie-Mellon University, 1984.
[56] D. Sha, C.-Y. Hsu, A hybrid particle swarm optimization for job shop
scheduling problem, Comput. Ind. Eng. 51 (2006) 791–808. [74] D. Applegate, W. Cook, A computational study of the job-shop
[57] B. Giffler, G.L. Thompson, Algorithms for solving production-scheduling scheduling problem, ORSA J. Comput. 3 (1991) 149–156.
problems, Oper. Res. 8 (1960) 487–503. [75] M. Friedman, A comparison of alternative tests of significance for the
[58] M.F. Tasgetiren, M. Sevkli, Y.-C. Liang, M.M. Yenisey, A particle problem of m rankings, Ann. Math. Stat. 11 (1940) 86–92.
swarm optimization and differential evolution algorithms for job shop [76] F. Wilcoxon, Individual comparisons by ranking methods, in:
scheduling problem, Int. J. Oper. Res. 3 (2006) 120–135. Breakthroughs in Statistics, 1992, pp. 196–202, Publishing.

15

You might also like