Hindawi

International Journal of Differential Equations

Volume 2020, Article ID 3521936, 7 pages
https://doi.org/10.1155/2020/3521936

Research Article
Variational Iteration Method and Differential Transformation
Method for Solving the SEIR Epidemic Model

A. Harir ,1 S. Melliani ,2 H. El Harfi,1 and L. S. Chadli1

1
Laboratory of Applied Mathematics and Scientific Computing, Sultan Moulay Slimane University, B.P. 523,
Beni Mellal, Morocco
2
Laboratory of Flows and Transfers Modelling, Sultan Moulay Slimane University, B.P. 523, Beni Mellal, Morocco

Correspondence should be addressed to A. Harir; [email protected]

Received 23 April 2020; Revised 29 May 2020; Accepted 3 June 2020; Published 2 July 2020

Academic Editor: Sining Zheng

Copyright © 2020 A. Harir et al. This is an open access article distributed under the Creative Commons Attribution License, which
permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The aim of the present study is to analyze and find a solution for the model of nonlinear ordinary differential equations (ODEs)
describing the so-called coronavirus (COVID-19), a deadly and most parlous virus. The mathematical model based on four
nonlinear ODEs is presented, and the corresponding numerical results are studied by applying the variational iteration method
(VIM) and differential transformation method (DTM).

1. Introduction model of nonlinear ordinary differential equations (ODEs)

describing the deadly and most parlous coronavirus
The whole world is experiencing hardship due to corona- (COVID-19). A mathematical model based on the four
virus (COVID-19), which was first identified in Wuhan, nonlinear ODEs is presented, and the corresponding nu-
China, in the month of December 2019. It has been con- merical results are studied by applying the variational it-
sidered that COVID-19 originated from wild animals (bats eration method (VIM) and differential transformation
[1]) and transmitted to humans as numerous infected pa- method (DTM).
tients claimed that they had been to a local wet market in The VIM was developed by He [10, 11]. In recent years, a
Wuhan during the end of November [2]. Later, some in- great deal of attention has been devoted to the study of this
vestigators confirmed that the virus transmission occurs method. The reliability of the method and the reduction in
from person to person [3]. the size of the computational domain make this method
Mathematical models can simulate the effects of a disease applicable to a wide range of model predictions. This method
at many levels, ranging from how the disease influences the is based on the use of restricted variations and a correction
interaction between cells in a single patient (within-host functional, and it was found to have wide applications in
models) to how it spreads across several geographically finding a solution for the nonlinear ordinary and partial
separated populations (metapopulation models). Models differential equations [12–14]. This method does not depend
simulating the disease spread within and among pop- on small parameters in the differential equation and pro-
ulations, such as those used to forecast the COVID-19 vides a solution (or an approximation to it) as a sequence of
outbreak [4], are typically based on the SEIR model. iterations. The method does not require that the nonline-
The SEIR model is based on the division of the pop- arities be differentiable with respect to the dependent var-
ulation under study into four compartments: an individual iable and its derivatives [15, 16].
can either be susceptible (S), exposed to the disease but not The DTM is a numerical method for solving differential
yet infectious (E), infectious (I), or recovered (R). The SEIR equations. The concept of the differential transformation
model can represent many human infectious diseases [5–9]. was first proposed by Zhou [17], and its main application
In this paper, we focus, analyze, and find a solution for the therein is solving both linear and nonlinear initial value
2 International Journal of Differential Equations

problems in electric circuit analysis. The DTM provides in a β α γ

fast manner exact values of the nth derivative of an analytical S E I R
function at a point in terms of known and unknown Figure 1: SEIR compartmental model.
boundary conditions. This method constructs, for differ-
ential equations, an analytical solution in the form of a
polynomial. (2) The rate of change in the number of exposed people
� the susceptible portion of the population × the
average number of people infected by an infectious
2. The SEIR Model person over the average duration of infection × the
The SEIR model in epidemiology for the spread of an in- number of people infected by infectious people + the
fectious disease is described by the following system of susceptible portion of the population × the rate of
differential equations: infectious animal source − the number of exposed
people over the average latency period − percentage
⎪
⎧ dS of population traveling out × the number of exposed
⎪
⎪ (t) � −βS(t)I(t),
⎪
⎪ dt people − the death rate of the exposed people × the
⎪
⎪
⎪
⎪ number of exposed people − testing and therapy rate
⎪
⎪
⎪
⎪ dE × the number of exposed people:
⎪
⎪ (t) � βS(t)I(t) − αE(t),
⎪
⎪
⎨ dt
⎪
⎪ (1)
⎪
⎪ dE S(t) Z
⎪
⎪ dI (t) � β I(t) + S(t) − αE(t)
⎪
⎪ (t) � αE(t) − cI(t), dt N N
⎪
⎪ dt (3)
⎪
⎪
⎪
⎪ ϱ ϱ
⎪
⎪ − 􏼒 I + E 􏼓E(t) − μE(t) − σE(t).
⎪
⎪ dR N N
⎪
⎩ (t) � cI(t).
dt
(3) The rate of change in the number of infected people �
Here, β, α, and c are positive parameters and S, E, I, and R the number of exposed people over the average la-
denote the fractions of the population that are susceptible, tency period − the number of infected people over
exposed, infectious, and recovered, respectively. the average duration of infection − percentage of
A schematic diagram of the disease transmission among population traveling out × the number of infected
the individuals is shown in Figure 1 using the SEIR model. people − the death rate of the infected people × the
For more information about the model refer to [18]. number of infected people:
The SEIR model of the novel coronavirus (COVID-19)
can be represented as follows:
dI ϱ ϱ
(1) The rate of change in the number of susceptible (t) � αE(t) − cI(t) − 􏼒 I + E 􏼓I(t) − μI(t). (4)
people � the susceptible portion of the population × dt N N
the average number of people infected by an infec-
tious person over the average duration of infection × (4) The rate of change in the number of recovered
the number of people infected by infectious people − people � the number of infected people over the
the susceptible portion of the population × the rate of average duration of infection − the death rate of the
infectious animal source + travelers entering − recovered people × the number of recovered peo-
percentage of population traveling out × the number ple + testing and therapy rate × the number of ex-
of susceptible people + natural birth rate × the total posed people Figure 2:
number of population − the death rate of susceptible
people × the number of susceptible people:
dR
(t) � cI(t) − μR(t) + σE(t). (5)
dS S(t) Z dt
(t) � −β I(t) − S(t) + ρI + ρE 􏼁
dt N N
(2) The transitions between model classes can now be
ϱ ϱ expressed by the following system of first-order differential
− 􏼒 I + E 􏼓S(t) + ]N(t) − μS(t). equations (Table 1):
N N
International Journal of Differential Equations 3

ρI , ρE

ν β α γ
S E I R

μ μ μ μ

ϱI , ϱE ϱI , ϱE ϱI , ϱE

Figure 2: The SEIR compartmental model of the novel coronavirus (COVID-19).

Table 1: Parameters and initial conditions of the SEIR model.

Parameter Definition
β Average number of people infected by an infectious person over the average duration of infection
α Number of exposed people over the average latency period
c Number of infected people over the average duration of infection
μ The death rate of people
] The birth rate parameter of people
σ The cure rate of people
Z Force of infection in the baseline scenario
ρI Average daily number of international inbound air passengers
ρE Average daily number of domestic inbound air passengers
ϱI Average daily number of international outbound air passengers
ϱE Average daily number of domestic outbound travellers
S0 Number of susceptible people at time t � 0
E0 Number of asymptomatic and noninfectious people at time t � 0
I0 Number of asymptomatic but infectious people at time t � 0
R0 Number of recovered people at time t � 0
N Total number of population, N(t) � S(t) + E(t) + I(t) + R(t)

⎪
⎧
⎪ dS S(t) Z ϱ ϱ
⎪
⎪ (t) � −β I(t) − S(t) + ρI + ρE 􏼁 − 􏼒 I + E 􏼓S(t) + ]N(t) − μS(t),
⎪
⎪ dt N N N N
⎪
⎪
⎪
⎪
⎪
⎪ dE S(t) Z ϱ ϱ
⎪
⎪ (t) � β I(t) + S(t) − αE(t) − 􏼒 I + E 􏼓E(t) − μE(t) − σE(t),
⎪
⎨ dt N N N N
⎪ (6)
⎪
⎪ dI ϱ ϱ
⎪
⎪ (t) � αE(t) − cI(t) − 􏼒 I + E 􏼓I(t) − μI(t),
⎪
⎪
⎪
⎪ dt N N
⎪
⎪
⎪
⎪
⎪
⎪ dR
⎩ (t) � cI(t) − μR(t) + σE(t),
dt

with the initial conditions where L is a linear operator, N is a nonlinear operator, and F(t)
is a known analytical function. We can construct a correction
S(0) � S0 , functional according to the variational method as follows:
E(0) � E0 , t
(7) 􏽥 n (s) − F(s)􏼉ds,
Un+1 (t) � Un (t) + 􏽚 λ􏼈LUn (s) + NU
I(0) � I0 , 0

R(0) � R0 . (9)

where λ is the general Lagrange multiplier [19], which can be

3. The Variational Iteration Method identified optimally via the variational theory, Un is the nth
approximate solution, and U 􏽥 n denotes a restricted variation,
To illustrate the basic concepts of the VIM, we consider the which means δNU 􏽥 n � 0. Successive approximations, Un+1 ,
following general nonlinear differential equation: will be obtained by applying the obtained Lagrange multi-
plier and a properly chosen initial approximation U0 .
LU(t) + NU(t) � F(t), (8) Consequently, the solution is given by U � limn⟶∞ Un . For
4 International Journal of Differential Equations

solving equation (6) by means of the VIM, we construct the

correctional functional as follows:

⎧
⎪ t dS (τ) 􏽥S (τ) Z ϱ ϱ
⎪
⎪ Sn+1 (t) � Sn (t) + 􏽚 λ1 (τ)􏼨 n + β n 􏽥In (τ) + 􏽥Sn (τ) − ρI + ρE 􏼁 + 􏼒 I + E 􏼓􏽥Sn (τ) − ]N􏽥 n (τ) + μ􏽥Sn (τ)􏼩dτ,
⎪
⎪ dτ N N N N
⎪
⎪ 0
⎪
⎪
⎪
⎪
⎪
⎪ t 􏽥S (τ)
⎪
⎪ dE (τ) Z ϱ ϱ
⎪ En+1 (t) � En (t) + 􏽚 λ2 (τ)􏼨 n
⎪
⎪ − β n 􏽥In (τ) − 􏽥Sn (τ) + 􏼒α + 􏼒 I + E 􏼓 + μ + σ 􏼓E􏽥 n (τ)􏼩dτ,
⎪
⎨ 0 dτ N N N N
⎪ (10)
⎪
⎪ t
⎪ dI (τ)
⎪
⎪
⎪ I (t) � I (t) + 􏽚 λ3 (τ)􏼨 n 􏽥 n (τ) + 􏼒c + ϱI + ϱE + μ􏼓􏽥In (τ)􏼩dτ,
− αE
⎪
⎪ n+1 n
0 dτ N N
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ t dR (τ)
⎪
⎪
⎩ Rn+1 (t) � Rn (t) + 􏽚 λ4 (τ)􏼨 n − c􏽥In (τ) + μR
􏽥 n (τ) − σ E
􏽥 n (τ)􏼩dτ.
0 dτ

Here, λ1 , λ2 , λ3 , and λ4 are general Lagrange multipliers. respect to Sn (t), En (t), In (t), and Rn (t), noticing that
Making the above correction functional stationary with δ􏽥Sn (τ) � δE
􏽥 n (τ) � δ􏽥In (τ) � δR
􏽥 n (τ) � 0, yields

⎪
⎧ t dSn (τ) 􏽥Sn (τ)
⎪
⎪
⎪ δS (t) � δS (t) + δ 􏽚 λ (τ)􏼨 + β 􏽥In (τ) + Z 􏽥Sn (τ) − ρI + ρE 􏼁 + 􏼒ϱI + ϱE 􏼓􏽥Sn (τ) − ]N􏽥 n (τ) + μ􏽥Sn (τ)􏼩dτ � 0,
⎪
⎪ n+1 n
0
1
dτ N N N N
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ t dEn (τ) 􏽥S (τ) Z ϱ ϱ
⎪
⎪ − β n 􏽥In (τ) − 􏽥Sn (τ) + 􏼒α + 􏼒 I + E 􏼓 + μ + σ 􏼓E 􏽥 n (τ)􏼩dτ � 0,
⎪ δEn+1 (t) � δEn (t) + δ 􏽚 λ2 (τ)􏼨
⎪
⎪
⎨ 0 dτ N N N N
⎪
⎪
⎪
⎪ t dI (τ)
⎪
⎪
⎪ δIn+1 (t) � δIn (t) + δ 􏽚 λ3 (τ)􏼨 n 􏽥 n (τ) + 􏼒c + ϱI + ϱE + μ􏼓􏽥In (τ)􏼩dτ � 0,
− αE
⎪
⎪ 0 dτ N N
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ t dRn (τ)
⎪
⎩ δRn+1 (t) � δRn (t) + δ 􏽚 λ4 (τ)􏼨 dτ − c􏽥In (τ) + μR
⎪ 􏽥 n (τ) − σ E
􏽥 n (τ)􏼩dτ � 0.
0

(11)

Therefore, the Lagrange multiplier can readily be λ1 (τ) � −1, λ2 (τ) � −1, λ3 (τ) � −1, and λ4 (τ) � −1. Con-
identified: sequently, the iteration formula can be obtained as follows:

⎪
⎧ t dS (τ) 􏽥S (τ) Z ϱ ϱ
⎪
⎪ S (t) � S (t) − 􏽚 􏼨 n + β n 􏽥In (τ) + 􏽥Sn (τ) − ρI + ρE 􏼁 + 􏼒 I + E 􏼓􏽥Sn (τ) − ]N 􏽥 n (τ) + μ􏽥Sn (τ)􏼩dτ,
⎪
⎪ n+1 n
dτ N N N N
⎪
⎪ 0
⎪
⎪
⎪
⎪
⎪
⎪ t dE (τ) 􏽥Sn (τ)
⎪
⎪ n 􏽥In (τ) − Z 􏽥Sn (τ) + 􏼒α + 􏼒ϱI + ϱE 􏼓 + μ + σ 􏼓E
􏽥 n (τ)􏼩dτ,
⎪
⎪ E n+1 (t) � E n (t) − 􏽚 􏼨 − β
⎪
⎪ 0 dτ N N N N
⎨
⎪ (12)
⎪
⎪ t dI (τ)
⎪
⎪
⎪
⎪ In+1 (t) � In (t) − 􏽚 􏼨
n 􏽥 n (τ) + 􏼒c + ϱI + ϱE + μ􏼓􏽥In (τ)􏼩dτ,
− αE
⎪
⎪ 0 dτ N N
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ t dR (τ)
⎪
⎪
⎩ Rn+1 (t) � Rn (t) − 􏽚 􏼨 n − c􏽥In (τ) + μR 􏽥 n (τ) − σ E
􏽥 n (τ)􏼩dτ.
0 dτ

With initial approximations S(0) � 2500, E(0) � 1, successive approximations, and considering the following
I(0) � 1, R(0) � 0, and N � 2502, which in turn gives values for parameters (see [20]) β � 0.8, α � 0.75,
International Journal of Differential Equations 5

σ � 0.1, c � 0.05, ] � 0.009/N, μ � 0.01, Similarly, we get the following system after two terms:
Z � 0.001, ρI � 0.15, ρE � 0.15, ϱI � 0.01, and ϱI � 0.03, we ⎧
⎪ S2 (t) � 2500 − 25.53132774t − 0.143869926t2 + 1.87756013310−3 t3 ,
⎪
⎪
obtain ⎪
⎪
⎨ E2 (t) � 1 − 0.059656274t + 0.29733881t2 − 1.87756013310−3 t3 ,
⎧ S1 (t) � 2500 − 25.53132774t,
⎪
⎪
⎪
⎪ I2 (t) � 1 + 0.689984012t − 0.043076138t2 ,
⎪
⎪
⎪ ⎪
⎪
⎩
⎪
⎨ E1 (t) � 1 − 0.059656274t, R2 (t) � 0.15t + 0.013516786t2 .
⎪ (13)
⎪
⎪ I1 (t) � 1 + 0.689984012t, (14)
⎪
⎪
⎩
R1 (t) � 0.15t. For the solution after three terms, we can write

⎧ S3 (t) � 2500 − 25.53132774t − 0.143869926t2 + 0.013851166t3 − 8.48550280810− 5 t4

⎪
⎪
⎪
⎪
⎪
⎪ −4.79158897410−7 t5 + 4.31004739710−9 t6 ,
⎪
⎪
⎪
⎪
⎪ E (t) � 1 − 0.059656274t + 0.29733881t2 − 0.09860941t3 + 4.8381106310−4 t4
⎨ 3
⎪ (15)
⎪
⎪ −4.79158897410−7 t5 + 4.31004739710−9 t6 ,
⎪
⎪
⎪
⎪ I (t) � 1 + 0.689984012t − 0.043076138t2 + 0.074410709t3 − 3.52042524910−4 t4 ,
⎪
⎪ 3
⎪
⎪
⎩
R3 (t) � 0.15t + 0.013516786t2 + 9.1483020810−3 t3 − 4.69390033310−5 t4 .

4. The Differential Transformation Method equations are given in [17, 21]. According to the operations
of differential transformation given in Table 1 in [21], we
The basic definition and the fundamental theorems of the have the following recurrence relation:
DTM and its applicability to various kinds of differential

⎪
⎧ 1 ⎡ k
⎪
⎪
⎪ S(k + 1) � ⎣− β 􏽘 S(m)I(k − m) − Z S(k) + ρ + ρ 􏼁δ(k) + ]N(k) − 􏼔􏼒ϱI + ϱE 􏼓 + μ􏼕S(k)⎤⎦,
⎪
⎪ k + 1 N m�0 N I E
N N
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ k
⎪ 1 ⎡
⎪
⎪
⎪ E(k + 1) � ⎣ β 􏽘 S(m)I(k − m) + Z S(k) − 􏼔α + 􏼒ϱI + ϱE 􏼓 + μ + σ 􏼕E(k)⎤⎦,
⎪
⎨ k + 1 N m�0 N N N
⎪ (16)
⎪
⎪
⎪
⎪
⎪
⎪ 1 ϱI ϱE
⎪
⎪
⎪ I(k + 1) � k + 1 􏼔αE(k) − 􏼔c + 􏼒N + N 􏼓 + μ􏼕I(k)􏼕,
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪ 1
⎪
⎩ R(k + 1) � [cI(k) − μR(k) + σE(k)].
k+1
3
The inverse differential transformation of S(k) is defined S(t) � 􏽘 S(k)tk ,
as follows: when t0 is taken as zero, the given function S(t) is k�0
declared by a finite series, and the above equation can be 3
written in the form E(t) � 􏽘 E(k)tk ,
∞ k�0
(18)
S(t) � 􏽘 S(k)tk . (17) 3
k
k�0 I(t) � 􏽘 I(k)t ,
k�0
By solving the above equations for 3
S(k + 1), E(k + 1), I(k + 1), and R(k + 1) up to order 3, we R(t) � 􏽘 R(k)tk .
obtain the functions of S(k), E(k), I(k), and R(k), k�0
respectively:
6 International Journal of Differential Equations

2500 600
2450
500
2400
2350
400
2300

I (t)
S (t)

2250 300

2200
200
2150
2100 100
2050
2000 0
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
Time in days (t) Time in days (t)

VIM VIM
DTM DTM

Figure 3: Comparison of solutions of the proposed model (6) for Figure 5: Comparison of solutions of the proposed model (6) for
compartment S using VIM and DTM. compartment I using VIM and DTM.

100 90

0 80

70
–100
60
–200
50
E (t)

R (t)

–300
40
–400
30
–500
20
–600 10

–700 0
0 2 4 6 8 10 12 14 16 18 20 0 2 4 6 8 10 12 14 16 18 20
Time in days (t) Time in days (t)
VIM VIM
DTM DTM
Figure 4: Comparison of solutions of the proposed model (6) for Figure 6: Comparison of solutions of the proposed model (6) for
compartment E using VIM and DTM. compartment R using VIM and DTM.

With initial approximations, S(0) � 2500, E(0) � 1,

I(0) � 1, R(0) � 0, and N � 2502 and parameters
5. Conclusions
β � 0.8, α � 0.75, σ � 0.1, c � 0.05, ] � 0.009/N, μ � 0.01, In this paper, we have developed the SEIR model of the
Z � 0.001, ρI � 0.15, ρE � 0.15, ϱI � 0.01, and ϱI � 0.03, COVID-19 epidemic in China that incorporates key features
and applying the conditions in equations (16) and (18), we of this pandemic. For solving this model, we used the
obtain the approximate solution after three terms as follows: variational iteration method (VIM) and differential trans-
formation method (DTM). It is found that these methods are
⎧
⎪
⎪ S(t) � 2500 − 25.53132774t − 0.143869688t2 + 0.013851335t3 , effective in providing analytic form solutions for such
⎪
⎪
⎪
⎨ E(t) � 1 − 0.059656274t + 0.297292652t2 − 0.099439782t3 , problems. The comparison of the results obtained by these
⎪
⎪ two methods is in excellent agreement.
⎪ I(t) � 1 + 0.689984012t − 0.043076138t2 + 0.075184915t3 ,
⎪
⎪
⎩ For further research, we propose the study of the frac-
R(t) � 0.15t + 0.013516786t2 + 9.1467634810−3 t3 .
tional-order model using the Caputo–Fabrizio derivative
(19) [22, 23]. In addition, we propose to extend the results of the
International Journal of Differential Equations 7

present paper and combine them with the results in [6] [14] J.-H. He, “Some asymptotic methods for strongly nonlinear
(Figures 3–6). equations,” International Journal of Modern Physics B, vol. 20,
no. 10, pp. 1141–1199, 2006.
[15] Q. Ain and J.-H. He, “On two-scale dimension and its ap-
Data Availability plications,” Thermal Science, vol. 23, no. 3 Part B,
pp. 1707–1712, 2019.
The data used to support the findings of this study are [16] J. H. He and H. Latifizadeh, “A general numerical algorithm
available from the corresponding author upon request. for nonlinear differential equations by the variational iter-
ation method,” International Journal of Numerical Methods
for Heat and Fluid Flow, 2020.
Conflicts of Interest [17] J. K. Zhou, Differential Transformation and Its Applications for
Electrical Circuits, Huazhong University Press, Wuhan,
The authors declare that they have no conflicts of interest. China, 1986.
[18] M. H. A. Biswas and L. T. Paiva, “A SEIR model for control of
References infectious diseases with constraints,” Mathematical Biosci-
ences and Engineering, vol. 11, no. 4, 2014.
[1] P. Zhou, X. L. Yang, X. G. Wang et al., “A pneumonia [19] M. Inokuti, H. Sekine, and T. Mura, “General use of the
outbreak associated with a new coronavirus of probable bat Lagrange multiplier in nonlinear mathematical physics,” in
origin,” Nature, vol. 579, no. 7798, pp. 270–283, 2020. Variational Method in the Mechanics of Solids, S. Nemat-
[2] H. Lu, C. W. Stratton, and Y. W. Tang, “Outbreak of Nassed, Ed., Pergemon Press, Oxford, UK, pp. 156–162, 1978.
pneumonia of unknown etiology in Wuhan China: the [20] T. M. Chen, J. Rui, Q. P. Wang, Z. Y. Zhao, J. A. Cui, and
mystery and the miracle,” Journal of Medical Virology, vol. 92, L. Yin, “A mathematical model for simulating the phase-based
no. 4, pp. 401-402, 2020. transmissibility of a novel coronavirus,” Infectious Diseases of
[3] W. Ji, W. Wang, X. Zhao, J. Zai, and X. Li, “Homologous Poverty, vol. 9, p. 24, 2020.
recombination within the spike glycoprotein of the newly [21] S. V. Kanth and K. Aruna, “Two-dimensional differential
identified coronavirus may boost cross-species transmission transform method for solving linear and non-linear Schro-
from snake to human,” Journal of Medical Virology, vol. 92, dinger equations,” Chaos, Solitons & Fractals, vol. 41, no. 5,
no. 4, 2020. pp. 2277–2281, 2009.
[4] M. S. Abdo, K. Shah, H. A. Wahash, and S. K. Panchal, “On a [22] K. Shah, M. A. Alqudah, F. Jarad, and T. Abdeljawad, “Semi-
comprehensive model of the novel coronavirus (COVID-19) analytical study of Pine Wilt Disease model with convex rate
under Mittag-Leffler derivative,” Chaos, Solitons Fractals, under Caputo-Febrizio fractional order derivative,” Chaos,
vol. 135, p. 2020, 109867. Solitons & Fractals, vol. 135, Article ID 109754, 2020.
[5] F. Awawdeh, A. Adawi, and Z. Mustafa, “Solutions of the SIR [23] K. Shah, F. Jarad, and T. Abdeljawad, “On a nonlinear
models of epidemics using HAM,” Chaos, Solitons & Fractals, fractional order model of dengue fever disease under Caputo-
vol. 42, no. 5, pp. 3047–3052, 2009. Fabrizio derivative,” Alexandria Engineering Journal, 2020, In
[6] M. Sajid, Z. Abbas, N. Ali, and T. Javed, “A note on solutions press.
of the SIR models of epidemics using HAM,” ISRN Applied
Mathematics, vol. 2013, Article ID 457072, 4 pages, 2013.
[7] Y. Li, F. Haq, K. Shah, M. Shahzad, and G. u. Rahman,
“Numerical analysis of fractional order pine wilt disease
model with bilinear incident rate,” Journal of Mathematics
and Computer Science, vol. 17, pp. 420–428, 2017.
[8] E. M. Lotfi, M. Maziane, K. Hattaf, and N. Yousfi, “Partial
differential equations of an epidemic model with spatial
diffusion,” International Journal of Partial Differential
Equations, vol. 2014, Article ID 186437, 6 pages, 2014.
[9] A. Harir, S. Melliani, and L. S. Chadli, “Fuzzy fractional
evolution equations and fuzzy solution operators,” Advances
in Fuzzy Systems, vol. 2019, Article ID 5734190, 10 pages,
2019.
[10] J. He, “A new approach to nonlinear partial differential
equations,” Communications in Nonlinear Science and Nu-
merical Simulation, vol. 2, no. 4, pp. 230–235, 1997.
[11] J.-H. He, “Approximate solution of nonlinear differential
equations with convolution product nonlinearities,” Com-
puter Methods in Applied Mechanics and Engineering, vol. 167,
no. 1-2, pp. 69–73, 1998.
[12] A. Harir, S. Melliani, and L. S. Chadli, “Solving fuzzy Burgers
equation by variational iteration method,” Journal of Math-
ematics and Computer Science, vol. 21, no. 2, pp. 136–149,
2020.
[13] J. H. He, “A variational iteration approach to nonlinear
problems and its applications,” Mechanic Applications,
vol. 20, no. 1, pp. 30-31, 1998.