Skip to main content

Theoretical and semi-analytical results to a biological model under Atangana–Baleanu–Caputo fractional derivative

Abstract

This manuscript is related to finding a solution of the SIR model under Mittag-Leffler type derivative. For the required results, we use Laplace transform together with Adomian decomposition method (LADM). The mentioned method is a powerful tool to deal with various linear and nonlinear problems of “fractional order differential equations (FODEs)”. Also, we study some results devoted to qualitative theory for the concerned model. Computational results show the verification of the established analysis. Briefly, we state that qualitative theory for the existence of solution is important to ensure whether the considered problem has a solution or not. Further ensuring the existence of solution, we investigate approximate solution which is computed in the form of infinite series. The results are graphically displayed to analyze the adopted procedure for solving nonlinear FODEs under ABC derivative.

Introduction

Infectious diseases are spread by pathogenic microorganisms. These diseases can transmit from one person to another or from animals or birds. Despite all the advancement in medicine to control the disease, it is still a major threat to the community. Major causes of infectious diseases are: change in human behavior, use of antibiotic drugs in larger and denser cities. Mathematical models for the infectious diseases are the major tools to study the process through which diseases spread in a population [13]. These models are used for the predictions about the future to evaluate strategies to control the disease. First-time authors of [4], formulated a simple model in 1927 that described the connection between “susceptible, infected, and recovered individuals in a population abbreviated as SIR” given by

$$ \textstyle\begin{cases} \frac{dx}{d\mathrm{t}}=\alpha N-\delta x(\mathrm{t}) y(\mathrm{t})-a x( \mathrm{t}), \\ \frac{dy}{d\mathrm{t}}=\delta x(\mathrm{t})y(\mathrm{t})-(\beta +a+b)y( \mathrm{t}), \\ \frac{dz}{d\mathrm{t}}=\beta y(\mathrm{t})- a z(\mathrm{t}), \end{cases} $$
(1)

where α is the birth rate, \(N=x(\mathrm{t})+y(\mathrm{t})+z(\mathrm{t})\), a represents the unrelated death rate, b is a disease-related death rate, δ is infectious rate, and β is the removal rate. Further, x stands for the density of susceptible, y for infected, and z for recovered individuals, respectively. Onward of the said model has been investigated very well, see [58].

Riemann, Liouville, Euler, and Fourier have made a significant contribution in the eighteenth century in the area mentioned above. Various aspects of mathematical modeling may not be well described via ordinary calculus since derivatives of noninteger order are in fact definite integrals that provide accumulation. The concerned accumulation includes the corresponding integer counterpart as a special case. Further such operators permit greater freedom in degree as compared to integer order (for details, see [921]). In the said area, by considering different aspects, great work has been done in [1214]. Differential operator with noninteger order has not been uniquely defined. There are several definitions in the literature. On the basis of kernels, there are two concepts. One definition involving a singular kernel is often called power law, while the second one contains nonsingular kernel of exponential and Mittag-Leffler type. The differential operators involving Mittag-Leffler and exponential type kernel have been recently introduced by Atangana, Baleanu, Caputo, and Fabrizo (see [19, 2225]). This derivative exhibits the singular kernel by a nonsingular kernel [20, 21, 2629]. Since the differential and integral operators of ABC type are nonlocal and nonsingular, such operators reduce the complication in numerical analysis of many problems. Further in some problems, the mentioned operators play excellent roles in description of many hereditary and memory terms. Therefore, the mentioned operators have been considered in the recent time in an increasing way for investigating physical and biological problems. In this regard, a number of methods available in literature have been applied to compute solutions under these derivatives. To compute the approximate and analytical solution, a famous decomposition method was used as the best tool for many problems. Therefore, in this article, we utilize Laplace Adomian decomposition method (LADM) for the series solution of SIR model (1) under ABC derivative. We consider the biological model (1) and use the ABC derivative for the model with order μ such that \(\mu \in (0,1]\) as given by

$$ \textstyle\begin{cases} {}^{ABC}\mathcal{D}^{\mu }_{0}(x)(\mathrm{t})=\alpha N-\delta x( \mathrm{t}) y(\mathrm{t})-a x(\mathrm{t}), \\ {}^{ABC}\mathcal{D}^{\mu }_{0}(y)(\mathrm{t})=\delta x(\mathrm{t}) y( \mathrm{t})-(\beta +a+b) y(\mathrm{t}), \\ {}^{ABC}\mathcal{D}^{\mu }_{0}(z)(\mathrm{t})=\beta y(\mathrm{t})-a z( \mathrm{t}) \end{cases} $$
(2)

under the condition

$$ x(0)=N_{1},\qquad y(0)=N_{2} ,\qquad z(0)=N_{3}. $$

Then, we get the results in the form of an analytical solution of the SIR model. Moreover, we exhibit the approximate solution for distinct fractional order \(\mu \in (0,1]\). In addition, we study some results about the qualitative analysis and stability analysis for the concerned model. Further, the right-hand sides of model (2) vanish at zero as for the general problem in [20], Theorem 3.1. Via fixed point theory and nonlinear analysis, we establish some results regarding the existence and stability of solution. Then, we compute the required series solution via the proposed method for model (2).

Auxiliary results

We recall some fundamental results here.

Definition 1

Let \(\phi \in \mathcal{H}^{1}(0,\tau )\) and \(\mu \in (0,1]\), then the ABC derivative is defined as

$$ {}^{\mathcal{ABC}}\mathcal{D}_{0}^{\mu }\bigl(\phi (\mathrm{t})\bigr)= \frac{\mathcal{ABC}(\mu )}{(1-\mu )} \int _{0}^{\mathrm{t}} \frac{d}{d\theta }\phi (\theta )E_{\mu } \biggl[\frac{-\mu }{1-\mu }(t- \theta )^{\mu } \biggr]\,d \theta . $$
(3)

Here, \(E_{\mu }\) is known as a Mittag-Leffler function.

Definition 2

The fractional integral of ABC is

$$ {}^{\mathcal{AB}}\mathcal{I}_{0}^{\mu }\bigl(\phi (\mathrm{t})\bigr)= \frac{(1-\mu )}{\mathcal{ABC}(\mu )}\phi (\mathrm{t})+ \frac{\mu }{\mathcal{ABC}(\mu )\Gamma (\mu )} \int _{0}^{\mathrm{t}}\phi ( \theta ) (\mathrm{t}-\theta )^{\mu -1}\,d\theta , $$
(4)

while \(\mathcal{ABC}(\mu )\) is a normalization constant with \(\mathcal{ABC}(0)=1\), \(\mathcal{ABC}(1)=1\).

Definition 3

“The Laplace transform of the \(\mathcal{ABC}\) derivative of a function \(\phi (\mathrm{t})\)” is defined by

$$ \mathscr{L} \bigl[{}^{\mathcal{ABC}}\mathcal{D}_{0}^{\mu }\phi ( \mathrm{t}) \bigr]=\frac{\mathcal{ABC}(\mu )}{s^{\mu }(1-\mu )+\mu } \bigl[s^{\mu }\bigl[\phi ( \mathrm{t})\bigr]-s^{\mu -1}\phi (0) \bigr]. $$
(5)

Lemma 1

For \(0<\mu <1\), the solution of the problem

$$ \begin{gathered} {} {}^{\mathcal{ABC}}\mathcal{D}^{\mu }_{0} \phi (\mathrm{t})=g( \mathrm{t}),\quad \mathrm{t}\in [0,T], \\ \phi (0)=\phi _{0}, \end{gathered} $$

is provided by

$$ \phi (\mathrm{t})=\frac{(1-\mu )}{\mathcal{ABC}(\mu )}g(\mathrm{t})+ \frac{\mu }{\mathcal{ABC}(\mu )\Gamma (\mu )} \int _{0}^{\mathrm{t}}( \mathrm{t}-\theta )^{\mu -1}g( \theta )\,d\theta . $$
(6)

Definition 4

The operator \(\varphi _{k}: Y\rightarrow Y\) for \(k=1,2,3\) defined as

$$ \textstyle\begin{cases} {}^{\mathcal{ABC}}\mathcal{D}^{\mu }_{0}x(\mathrm{t})=\chi _{1}(x,y,z)( \mathrm{t}), \\ {}^{\mathcal{ABC}}\mathcal{D}^{\mu }_{0}y(\mathrm{t})=\chi _{2}(x,y,z)( \mathrm{t}), \\ {}^{\mathcal{ABC}}\mathcal{D}^{\mu }_{0}z(\mathrm{t})=\chi _{3}(x,y,z)( \mathrm{t}), \end{cases} $$
(7)

is Hyers–Ulam (HU) stable if, for any positive number \(c_{l}\) (\(l=1,2,3,\ldots 9\)), \(\Lambda _{l}\) (\(l=1,2,3\)) and for every solution \((\widehat{x},\widehat{y},\widehat{z})\in Y\) obeying the relation

$$ \textstyle\begin{cases} \Vert {x}-\widehat{x} \Vert \leq \Lambda _{1}, \\ \Vert {y}-\widehat{y} \Vert \leq \Lambda _{2}, \\ \Vert {z}-\widehat{z} \Vert \leq \Lambda _{3}, \end{cases} $$
(8)

with \(({x},{y},{z})\in Y\) of (7), the following hold:

$$ \textstyle\begin{cases} \Vert {x}-\widehat{x} \Vert \leq c_{1}\Lambda _{1}+c_{2}\Lambda _{2}+c_{3} \Lambda _{3}, \\ \Vert {y}-\widehat{y} \Vert \leq c_{4}\Lambda _{1}+c_{5}\Lambda _{2}+c_{6} \Lambda _{3}, \\ \Vert {z}-\widehat{z} \Vert \leq c_{7}\Lambda _{1}+c_{8}\Lambda _{2}+c_{9} \Lambda _{3}. \end{cases} $$
(9)

Definition 5

If \(\delta _{l}\) for \(l=1,2,3, \ldots n\) are eigenvalues of the matrix \(\mathcal{N}\), then the spectral radius is denoted as \(\Theta (\mathcal{N})\) and is defined as

$$ \Theta (\mathcal{N})=\max \bigl\{ \vert \delta _{l} \vert , \text{for } l=1,2,\ldots n\bigr\} . $$

Moreover, if \(\Theta (\mathcal{N})< 1\), this implies that \(\mathcal{N}\) tends to zero.

Theorem 1

For the operator \(\varphi _{k}: Y\rightarrow Y\) for \(k=1,2,3\), such that

$$ \textstyle\begin{cases} \Vert \varphi _{1}(x,y,z)-\varphi _{1}(\widehat{x},\widehat{y}, \widehat{z}) \Vert \leq c_{1} \Vert x-\widehat{x} \Vert +c_{2} \Vert y-\widehat{y} \Vert +c_{3} \Vert z-\widehat{z} \Vert , \\ \Vert \varphi _{2}(x,y,z)-\varphi _{2}(\widehat{x},\widehat{y}, \widehat{z}) \Vert \leq c_{4} \Vert x-\widehat{x} \Vert +c_{5} \Vert y-\widehat{y} \Vert +c_{6} \Vert z-\widehat{z} \Vert , \\ \Vert \varphi _{3}(x,y,z)-\varphi _{3}(\widehat{x},\widehat{y}, \widehat{z}) \Vert \leq c_{7} \Vert x-\widehat{x} \Vert +c_{8} \Vert y-\widehat{y} \Vert +c_{9} \Vert z-\widehat{z} \Vert , \\ \forall (x,y,z),\quad (\check{x},\check{y},\check{z})\in Y, \end{cases} $$
(10)

and the matrix

$$ \mathcal{N}= \begin{pmatrix} c_{1} & c_{2} &c_{3} \\ c_{4} & c_{5}&c_{6} \\ c_{7} &c_{8} &c_{9} \end{pmatrix} $$
(11)

tends to zero, then (7) is Hyers–Ulam stable.

Qualitative results for the proposed model (2)

In this part of the manuscript, we study qualitative results for problem (2). Here, we express right-hand sides of (2) as follows:

$$\begin{aligned}& \chi _{1}(\mathrm{t},x,y,z) = \alpha N-\delta x(\mathrm{t}) y( \mathrm{t})-a x(\mathrm{t}), \\& \chi _{2}(\mathrm{t},x,y,z) = \delta x(\mathrm{t}) y(\mathrm{t})-( \beta +a+b)y(\mathrm{t}), \\& \chi _{3}(\mathrm{t},x,y,z) = \beta y(\mathrm{t})-a z(\mathrm{t}). \end{aligned}$$
(12)

We select

$$\begin{aligned} \mathbf{M}_{j}= \sup _{A[d, b_{j}]} \bigl\Vert \varphi _{1}(\mathrm{t},x,y,z) \bigr\Vert \end{aligned}$$
(13)

such that

$$ A[d, b_{j}]=[\mathrm{t}-d, \mathrm{t}+d]\times [\mathrm{t}-b_{j}, \mathrm{t}+b_{j}]=B\times B_{j}\quad \text{for } j=1,2,3. $$

The concerned norm may be defined as

$$ \Vert \mathbf{Y} \Vert =\sup_{\mathrm{t}\in [\mathrm{t}-d, \mathrm{t}+b]} \bigl\vert \phi (\mathrm{t}) \bigr\vert . $$
(14)

Then the Picard operator is given as

$$ \mathbf{T}:A(B, B_{1}, B_{2}, B_{3})\rightarrow A(B, B_{1}, B_{2}, B_{3}). $$
(15)

We present the following theorem.

Theorem 2

In view of the Banach contraction theorem under the Picard operator as defined in (15), there exists at most one solution to the considered model (2).

Proof

In this regard, applying \({}^{\mathcal{AB}}\mathcal{I}^{\mu }\) on model (2), we obtain

$$ \textstyle\begin{cases} x(\mathrm{t})-x(0) = {}^{\mathcal{AB}}\mathcal{I}^{\mu }[ \chi _{1}(\mathrm{t},x,y,z)], \\ y(\mathrm{t})-y(0) = {}^{\mathcal{AB}}\mathcal{I}^{\mu }[\chi _{2}( \mathrm{t},x,y,z)], \\ z(\mathrm{t})-z(0) = {}^{\mathcal{AB}}\mathcal{I}^{\mu }[\chi _{3}( \mathrm{t},x,y,z)]. \end{cases} $$
(16)

Using Lemma 1 and writing (16) in a simple form, one has

$$ \mathbf{Y}(\mathrm{t})=\mathbf{Y}_{0}(\mathrm{t})+ \bigl[\Phi \bigl( \mathrm{t}, \mathbf{Y}(\mathrm{t})\bigr)-\Phi _{0}( \mathrm{t}) \bigr] \vartheta (\mu )+\bar{\vartheta }(\mu ) \int _{0}^{\mathrm{t}} ( \mathrm{t}-\zeta )^{\mu -1}\Phi \bigl(\zeta , \mathbf{Y}(\zeta )\bigr)\,d\zeta , $$
(17)

where

$$ \vartheta (\mu )=\frac{(1-\mu )}{\mathcal{ABC}(\mu )},\qquad \bar{\vartheta }( \mu )=\frac{\mu }{\mathcal{ABC}(\mu )\Gamma (\mu )}, $$

and

$$ \begin{gathered} \mathbf{Y}(\mathrm{t})= \textstyle\begin{cases}x(\mathrm{t}), \\ y(\mathrm{t}), \\ z(\mathrm{t}), \end{cases}\displaystyle \qquad \mathbf{Y}_{0}(\mathrm{t})= \textstyle\begin{cases}x(0), \\ y(0), \\ z(0), \end{cases}\displaystyle \qquad \Phi \bigl(\mathrm{t}, \mathbf{Y}(\mathrm{t})\bigr)= \textstyle\begin{cases}\chi _{1}(\mathrm{t},x,y,z), \\ \chi _{2}(\mathrm{t},x,y,z), \\ \chi _{3}(\mathrm{t},x,y,z), \end{cases}\displaystyle \\ \Phi _{0}(\mathrm{t})= \textstyle\begin{cases}\chi _{1}(0, x(0), y(0), z(0)), \\ \chi _{2}(0, x(0), y(0), z(0)), \\ \chi _{3}(0, x(0), y(0), z(0)). \end{cases}\displaystyle \end{gathered} $$
(18)

Using (17) and (18), the operator in (15) is defined as follows:

$$ \mathbf{T}\mathbf{Y}(\mathrm{t})=\mathbf{Y}_{0}( \mathrm{t})+ \bigl[ \Phi \bigl(\mathrm{t}, \mathbf{Y}(\mathrm{t})\bigr)-\Phi _{0}(\mathrm{t}) \bigr] \vartheta (\mu )+\bar{\vartheta }(\mu ) \int _{0}^{\mathrm{t}}(\mathrm{t}- \zeta )^{\nu -1} \Phi \bigl(\zeta , \mathbf{Y}(\zeta )\bigr)\,d\zeta . $$
(19)

Thus, the model under our study satisfies the result

$$\begin{aligned}& \Vert \mathbf{Y} \Vert \leq \max \{d_{1}, d_{2}, d_{3}\}, \end{aligned}$$
(20)
$$\begin{aligned}& \begin{aligned}[t] & \bigl\Vert T \mathbf{Y}(\mathrm{t})-\mathbf{Y}_{0}(\mathrm{t}) \bigr\Vert \\ &\quad =\sup_{ \mathrm{t}\in B} \biggl\vert \Phi \bigl(\mathrm{t}, \mathbf{Y}(\mathrm{t})\bigr) \vartheta (\mu )+\bar{\vartheta }(\mu ) \int _{0}^{\mathrm{t}}(\mathrm{t}- \zeta )^{\mu -1}\Phi \bigl(\xi , \mathbf{Y}(\zeta )\bigr)\,d\zeta \biggr\vert \\ &\quad \leq \sup_{\mathrm{t}\in B}\vartheta (\mu ) \bigl\vert \Phi \bigl(\mathrm{t}, \mathbf{Y}(\mathrm{t})\bigr) \bigr\vert +\sup _{\mathrm{t}\in B}\bar{\vartheta }(\mu ) \int _{0}^{\mathrm{t}}(\mathrm{t}-\zeta )^{\mu -1} \bigl\vert \Phi \bigl(\zeta , \mathbf{Y}(\zeta )\bigr) \bigr\vert \,d\zeta \\ &\quad \leq \vartheta (\mu )\mathbf{M}+\bar{\vartheta }(\mu ) \mathrm{t}^{ \mu }_{0}\mathbf{M}d, \quad\mathbf{M}=\max \{ \mathbf{M}_{j}\} \text{ for } j=1,2,3, \mathrm{t}_{0}= \max \{\mathrm{t}\in B\} \\ &\quad < d\mathbf{M}\leq \max \{d_{1}, d_{2}, d_{3}\}=\bar{d},\quad \text{where } d= \frac{(\Gamma (\mu )+ \mu \mathrm{t}^{\mu }_{0})}{\mathcal{ABC}(\mu )\Gamma (\mu )}, \end{aligned} \end{aligned}$$
(21)

such that

$$ d< \frac{\bar{d}}{\mathbf{M}}. $$

On further simplification, one has

$$\begin{aligned} \Vert \mathbf{T}\mathbf{Y}_{1}- \mathbf{T}\mathbf{Y}_{2} \Vert =\sup_{ \mathrm{t}\in B} \vert \mathbf{Y}_{1}-\mathbf{Y}_{2} \vert . \end{aligned}$$
(22)

To compute (22), we proceed as follows:

$$\begin{aligned} \Vert \mathbf{T}\mathbf{Y}_{1}-\mathbf{T} \mathbf{Y}_{2} \Vert =&\sup_{ \mathrm{t}\in B} \biggl\vert \vartheta (\mu ) \bigl(\Phi \bigl(\mathrm{t}, \mathbf{Y}_{1}(\mathrm{t}) \bigr)-\Phi \bigl(\mathrm{t}, \mathbf{Y}_{2}( \mathrm{t})\bigr) \bigr) \\ &{}+\bar{\vartheta }(\mu ) \int _{0}^{\mathrm{t}}(\mathrm{t}-\zeta )^{\nu -1} \bigl( \Phi \bigl(\zeta , \mathbf{Y}_{1}(\zeta )\bigr)-\Phi \bigl(\zeta , \mathbf{Y}_{2}( \zeta )\bigr) \bigr)\,d\zeta \biggr\vert , \\ \leq& \vartheta (\mu )k \Vert \mathbf{Y}_{1}-\mathbf{Y}_{2} \Vert + \bar{\vartheta }(\mu )k\mathrm{t}^{\mu } \Vert \mathbf{Y}_{1}-\mathbf{Y}_{2} \Vert ,\quad \text{with } k< 1 \\ \leq& \bigl\{ \vartheta (\mu )k+\bar{\vartheta }(\mu )\mathrm{t}^{\mu } k \bigr\} \Vert \mathbf{Y}_{1}-\mathbf{Y}_{2} \Vert , \\ \leq& d k \Vert \mathbf{Y}_{1}-\mathbf{Y}_{2} \Vert . \end{aligned}$$
(23)

As Φ is a contraction, so we have \(kd<1\), thus T is a contraction. Therefore, our concerned problem (18) has the required solution. □

Stability results

Theorem 3

If \(d<1 \) holds, then the matrix \(\mathcal{N}\) also converging to zero is Hyers–Ulam stable.

Proof

Taking any two solutions \((x,y,z)\), \((\widehat{x},\widehat{y},\widehat{z})\), we have

$$\begin{aligned} & \bigl\Vert \chi _{1}(x,y,z)-\chi _{1}(\widehat{x},\widehat{y}, \widehat{z}) \bigr\Vert \\ &\quad \leq \bigl[ \bigl\vert \Phi \bigl(\mathrm{t}, \mathbf{(x,y,z)}( \mathrm{t})\bigr)-\Phi \bigl(\mathrm{t}, \mathbf{(\widehat{x},\widehat{y},\widehat{z})}(\mathrm{t})\bigr) \bigr\vert \bigr] \vartheta (\mu ) \\ &\qquad {}+\bar{\vartheta }(\mu ) \int _{0}^{\mathrm{t}} (\mathrm{t}-\zeta )^{\mu -1} \bigl\vert \Phi \bigl(\zeta , \mathbf{(x,y,z)}(\zeta )\bigr)-\Phi \bigl(\zeta , \mathbf{(\widehat{x},\widehat{y},\widehat{z})}(\zeta )\bigr) \bigr\vert \,d\zeta \\ &\quad \leq \vartheta (\mu )k \bigl[ \Vert x-\widehat{x} \Vert + \Vert y- \widehat{y} \Vert + \Vert z- \widehat{z} \Vert \bigr] \\ &\qquad {}+\bar{\vartheta }(\mu ) \mathrm{t}^{\mu }k \bigl[ \Vert x- \widehat{x} \Vert + \Vert y- \widehat{y} \Vert + \Vert z-\widehat{z} \Vert \bigr] \\ &\quad \leq \bigl(\vartheta (\mu )+\bar{\vartheta }(\mu ) \mathrm{t}^{\mu } \bigr)k \Vert x-\widehat{x} \Vert \\ &\qquad {}+ \bigl(\vartheta (\mu )+\bar{\vartheta }(\mu ) \mathrm{t}^{\mu } \bigr)k \Vert y-\widehat{y} \Vert \\ &\qquad {}+ (\vartheta (\mu )+\bar{\vartheta }(\mu ) \mathrm{t}^{\mu }k \Vert z- \widehat{z} \Vert \\ &\quad \leq c_{1} \Vert x-\widehat{x} \Vert +c_{2} \Vert y- \widehat{y} \Vert +c_{3} \Vert z- \widehat{z} \Vert . \end{aligned}$$
(24)

In the same fashion, one has

$$ \begin{gathered} \bigl\Vert \chi _{2}(x,y,z)-\chi _{2}(\widehat{x},\widehat{y}, \widehat{z}) \bigr\Vert \leq c_{4} \Vert x-\widehat{x} \Vert +c_{5} \Vert y- \widehat{y} \Vert +c_{6} \Vert z-\widehat{z} \Vert , \\ \bigl\Vert \chi _{3}(x,y,z)-\chi _{3}(\widehat{x}, \widehat{y},\widehat{z}) \bigr\Vert \leq c_{7} \Vert x-\widehat{x} \Vert +c_{8} \Vert y-\widehat{y} \Vert +c_{9} \Vert z- \widehat{z} \Vert , \end{gathered} $$
(25)

where

$$ c_{i}= \bigl(\vartheta (\mu )+\bar{\vartheta }(\mu ) \mathrm{t}^{\mu } \bigr)k\quad \text{for each }i=1,2,\ldots,9. $$

Now, the matrix \(\mathcal{N}\) given by

$$ \mathcal{N}= \begin{pmatrix} c_{1} & c_{2} &c_{3} \\ c_{4} & c_{5}&c_{6} \\ c_{7} &c_{8} &c_{9} \end{pmatrix} $$
(26)

converges to zero. Hence, the system is Hyers–Ulam stable. □

Analytical results for the proposed model

Here, we are going to apply LADM to obtain general results for the considered model (2).

$$ \textstyle\begin{cases} \mathcal{L}[x(\mathrm{t})]=\frac{x(0)}{s}+ \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L}[ \alpha N-\delta x(\mathrm{t}) y(\mathrm{t})- ax(\mathrm{t})], \\ \mathcal{L}[y(\mathrm{t})]=\frac{y(0)}{s}+ \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L}[ \delta x(\mathrm{t}) y(\mathrm{t})-(\beta +a+b) y(\mathrm{t})], \\ \mathcal{L}[z(\mathrm{t})]=\frac{z(0)}{s}+ \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L}[ \beta y(\mathrm{t})-a z(\mathrm{t})]. \end{cases} $$
(27)

Now, we are going to consider \(x(\mathrm{t})\), \(y(\mathrm{t})\), \(z(\mathrm{t})\) in terms of infinite series as follows:

$$ x(\mathrm{t})=\sum_{q=0}^{\infty }x_{q}( \mathrm{t}),\qquad y(\mathrm{t})= \sum_{q=0}^{\infty }y_{q}( \mathrm{t}),\qquad z(\mathrm{t})=\sum_{q=0}^{\infty }z_{q}( \mathrm{t}). $$
(28)

We resolve nonlinear terms as follows:

$$ x(\mathrm{t})y(\mathrm{t})=\sum _{q=0}^{\infty } A_{q} (x,y), $$
(29)

where \(A_{q}(x,y)\) can be defined as

$$ A_{q}(x,y)=\frac{1}{q!}\frac{d^{q}}{d\lambda ^{q}} \Biggl[\sum _{j=0}^{p} \lambda ^{j}x_{j}( \mathrm{t})\sum_{j=0}^{p}\lambda ^{j}y_{j}( \mathrm{t}) \Biggr]\Bigg|_{\lambda =0}. $$

Hence, by using (28) and (29), our system (27) becomes

$$ \textstyle\begin{cases} \mathcal{L} [\sum_{q=0}^{\infty }x_{q}(\mathrm{t}) ]= \frac{x(0)}{s}+\frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )} \mathcal{L} [\alpha N-\delta \sum_{q=0}^{\infty }A_{q}(x,y)-a\sum_{q=0}^{\infty }x_{q} ], \\ \mathcal{L} [\sum_{q=0}^{\infty }y_{q}(\mathrm{t}) ]= \frac{y(0)}{s}+\frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )} \mathcal{L} [\delta \sum_{q=0}^{\infty }A_{q}(x,y)-(\beta +a+b) \sum_{q=0}^{\infty }y_{q} ], \\ \mathcal{L} [\sum_{q=0}^{\infty }z_{q}(\mathrm{t}) ]= \frac{z(0)}{s}+\frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )} \mathcal{L} [\beta \sum_{q=0}^{\infty }y_{q}-a\sum_{q=0}^{\infty }z_{q} ]. \end{cases} $$
(30)

Upon comparing terms wise (30), one has

$$ \textstyle\begin{cases} \mathcal{L}[x_{0}(\mathrm{t})]=N_{1},\qquad \mathcal{L}[y_{0}(\mathrm{t})]=N_{2},\qquad \mathcal{L}[z_{0}(\mathrm{t})]=N_{3}, \\ \mathcal{L}[x_{1}(\mathrm{t})]= \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L} [ \alpha N-\delta A_{0}(x,y)-a N_{1} ], \\ \mathcal{L}[y_{1}(\mathrm{t})] = \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L} [ \delta A_{0}(x,y)-(\beta +a+b)N_{2} ], \\ \mathcal{L}[z_{1}(\mathrm{t})] = \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L} [ \beta N_{2}-a N_{3} ], \\ \mathcal{L}[x_{2}(\mathrm{t})]= \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L} [ \alpha N-\delta A_{2}(x,y)-a x_{1} ], \\ \mathcal{L}[y_{2}(\mathrm{t})] = \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L} [ \delta A_{q}(x,y)-(\beta +a+b) y_{1} ], \\ \mathcal{L}[z_{2}(\mathrm{t})] = \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L} [ \beta y_{1}-a z_{1} ], \\ \vdots \\ \mathcal{L}[x_{q+1}(\mathrm{t})]= \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L} [ \alpha N-\delta A_{q}(u,x)-a x_{q} ], \\ \mathcal{L}[y_{q+1}(\mathrm{t})] = \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L} [ \delta A_{q}(x,y)-(\beta +a+b)y_{q} ], \\ \mathcal{L}[z_{q+1}(\mathrm{t})] = \frac{s^{\mu }(1-\mu )+\mu }{s^{\mu }\mathcal{ABC}(\mu )}\mathcal{L} [ \beta y_{q}-a z_{q} ], \quad q\geq 0. \end{cases} $$
(31)

Simplifying the Laplace transform in (31), we get

$$ \textstyle\begin{cases} x_{0}(\mathrm{t})=N_{1},\qquad y_{0}(\mathrm{t})=N_{2},\qquad z_{0}( \mathrm{t})=N_{3}, \\ x_{1}(\mathrm{t})= (\alpha N-\delta N_{1} N_{2}-a N_{1} ) (1-\mu +\frac{\mu \mathrm{t}^{\mu }}{\Gamma (\mu )} ), \\ y_{1}(\mathrm{t})= (1-\mu + \frac{\mu \mathrm{t}^{\mu }}{\Gamma (\mu )} ) (\delta N_{1}N_{2}-( \beta +a+b)N_{2} ), \\ z_{1}(\mathrm{t})= (\beta N_{1}-a N_{3} ) (1-\mu + \frac{\mu \mathrm{t}^{\mu }}{\Gamma (\mu )} ), \\ x_{2}(\mathrm{t})= (1-\mu + \frac{\mu \mathrm{t}^{\mu }}{\Gamma (\mu )} )\alpha N- ((\delta N_{2}+ \alpha )(\alpha N-\delta N_{1} N_{2}-\alpha N_{1}) ) [ (1- \mu +\frac{\mu \mathrm{t}^{\mu }}{\Gamma (\mu )} ) ]^{2}, \\ y_{2}(\mathrm{t})= ((\delta N_{1}-a-\beta -b)(\delta N_{1} N_{2}+(a+b- \beta )N_{2})\\ \hphantom{y_{2}(\mathrm{t})=}{}+\delta N_{2}(\alpha N-\delta N_{1}N_{2}-\alpha N_{1}) ) [ (1-\mu +\frac{\mu \mathrm{t}^{\mu }}{\Gamma (\mu )} ) ]^{2}, \\ z_{2}(\mathrm{t})= [1-\mu + \frac{\mu \mathrm{t}^{\mu }}{\Gamma (\mu )} ]^{2} (\beta ( \delta N_{1}N_{2}-(\beta +a+b)N_{2})-a(\beta N_{1}-aN_{3}) ), \end{cases} $$

and so on.

In this way, the reaming terms may be computed. Finally, the required solutions can be expressed as follows:

$$ x(\mathrm{t})=\sum_{k=0}^{\infty }x_{k}( \mathrm{t}),\qquad y(\mathrm{t})=\sum_{k=0}^{\infty }y_{k}( \mathrm{t}),\qquad z(\mathrm{t})=\sum_{k=0}^{\infty }z_{k}( \mathrm{t}), $$
(32)

Computational results

In this section of the paper, our computational results about a series solution of the concerned model are represented. To obtain the main goal, we apply LADM for the solution corresponding to the values given in Table 1. In view of Table 1, we exhibit the results, which are represented in (32) for different fractional order in the following Figs. 13 using Matlab. Figures 13 show the graphs for the population of three compartments (susceptible, infected, and recovered) for distinct values of μ. One can observe that as we increase the values of μ, the corresponding solutions converge to the solution at integer order. Moreover, with passage of time the population of the susceptible class is decreasing when starts from 0.1 in given time of 250 days under proper cure or vaccination. The decreasing process of a class is different at different fractional order, while the density of infected papulation is decreasing in given time of 250 days. Hence, the recovered class is increasing with passage of time. As we observe, for different values of μ, the distinct trajectories are obtained as exhibited in Figs. 13. Increasing or decreasing (growth or decay) process is somewhat greater at small fractional order values as compared to those of greater fractional order.

Figure 1
figure1

Dynamical behavior of “susceptible papulation” corresponding to various values of μ

Figure 2
figure2

Dynamical behavior of “infected papulation” corresponding to various values of μ

Figure 3
figure3

Dynamical behavior of “recovered papulation” corresponding to various values of μ

Table 1 Parameters and their numerical values in model (2)

Concluding remarks

We have discussed LADM for a biological model of the SIR model using the ABC operator. Also, we developed some results about the qualitative theory and Hyers–Ulam stability analysis. The methodology utilized here for dynamical problems under ABC operator of derivative is very rarely applied in the literature. Further on providing some graphs of approximate results, we have illustrated the procedure. The mentioned tool may be used in the future to handle more complicated problems under the aforementioned operator.

Availability of data and materials

The authors confirm that the data supporting the findings of this study are available within the article cited therein.

References

  1. 1.

    Chanprasopchai, P., Tang, I.M., Pongsumpun, P.: SIR model for Dengue disease with effect of Dengue vaccination. Comput. Math. Methods Med. 2018, Article ID 9861572 (2018)

    MathSciNet  Article  Google Scholar 

  2. 2.

    Shulgin, B., Stone, L., Agur, Z.: Pulse vaccination strategy in SIR epidemic model. Bull. Math. Biol. 60(6), 1123–1148 (1988)

    Article  Google Scholar 

  3. 3.

    Zaman, G., Han, Y., Kang, I., Jung, H.: Stability analysis and optimal vaccination of SIR epidemic model. Biosystems 93(3), 240–249 (2008)

    Article  Google Scholar 

  4. 4.

    Kermack, W.O., McKendrick, A.G.: Contribution to the mathematical theory of epidemics. J. R. Stat. Soc. A 115, 700–721 (1927)

    MATH  Google Scholar 

  5. 5.

    Allen, L.J.S., Burgin, A.M.: Comparison of deterministic and stochastic SIS and SIR models in discrete time. Math. Biosci. 163(1), 1–33 (2000)

    MathSciNet  Article  Google Scholar 

  6. 6.

    Pandey, A., Mubayi, A., Medlock, J.: Comparing vector-host and SIR models for Dengue transmission. Math. Biosci. 246(2), 252–259 (2013)

    MathSciNet  Article  Google Scholar 

  7. 7.

    Angstmann, C.N., Henry, B.I., McGann, A.V.: A fractional-order infectivity SIR model. Phys. A, Stat. Mech. Appl. 452, 86–93 (2016)

    MathSciNet  Article  Google Scholar 

  8. 8.

    Song, H., Jiang, W., Liu, S.: Global dynamics of two heterogeneous SIR models with nonlinear incidence and delays. Int. J. Biomath. 9(3), 1650046 (2016)

    MathSciNet  Article  Google Scholar 

  9. 9.

    Podlubny, I.: Fractional Differential Equations, Mathematics in Science and Engineering. Academic Press, New York (1999)

    Google Scholar 

  10. 10.

    Kilbas, A.A., Srivastava, H., Trujillo, J.: Theory and Application of Fractional Differential Equations. North Holland Mathematics Studies, vol. 204. Elseveir, Amsterdam (2006)

    Google Scholar 

  11. 11.

    Kilbas, A.A., Marichev, O.I., Samko, S.G.: Fractional Integrals and Derivatives (Theory and Applications). Gordon & Breach, New York (1993)

    Google Scholar 

  12. 12.

    Toledo-Hernandez, R., Rico-Ramirez, V., Iglesias-Silva, G.A., Diwekar, U.M.: A fractional calculus approach to the dynamic optimization of biological reactive systems. Part I: fractional models for biological reactions. Chem. Eng. Sci. 117, 217–228 (2014)

    Article  Google Scholar 

  13. 13.

    Miller, K.S., Ross, B.: An Introduction to the Fractional Calculus and Fractional Differential Equations. Wiley, New York (1993)

    Google Scholar 

  14. 14.

    Lakshmikantham, V., Leela, S., Vasundhara, J.: Theory of Fractional Dynamic Systems. Cambridge Academic, Cambridge (2009)

    Google Scholar 

  15. 15.

    Rossikhin, Y.A., Shitikova, M.V.: Applications of fractional calculus to dynamic problems of linear and nonlinear hereditary mechanics of solids. Appl. Mech. Rev. 50, 15–67 (1997)

    Article  Google Scholar 

  16. 16.

    Shah, K., Abdeljawad, F.J.T.: On a nonlinear fractional order model of Dengue fever disease under Caputo–Fabrizio derivative. Alex. Eng. J. 59(4), 2305–2313 (2020)

    Article  Google Scholar 

  17. 17.

    Biazar, J.: Solution of the epidemic model by Adomian decomposition method. Appl. Math. Comput. 173, 1101–1106 (2006)

    MathSciNet  MATH  Google Scholar 

  18. 18.

    Rafei, M., Ganji, D.D., Daniali, H.: Solution of the epidemic model by homotopy perturbation method. Appl. Math. Comput. 187, 1056–1062 (2007)

    MathSciNet  MATH  Google Scholar 

  19. 19.

    Al-Refai, M., Abdeljawad, T.: Analysis of the fractional diffusion equations with fractional derivative of non-singular kernel. Adv. Differ. Equ. 2017, 315 (2017)

    MathSciNet  Article  Google Scholar 

  20. 20.

    Jarad, F., Abdeljawad, T., Hammouch, Z.: On a class of ordinary differential equations in the frame of Atangana–Baleanu fractional derivative. Chaos Solitons Fractals 117, 16–20 (2018)

    MathSciNet  Article  Google Scholar 

  21. 21.

    Shatha, H.: Atangana–Baleanu fractional framework of reproducing kernel technique in solving fractional population dynamics system. Chaos Solitons Fractals 133, 109–624 (2020)

    MathSciNet  Google Scholar 

  22. 22.

    Atangana, A., Koca, I.: Chaos in a simple nonlinear system with Atangana–Baleanu derivatives with fractional order. Chaos Solitons Fractals 89, 447–454 (2016)

    MathSciNet  Article  Google Scholar 

  23. 23.

    Atangana, A., Gómez-Aguilar, J.F.: Numerical approximation of Riemann–Liouville definition of fractional derivative: from Riemann–Liouville to Atangana–Baleanu. Numer. Methods Partial Differ. Equ. 34(5), 1502–1523 (2018)

    MathSciNet  Article  Google Scholar 

  24. 24.

    Gómez-Aguilar, J.F, Atangana, A., Morales-Delgado, V.F.: Electrical circuits RC, LC, and RL described by Atangana–Baleanu fractional derivatives. Int. J. Circuit Theory Appl. 45(11), 1514–1533 (2017)

    Article  Google Scholar 

  25. 25.

    Koca, I., Atangana, A.: Solutions of Cattaneo–Hristov model of elastic heat diffusion with Caputo–Fabrizio and Atangana–Baleanu fractional derivatives. Therm. Sci. 21(6), 2299–2305 (2017)

    Article  Google Scholar 

  26. 26.

    Arqub, O.A.: Atangana–Baleanu fractional approach to the solutions of Bagley–Torvik and Painlevé equations in Hilbert space. Chaos Solitons Fractals 117, 161–167 (2018)

    MathSciNet  Article  Google Scholar 

  27. 27.

    Abdeljawad, T., et al.: Analysis of some generalized ABC-fractional logistic models. Alex. Eng. J. 59(4), 2141–2148 (2020)

    Article  Google Scholar 

  28. 28.

    Khan, H., Khan, A., Jarad, F., Shah, A.: Existence and data dependence theorems for solutions of an ABC-fractional order impulsive system. Chaos Solitons Fractals 131, 109477 (2020)

    MathSciNet  Article  Google Scholar 

  29. 29.

    Khan, H., Jarad, F., Abdeljawad, T., Khan, A.: A singular ABC-fractional differential equation with p-Laplacian operator. Chaos Solitons Fractals 129, 56–61 (2019)

    MathSciNet  Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Contributions

Authors have equally contributed in preparing this manuscript. All authors read and approved the final manuscript.

Corresponding author

Correspondence to Anwar Zeb.

Ethics declarations

Competing interests

The authors declare that they have no competing interests.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Zeb, A., Nazir, G., Shah, K. et al. Theoretical and semi-analytical results to a biological model under Atangana–Baleanu–Caputo fractional derivative. Adv Differ Equ 2020, 654 (2020). https://doi.org/10.1186/s13662-020-03117-7

Download citation

Keywords

  • Fractional order differential equation
  • Atangana–Baleanu–Caputo fractional derivative
  • Laplace Adomian decomposition method
  • SIR model
\