Open Mathematics 2021; 19: 760–772

Research Article

Ahmed Alsaedi, Bashir Ahmad*, Badrah Alghamdi, and Sotiris K. Ntouyas

On a nonlinear system of Riemann-Liouville

fractional differential equations with
semi-coupled integro-multipoint boundary
conditions
https://doi.org/10.1515/math-2021-0069
received February 16, 2021; accepted June 29, 2021

Abstract: We study a nonlinear system of Riemann-Liouville fractional differential equations equipped with
nonseparated semi-coupled integro-multipoint boundary conditions. We make use of the tools of the fixed-
point theory to obtain the desired results, which are well-supported with numerical examples.

Keywords: Riemann-Liouville fractional derivative, integro-differential inclusions, nonlocal multi-point

boundary conditions, existence, fixed point theorems

MSC 2020: 34A08, 34B15

1 Introduction
Nonlinear boundary value problems involving Riemann-Liouville fractional derivatives have been studied
by many researchers, for example, see [1–6]. In a recent article [7], the authors discussed the existence and
Ulam-type stability for nonlinear Riemann-Liouville fractional differential equations with constant delay.
The nonlocal nature of fractional derivative operators significantly contributed to the popularity of
fractional calculus. Nowadays, one can find extensive application of fractional-order operators in the
mathematical models of several real-world phenomena occurring in physical and applied sciences, such
as continuum mechanics [8], bioengineering [9], financial economics [10], fractals [11], etc.
The topic of fractional differential systems also received considerable attention in view of their applica-
tions in diverse fields such as anomalous diffusion [12], disease models [13] biological models [14], hybrid
systems [15], rheological models [16], diffusion systems [17], ecological models [18], etc. For theoretical
details of such systems, see [19–29].
In this paper, we investigate the existence of solutions for a system of nonlinear Riemann-Liouville
fractional differential equations
⎧ D αu(t ) = F (t , u(t ) , v (t )) , 1 < α ≤ 2, t ∈ [0, T ],
⎨ β (1)
⎩ D v (t ) = G (t , u(t ) , v (t )) , 1 < β ≤ 2, t ∈ [0, T ],


* Corresponding author: Bashir Ahmad, Department of Mathematics, Faculty of Science, Nonlinear Analysis and Applied
Mathematics (NAAM)-Research Group, King Abdulaziz University, P.O. Box 80203, Jeddah 21589, Saudi Arabia,
e-mail: bashirahmad-qau@yahoo.com
Ahmed Alsaedi: Department of Mathematics, Faculty of Science, Nonlinear Analysis and Applied Mathematics (NAAM)-Research
Group, King Abdulaziz University, P.O. Box 80203, Jeddah 21589, Saudi Arabia, e-mail: aalsaedi@hotmail.com
Badrah Alghamdi: Department of Mathematics, Faculty of Science, Nonlinear Analysis and Applied Mathematics (NAAM)-
Research Group, King Abdulaziz University, P.O. Box 80203, Jeddah 21589, Saudi Arabia, e-mail: bsalghamdi@kau.edu.sa
Sotiris K. Ntouyas: Department of Mathematics, University of Ioannina, 451 10, Ioannina, Greece, e-mail: sntouyas@uoi.gr

Open Access. © 2021 Ahmed Alsaedi et al., published by De Gruyter. This work is licensed under the Creative Commons
Attribution 4.0 International License.
 Semi-coupled fractional differential systems  761

equipped with nonlocal semi-coupled fractional integro-multipoint boundary conditions of the form:
⎧ D α − 2 u(0+) + a0D α − 2 u(T −) = λ1 ,
⎪ m
⎪ α−1 +
⎪ D u ( 0 ) + a1D α − 1u(T −) = νI α − 1v (η ) +
1 ∑ μ i v (ξ i ) ,
⎪ i=1
⎨ β−2 + (2)
⎪ D v (0 ) + b0D β − 2 v (T −) = λ2 ,
⎪ n
⎪ D β − 1v (0+) + b1D β − 1v (T −) = μI β − 1u(η2 ) + ∑ σju(ζ j ) ,
⎪
⎩ j=1

where D χ is the Riemann-Liouville fractional derivative of order χ ∈ {α, β } , 0 < η1 < η2 < ξ1 < ξ2 < … <
ξm < ζ1 < ζ2 < … < ζm < T , a0 , a1, b0, b1, ν , μ, λ1, λ2 , μi (i = 1, 2, … , m), σj ( j = 1, 2, … , n) are real constants
and a0 ≠ −1, a1 ≠ −1, b0 ≠ −1, and F , G : [0, T ] ×  ×  ⟶  are continuous functions.
Here we emphasize that our results are new with respect to the semi-coupled boundary conditions (2)
and enrich the related literature on the topic.
In Section 2, we prove an auxiliary lemma dealing with the linear variant of systems (1)–(2), which plays
a fundamental role in establishing the existence and uniqueness results for the given nonlinear problem,
presented in Section 3. Illustrative examples demonstrating the application of the obtained results are given
in Section 4.

2 Preliminaries
Let us begin this section with some related definitions [30,31].

Definition 2.1. The (left) Riemann-Liouville fractional integral of order χ > 0 for ϱ ∈ L1[a , b], −∞ < a ≤
t ≤ b < +∞, existing almost everywhere on [a , b], is defined as
t
1
(Iaχ+ϱ)(t ) =
Γ (χ )
∫(t − s) χ−1ϱ(s)ds, χ > 0,
a

where Γ(⋅ ) is the gamma function and (Ia0+ϱ)(x ) = ϱ(x ).

Definition 2.2. For ϱ, ϱ(m) ∈ L1[a , b], the (left) Riemann-Liouville fractional derivative Daχ+ϱ of order
χ ∈ (m − 1, m], m ∈  is defined as
t
1 dm
Daχ+ϱ(t ) =
Γ(m − χ ) d t m
∫(t − s)m−1−χ ϱ(s)ds, −∞ ≤ a < t < b ≤ +∞ .
a

In the sequel, we will write Riemann-Liouville fractional integral and derivative operators as I χ and D χ
instead of Iaχ+ and Daχ+, respectively.

Lemma 2.1. Let ψ1, ψ2 ∈ C[0, T ] ∩ L[0, T ] and Λ ≠ 0. Then the unique solution of the following linear system
⎧ D αu(t ) = ψ (t ) ,
⎪ 1 < α ≤ 2, t ∈ [0, T ],
1
⎨ (3)
β v (t ) = ψ (t ) ,
⎩
⎪
D 2 1 < β ≤ 2, t ∈ [0, T ],

subject to the boundary conditions (2) is given by

u(t ) = λ1ρ1 (t ) + λ2 ω1ρ2 (t ) − a1ρ2 (t )I 1ψ1(T ) − b1ρ3 (t )I 1ψ2 (T ) − a0ρ1 (t )I 2ψ1(T ) − b0ω1ρ2 (t )I 2ψ2 (T )
n m
(4)
+ μρ3 (t )I α + β − 1ψ1(η2 ) + νρ2 (t )I α + β − 1ψ2 (η1) + ρ3 (t ) ∑ σjI αψ1(ζ j ) + ρ2 (t )∑μi I βψ2 (ξi ) + I αψ1(t ) ,
j=1 i=1
762  Ahmed Alsaedi et al.

v (t ) = λ2 ρ1∗(t ) + λ1ν1ρ2∗(t ) − a1ρ3∗(t )I 1ψ1(T ) − b1ρ2∗(t )I 1ψ2 (T ) − a0ν1ρ2∗(t )I 2ψ1(T ) − b0ρ1∗(t )I 2ψ2 (T )
n m
(5)
+ μρ2∗(t )I α + β − 1ψ1(η2 ) + νρ3∗(t )I α + β − 1ψ2 (η1) + ρ2∗(t ) ∑ σjI αψ1(ζ j ) + ρ3∗(t )∑μi I βψ2 (ξi ) + I βψ2 (t ) ,
j=1 i=1

where
⎡ 1 ν a0T Γ(α) ⎛ b0T Γ(β ) ⎞⎤ ν ⎡ b0T Γ(β ) ⎤
ρ1 (t ) = t α − 2⎢ + 1 ⎜ω 2 − ω1⎟⎥ − t α − 1 1 ⎢ω2 − ω1⎥ ,
⎣ (1 + a0)Γ(α − 1) Λ (1 + a0)Γ(α − 1) ⎝ (1 + a1)Γ(α) ⎠⎦ Λ ⎣ (1 + a1)Γ(α) ⎦
1 ⎡ α−2 a0T (b1 + 1)Γ(β ) (1 + b1)Γ(β ) ⎤
ρ2 (t ) = ⎢t − t α−1 ⎥,
Λ⎣ (1 + a1)(1 + a0)Γ(α − 1) (1 + a1)Γ(α) ⎦
1⎡ b0T Γ(β ) ⎤⎡ a 0 T Γ( α ) ⎤
ρ3 (t ) = ⎢ω2 − ω 1 ⎥⎢ t α − 2 − t α − 1⎥ ,
Λ⎣ (1 + a1)Γ(α) ⎦⎣ (1 + a0)Γ(α − 1) ⎦
⎡ 1 ω b0T Γ(β ) ⎛ a0T ⎞⎤ ω ⎡ a0T ⎤
ρ1∗(t ) = t β − 2⎢ + 1 ⎜ν 2 − ν1⎟⎥ − t β − 1 1 ⎢ν2 − ν1⎥ ,
⎣ (1 + b0)Γ(β − 1) Λ (1 + b0)Γ(β − 1) ⎝ (1 + a1) ⎠⎦ Λ⎣ (1 + a1) ⎦
1 ⎡ β−2 b0T Γ(β ) ⎤
ρ2∗(t ) = ⎢t − t β − 1⎥ ,
Λ⎣ (1 + b0)Γ(β − 1) ⎦
1⎡ a0 T ⎤⎡ b0T Γ(β ) ⎤ (6)
ρ3∗(t ) = ⎢ν 2 − ν1⎥⎢t β − 2 − t β − 1⎥ ,
Λ⎣ (1 + a1) ⎦⎣ (1 + b0)Γ(β − 1) ⎦
⎡ a0 T ⎤
Λ = − Γ(β )(1 + b1) + ⎢ν2 − ν1⎥[Γ(α)(1 + a1)ω2 − b0T Γ(β )ω1],
⎣ (1 + a1) ⎦
⎡ n ⎤
1 ⎢μη α + β − 3 Γ(α − 1) + ∑ σjζ α − 2 ⎥ ,
ν1 =
(1 + a0)Γ(α − 1) ⎢⎣ 2
Γ(α + β − 2) j=1
j
⎥⎦
⎡ n ⎤
1 ⎢μη α + β − 2 Γ( α )
ν2 = + ∑ σjζ jα−1⎥⎥,
(1 + a1)Γ(α) ⎢⎣ 2 Γ(α + β − 1) j=1 ⎦
1 ⎡ Γ(β − 1) m ⎤
ω1 = ⎢νη1α + β − 3 + ∑μi ξiβ −2⎥⎥,
(1 + b0)Γ(β − 1) ⎢⎣ Γ(α + β − 2) i=1 ⎦
1 ⎡ Γ( β ) m ⎤
ω2 = ⎢νη α + β − 2 + ∑μi ξiβ −1⎥⎥.
(1 + a1)Γ(α) ⎢⎣ 1 Γ(α + β − 1) i=1 ⎦

Proof. It is well known that the solutions of fractional differential equations in (3) can be written as
u(t ) = I αψ1(t ) + c0t α − 2 + c1t α − 1, (7)

v (t ) = I βψ2 (t ) + c2t β − 2 + c3t β − 1, (8)

where ci ∈  , i = 0, 1, 2, 3 are unknown arbitrary constants. From (7) and (8), we have
D α − 2 u(t ) = c0Γ(α − 1) + c1Γ(α)t + I 2ψ1(t ) , (9)

D β − 2 v (t ) = c2 Γ(β − 1) + c3Γ(β )t + I 2ψ2 (t ) , (10)

D α − 1u(t ) = c1Γ(α) + I 1ψ1(t ) , (11)

D β − 1v (t ) = c3Γ(β ) + I 1ψ2 (t ) . (12)

Using (9) and (10) in (2), we get

(1 + a0)Γ(α − 1)c0 + a0T Γ(α)c1 = λ1 − a0I 2ψ1(T ) , (13)

(1 + b0)Γ(β − 1)c2 + b0T Γ(β )c3 = λ2 − b0I 2ψ2 (T ) . (14)

 Semi-coupled fractional differential systems  763

Combining (11) and (12) with (2), we obtain

⎡ νη α + β − 2 Γ(β ) m
b0T Γ(β ) ⎛ Γ(β − 1) m ⎞⎤
Γ(α)(1 + a1)c1 − ⎢ 1 + ∑ μi ξ i β − 1 − ⎜⎜νη1α + β − 3 + ∑μi ξiβ −2⎟⎟⎥⎥c3
⎢⎣ Γ(α + β − 1) i=1
(1 + b0)Γ(β − 1) ⎝ Γ(α + β − 2) i=1 ⎠⎦

b0 ⎡ Γ(β − 1) m ⎤ m
=− ⎢νη α + β − 3 + ∑μi ξiβ −2⎥⎥I 2ψ2(t ) + ∑μi I βψ2(ξi) − a1I 1ψ1(T ) (15)
(1 + b0)Γ(β − 1) ⎢⎣ 1 Γ(α + β − 2) i=1 ⎦ i=1

λ2 ⎡ Γ(β − 1) m ⎤
+ ⎢νη1α + β − 3 + ∑μi ξiβ −2⎥⎥ + νI α+β −1ψ2(η1),
⎢
(1 + b0)Γ(β − 1) ⎣ Γ(α + β − 2) i=1 ⎦

⎡ n ⎤
Γ( α )
− ⎢μη2α + β − 2 + ∑ σjζ jα−1⎥⎥c1 + Γ(β)(1 + b1)c3
⎢⎣ Γ(α + β − 1) j=1 ⎦
n ⎡ n ⎤ (16)
1 ⎢μη α + β − 3 Γ(α − 1) +
= ∑ σjI αψ1(ζ j ) − b1I 1ψ2 (T ) + ∑ σjζ jα−2⎥⎥
j=1
(1 + a0)Γ(α − 1) ⎢⎣ 2 Γ(α + β − 2) j=1 ⎦
× [λ1 − a0c1T Γ(α) − a0I 2ψ1(t )] + μI α + β − 1ψ1(η2 ) .

Solving (15) and (16) for c1 and c3 , we find that

⎡ 1 (b1 + 1)Γ(β ) ⎤⎡ m ⎤
c1 = ⎢ ⎥⎢ −ω1λ2 + b0ω1I 2ψ2 (T ) − νI α + β − 1ψ2 (η1) − ∑μi I βψ2(ξi) + a1I 1ψ1(T )⎥⎥
⎣ Λ (1 + a1)Γ(α) ⎦⎢⎣ i=1 ⎦
(17)
1⎡ b0T Γ(β ) ⎤⎡ n ⎤
+ ⎢ω 2 − ω1⎥⎢ −λ1ν1 − μI α + β − 1ψ1(η2 ) − ∑ j 1 j
σ I αψ (ζ ) + b I 1ψ (T ) + a ν I 2ψ (T )⎥ ,
1 0 1
Λ⎣ (1 + a1)Γ(α) ⎦⎢⎣ j=1
2 1
⎥⎦

1⎡ ⎛ a0T ⎞ ⎛ a0T ⎞
c3 = ⎢−λ1ν1 − λ2 ω1⎜ν2 − ν1⎟ + a0ν1I 2ψ1(t ) + b0ω1⎜ν2 − ν1⎟I 2ψ2 (T )
Λ⎣ ⎝ (1 + a1) ⎠ ⎝ (1 + a1) ⎠
⎛ a0T ⎞ ⎛ a0T ⎞ m
− ν ⎜ν 2 − ν1⎟I α + β − 1ψ2 (η1) − μI α + β − 1ψ1(η2 ) − ⎜ν2 − ν1⎟ ∑μi I βψ2 (ξi ) (18)
⎝ (1 + a1) ⎠ ⎝ (1 + a1) ⎠ i = 1
n ⎛ ⎞ ⎤
a0T
− ∑ σjI αψ1(ζ j ) + a1⎜⎝ν2 − ν1⎟I 1ψ1(T ) + b1I 1ψ2 (T )⎥ .
j=1
(1 + a1) ⎠ ⎥⎦

Substituting the values of c1 and c3 in (13) and (14), respectively, we get

⎡ 1 ν a 0 T Γ(α ) ⎛ b0T Γ(β ) ⎞⎤
c0 = [λ1 − a0I 2ψ1(T )]⎢ + 1 ⎜ω 2 − ω1⎟⎥
⎣ (1 + a0)Γ(α − 1) Λ (1 + a0)Γ(α − 1) ⎝ (1 + a1)Γ(α) ⎠⎦
⎡ω a0(1 + b1)T ∣Γ(β ) ⎤ ⎡ ω1 a0b0(1 + b1)T Γ(β ) ⎤
+ λ2 ⎢ 1 ⎥ − I 2ψ2 (T )⎢ ⎥
⎣ Λ (1 + a0)(1 + a1)Γ(α − 1) ⎦ ⎣ Λ (1 + a0)(1 + a1)Γ(α − 1) ⎦
⎡ν a 0 ( 1 + b 1 ) T Γ( β ) ⎤ m ⎡1 a 0 ( 1 + b 1 ) T Γ( β ) ⎤
+ I α + β − 1ψ2 (η1)⎢ ⎥ + ∑μi I βψ2 (ξi )⎢ ⎥
⎣ Λ (1 + a0)(1 + a1)Γ(α − 1) ⎦ i = 1 ⎣ Λ (1 + a0)(1 + a1)Γ(α − 1) ⎦
⎡1 a0a1(1 + b1)T Γ(β ) ⎤ ⎡μ a 0 T Γ( α ) ⎛ b0T Γ(β ) ⎞⎤
− I 1ψ1(T )⎢ ⎥ + I α + β − 1ψ1(η2 )⎢ ⎜ω 2 − ω1⎟⎥
⎣ Λ (1 + a0)(1 + a1)Γ(α − 1) ⎦ ⎣ Λ (1 + a0)Γ(α − 1) ⎝ (1 + a1)Γ(α) ⎠⎦
n ⎡1 a0T Γ(α) ⎛ b0T Γ(β ) ⎞⎤
+ ∑ σjI αψ1(ζ j )⎢ ⎜ω 2 − ω1⎟⎥
j=1 ⎣ Λ (1 + a0)Γ(α − 1) ⎝ (1 + a1)Γ(α) ⎠⎦
⎡ −b a0T Γ(α) ⎛ b0T Γ(β ) ⎞⎤
+ I 1ψ2 (T )⎢ 1 ⎜ω 2 − ω1⎟⎥ ,
⎣ Λ (1 + a0)Γ(α − 1) ⎝ (1 + a1)Γ(α) ⎠⎦
764  Ahmed Alsaedi et al.

⎡ 1 ω b0T Γ(β ) ⎛ a0T ⎞⎤

c2 = [λ2 − b0I 2ψ2 (T )]⎢ + 1 ⎜ν 2 − ν1⎟⎥
⎣ (1 + b0)Γ(β − 1) Λ (1 + b0)Γ(β − 1) ⎝ (1 + b 1 ) ⎠⎦
⎡ν b0T Γ(β ) ⎤ ⎡ ν1 a0b0T Γ(β ) ⎤
+ λ1⎢ 1 ⎥ − I 2ψ1(T )⎢ ⎥
⎣ Λ (1 + b0)Γ(β − 1) ⎦ ⎣ Λ (1 + b0)Γ(β − 1) ⎦
⎡ν b0T Γ(β ) ⎛ a0T ⎞⎤
+ I α + β − 1ψ2 (η1)⎢ ⎜ν 2 − ν1⎟⎥
⎣ Λ (1 + b0)Γ(β − 1) ⎝ (1 + a1) ⎠⎦
⎡μ b0T Γ(β ) ⎤ m ⎡1 b0T Γ(β ) ⎛ a0T ⎞⎤
+ I α + β − 1ψ1(η2 )⎢ ⎥+ ∑μi I βψ2(ξi)⎢ ⎜ν 2 − ν1⎟⎥
⎣ Λ (1 + b0)Γ(β − 1) ⎦ i=1 ⎣ Λ (1 + b0)Γ(β − 1) ⎝ (1 + a1) ⎠⎦
⎡1 b0a1T Γ(β ) ⎛ a0T ⎞⎤
− I 1ψ1(T )⎢ ⎜ν 2 − ν1⎟⎥
⎣ Λ (1 + b0)Γ(β − 1) ⎝ (1 + a1) ⎠⎦
n ⎡1 b0T Γ(β ) ⎤ ⎡1 b0b1T Γ(β ) ⎤
+ ∑ σjI αψ1(ζ j )⎢ ⎥ − I 1ψ2 (T )⎢ ⎥.
j=1 ⎣ Λ (1 + b0)Γ(β − 1) ⎦ ⎣ Λ (1 + b0)Γ(β − 1) ⎦

Inserting the value of c0, c1, c2 , and c3 in (7) and (8) together with the notations (6) leads to the solutions (4)
and (5). The converse of this lemma follows by direct computation. The proof is completed. □

3 Main results
Let C ([0, T ], ) denote the Banach space of all continuous real-valued functions defined on [0, T ] with
norm ‖u‖ = supt ∈ [0, T ]∣u(t )∣. For t ∈ [0, T ], let r([0, T ], ) denote the space of all functions ur such that
ur ∈ C ([0, T ], ), which is a Banach space endowed with norm ‖u‖r = supt ∈ [0, T ]{t r∣u(t )∣} . Let X = {u : u ∈
2 − α([0, T ], )} and Y = {v : v ∈ 2 − β([0, T ], )} be equipped with the norm ‖u‖X = supt ∈ [0, T ]{t 2 − α∣u(t )∣} and
‖v‖Y = supt ∈ [0, T ]{t 2 − β∣v (t )∣} , respectively. Then the product space (X × Y , ‖⋅‖X × Y ) is a Banach space with the
norm
‖(u , v )‖X × Y = ‖u‖X + ‖v‖Y .

Next we introduce an operator  : X × Y → X × Y by

 (u , v )(t ) = (1(u , v )(t ) , 2(u , v )(t )) ,

where

1(u , v )(t ) = λ1ρ1 (t ) + λ2 ω1ρ2 (t ) − a1ρ2 (t )I 1F (T , u(T ) , v (T )) − b1ρ3 (t )I 1G (T , u(T )v (T ))

− a0ρ1 (t )I 2F (T , u(T ) , v (T )) − b0ω1ρ2 (t )I 2G (T , u(T ) , v (T ))
+ μρ3 (t )I α + β − 1F (η2 , u(η2 ) , v (η2 )) + νρ2 (t )I α + β − 1G (η1 , u(η1) , v (η1))
n m
+ ρ3 (t ) ∑ σjI αF (ζ j, u(ζ j ) , v (ζ j )) + ρ2 (t )∑μi I βG (ξi, u(ξi ) , v (ξi ))
j=1 i=1
+ I αF (t , u(t ) , v (t )) , t ∈ [0, T ],

2(u , v )(t ) = λ2 ρ1∗(t ) + λ1ν1ρ2∗(t ) − a1ρ3∗(t )I 1F (T , u(T ) , v (T )) − b1ρ2∗(t )I 1G (T , u(T ) , v (T ))

− a0ν1ρ2∗(t )I 2F (T , u(T ) , v (T )) − b0ρ1∗(t )I 2G (T , u(T ) , v (T ))
+ μρ2∗(t )I α + β − 1F (η2 , u(η2 ) , v (η2 )) + νρ3∗(t )I α + β − 1G (η1 , u(η1) , v (η1))
n m
(19)
+ ρ2∗(t ) ∑ σjI αF (ζ j, u(ζ j ) , v (ζ j )) + ρ3∗(t ) ∑ μi I β G (ξ i, u(ξi ) , v (ξi ))
j=1 i=1

+ I βG (t , u(t ) , v (t )) , t ∈ [0, T ].
 Semi-coupled fractional differential systems  765

For the sake of brevity, we set

⎡ η 2α + β − 3Γ(α − 1) η α + 2β − 3Γ(β − 1) ⎤ ⎡ α β ⎤
N1 = ∣μ∣δ3⎢ 2 + 2 ⎥ + ∣a0∣δ1⎢ T Γ(α − 1) + T Γ(β − 1) ⎥
⎢⎣ Γ(2α + β − 2) Γ(α + 2β − 2) ⎥⎦ ⎣ Γ(α + 1) Γ(β + 1) ⎦
⎡ ζ 2α − 2 Γ(α − 1) ζ jα + β − 2 Γ(β − 1) ⎤ ⎡ α−1 T β−1 ⎤
n
+ δ3 ∑ ∣σj∣⎢
j
+ ⎥ + ∣a1∣δ2⎢ T + ⎥ (20)
j=1
⎢
⎣ Γ(2α − 1) Γ(α + β − 1) ⎦ ⎥ ⎣ (α − 1) (β − 1) ⎦

T αΓ(α − 1) T β Γ(β − 1)
+ + ,
Γ(2α − 1) Γ(α + β − 1)

⎡ η 2α + β − 3Γ(α − 1) η α + 2β − 3Γ(β − 1) ⎤ ⎡ α β ⎤
N2 = ∣ν∣δ2⎢ 1 + 1 ⎥ + ∣b0∣∣ω1∣δ2⎢ T Γ(α − 1) + T Γ(β − 1) ⎥
⎢⎣ Γ(2α + β − 2) Γ(α + 2β − 2) ⎥⎦ ⎣ Γ(α + 1) Γ(β + 1) ⎦
(21)
⎡ ξ α + β − 2 Γ(α − 1) ξ 2β − 2 Γ(β − 1) ⎤ ⎡ α−1 T β−1 ⎤
m
+ δ2 ∑∣μi ∣⎢ i + i ⎥ + ∣b1∣δ3⎢ T + ⎥,
i=1 ⎢⎣ Γ(α + β − 1) Γ(2β − 1) ⎥⎦ ⎣α − 1 β − 1⎦

⎡ η 2α + β − 3Γ(α − 1) η α + 2β − 3Γ(β − 1) ⎤ ⎡ α β ⎤
N1∗ = ∣μ∣δ2∗⎢ 2 + 2 ⎥ + ∣a0ν1∣δ2∗⎢ T Γ(α − 1) + T Γ(β − 1) ⎥
⎢⎣ Γ(2α + β − 2) Γ(α + 2β − 2) ⎥⎦ ⎣ Γ(α + 1) Γ(β + 1) ⎦
(22)
⎡ ζ 2α − 2 Γ(α − 1) ζ jα + β − 2 Γ(β − 1) ⎤ ⎡ α−1 T β−1 ⎤
n
+ δ2∗ ∑ j⎢∣σ ∣⎢ j
+ ⎥ + ∣a1∣δ ∗⎢ T + ⎥,
Γ(α + β − 1) ⎥⎦
3
j=1 ⎣ Γ(2α − 1) ⎣ (α − 1) (β − 1) ⎦

⎡ η 2α + β − 3Γ(α − 1) η α + 2β − 3Γ(β − 1) ⎤ ⎡ α β ⎤
N2∗ = ∣ν∣δ3∗⎢ 1 + 1 ⎥ + ∣b0∣δ1∗⎢ T Γ(α − 1) + T Γ(β − 1) ⎥
⎢⎣ Γ(2α + β − 2) Γ(α + 2β − 2) ⎥⎦ ⎣ Γ(α + 1) Γ(β + 1) ⎦
⎡ ξ α + β − 2 Γ(α − 1) ξ 2β − 2 Γ(β − 1) ⎤ ⎡ α−1 T β−1 ⎤
m
+ δ3∗ ∑∣μi ∣⎢ i + i ⎥ + ∣b1∣δ2∗⎢ T + ⎥ (23)
i=1 ⎢⎣ Γ(α + β − 1) Γ(2β − 1) ⎥⎦ ⎣α − 1 β − 1⎦
T αΓ(α − 1) T β Γ(β − 1)
+ + .
Γ(α + β − 1) Γ(2β − 1)

Theorem 3.1. Assume that:

(A1) F , G : [0, T ] ×  ×  →  are continuous functions and there exist positive constants L1 and L2 such
that, for all t ∈ [0, T ] and ui , vi ∈  , i = 1, 2 ,
∣F (t , u1 , v1) − F (t , u2 , v2 )∣ ≤ L1(∣u1 − u2 ∣ + ∣v1 − v2∣) ,
∣G (t , u1 , v1) − G (t , u2 , v2 )∣ ≤ L2 (∣u1 − u2 ∣ + ∣v1 − v2∣) .

Then the system (1)–(2) has a unique solution on [0, T ], provided that
L1(N1 + N1∗) + L2 (N2 + N2∗) < 1, (24)

where, N1, N2 , N1∗, and N2∗ are, respectively, given by (20), (21), (22), and (23).

Proof. Define supt ∈ [0, T ]F (t , 0, 0) = M1, supt ∈ [0, T ]G (t , 0, 0) = M2 , and choose r > 0 such that
∣λ1∣(δ1 + δ2∗∣ν1∣) + ∣λ2 ∣(∣ω1∣δ2 + δ1∗) + M1(e1 + e1∗) + M2 (e2 + e2∗)
r≥ , (25)
1 − [L1(N1 + N1∗) + L2 (N2 + N2∗)]

where

T2 η2α + β − 1 n ζ jα T2
e1 = ∣a0∣δ1 + ∣a1∣δ2T + ∣μ∣δ3 + δ3 ∑ ∣σj∣ + ,
2 Γ (α + β ) j=1
Γ(α + 1) Γ(α + 1)
(26)
η1α + β − 1 m
ξi βT2
e2 = ∣b1∣δ3T + ∣ν∣δ2 + δ2 ∑∣μi ∣ + ∣b0ω1∣δ2 ,
Γ (α + β ) i=1
Γ(β + 1) 2
766  Ahmed Alsaedi et al.

T 2 η2α + β − 1 n ζ jα
e1∗ = ∣a0ν1∣δ2∗ + ∣a1∣δ3∗T + ∣μ∣δ2∗ + δ2∗ ∑ ∣σj∣ ,
2 Γ (α + β ) j=1
Γ(α + 1)
η1α + β − 1 m
ξi β T2 T2
e2∗ = ∣b1∣δ2∗T + ∣ν∣δ3∗ + δ3∗ ∑∣μi ∣ + ∣b0∣δ1∗ + ,
Γ (α + β ) i=1
Γ(β + 1) 2 Γ(β + 1)

δm = sup {t 2 − α∣ρm (t )∣} , δm∗ = sup {t 2 − β∣ρm∗ (t )∣} , m = 1, 2, 3. (27)

t ∈ [0, T ] t ∈ [0, T ]

In the first step, it will be shown that   r ⊂  r , where  r = {(u , v ) ∈ X × Y : ‖(u , v )‖X × Y ≤ r } . By the assump-
tion (A1 ), for (u , v ) ∈  r , t ∈ [0, T ], we have
∣F (t , u(t ) , v (t ))∣ ≤ ∣F (t , u(t ) , v (t )) − F (t , 0, 0)∣ + ∣F (t , 0, 0)∣ ≤ L1(∣u∣ + ∣v∣) + M1

and
∣G (t , u(t ) , v (t ))∣ ≤ ∣G (t , u(t ) , v (t )) − G (t , 0, 0)∣ + ∣G (t , 0, 0)∣ ≤ L2 (∣u∣ + ∣v∣) + M2 .

In consequence, by using the relation for beta function B(⋅ , ⋅ ):

1
Γ(a )Γ(b )
B (a , b ) = ∫ua−1(1 − u)b−1du = Γ(a + b )
,
0

we obtain

‖1(u , v )‖X = sup {t 2 − α∣1(u , v )(t )∣}

t ∈ [0, T ]
⎧ ⎡ T
⎪ 2 − α⎢ (T − s )
≤ sup ⎨t ⎢∣λ1∣∣ρ1 (t )∣ + ∣λ2 ω1∣∣ρ2 (t )∣ + ∣a0∣∣ρ1 (t )∣
t ∈ [0, T ] ⎪ Γ(2)
(L1(∣u∣ + ∣v∣) + M1)ds ∫
⎩ ⎣ 0
T T

∫
+ ∣a1∣∣ρ2 (t )∣ (L1(∣u∣ + ∣v∣) + M1)ds + ∣b1∣∣ρ3 (t )∣ (L 2 (∣u∣ + ∣v∣) + M2 )ds ∫
0 0
η1 η2
(η1 − s)α + β − 2 (η − s ) α + β − 2
+ ∣ν∣∣ρ2 (t )∣ ∫ Γ(α + β − 1) (L2 (∣u∣ + ∣v∣) + M2 )ds + ∣μ∣∣ρ3 (t )∣ ∫ Γ(2α + β − 1) (L1(∣u∣ + ∣v∣) + M1)ds
0 0
ξi ζj
m
(ξ i − s ) β − 1 n (ζ j − s ) α − 1
+ ∣ρ2 (t )∣∑∣μi ∣ ∫ Γ( β )
(L2 (∣u∣ + ∣v∣) + M2 )ds + ∣ρ3 (t )∣ ∑ ∣σj∣ ∫ Γ(α )
(L1(∣u∣ + ∣v∣) + M1)ds
i=1 j=1
0 0
T t ⎤⎫
(t − s ) α − 1 ⎪
(L1(∣u∣ + ∣v∣) + M1)ds⎥⎥⎬
(T − s )
+ ∣b0ω1∣∣ρ2 (t )∣ ∫ (L2 (∣u∣ + ∣v∣) + M2 )ds + ∫
⎦⎪
Γ(2) Γ( α )
0 0 ⎭
T
2
≤ ∣λ1∣δ1 + ∣λ2 ω1∣δ2 + ∣a0∣δ1L1 ∫(T − s)[sα−2‖u‖X + s β−2‖v‖Y ]ds + M1∣a0∣δ1 T2
0
T T

+ ∣a1∣δ2L1 ∫ [sα − 2‖u‖X + s β − 2‖v‖Y ]ds + ∣a1∣δ2M1T + ∣b1∣δ3L2 ∫[sα−2‖u‖X + s β−2‖v‖Y ]ds
0 0
η1
(η − s ) α + β − 2
+ M2∣b1∣δ3T + ∣ν∣δ2L2 ∫ Γ(1α + β − 1) [sα−2‖u‖X + s β−2‖v‖Y ]ds
0
η2
(η1 )α + β − 1 (η − s ) α + β − 2
+ M2∣ν∣δ2
Γ( α + β )
+ ∣μ∣δ3L1 ∫ Γ(2α + β − 1) [sα−2‖u‖X + s β−2‖v‖Y ]ds
0
ξi
m
(η2 )α + β − 1 (ξ i − s ) β − 1
+ M1∣μ∣δ3
Γ(α + β )
+ δ2 ∑∣μi ∣L2 ∫ Γ(β )
[sα − 2‖u‖X + s β − 2‖v‖Y ]ds
i=1
0
 Semi-coupled fractional differential systems  767

ζj
m
∣μi ∣ξiβ n (ζ j − s ) α − 1
+ M2δ2 ∑
Γ(β + 1)
+ δ3 ∑ ∣σj∣L1 ∫ Γ(α )
[sα − 2‖u‖X + s β − 2‖v‖Y ]ds
i=1 j=1
0
T
n ∣σj∣ζ jα
+ M1δ3 ∑
Γ(α + 1)
+ ∣b0ω1∣δ2L2 ∫(T − s)[sα−2‖u‖X + s β−2‖v‖Y ]ds (28)
j=1
0
t
T2 α−1 T2
+ M1∣b0ω1∣δ2
2
+ L 1t 2 − α ∫ (t −Γ(sα)) [sα − 2‖u‖X + s β − 2‖v‖Y ]ds + M1
Γ(α + 1)
0
≤ ∣λ1∣δ1 + ∣λ2 ω1∣δ2 + [L1N1 + L2N2]r + M1e1 + M2e2 .

Similarly, one can get

∣∣2(u , v )∣∣Y ≤ ∣λ2 ∣δ1∗ + ∣λ1ν1∣δ2∗ + [L1N1∗ + L2N2∗]r + M1e1∗ + M2e2∗. (29)

In view of (28) and (29) together with (25), we have

‖ (u , v )‖X × Y ≤ ∣λ1∣[δ1 + ∣ν1∣δ2∗] + ∣λ2 ∣[δ2∣ω1∣ + δ1∗] + [L1(N1 + N1∗) + L2 (N2 + N2∗)]r + M1(e1 + e1∗) + M2 (e2 + e2∗) ≤ r .

Now, for (u1 , v1) , (u2 , v2 ) ∈ X × Y , and for any t ∈ [0, T ], we get

t 2 − α∣1(u2 , v2 )(t ) − 1(u1 , v1)(t )∣

⎧ ⎡ T
⎪ 2 − α⎢
∫
≤ sup ⎨t ⎢∣a0ρ1 (t )∣ (T − s)∣F (s, u2 , v2 ) − F (s, u1 , v1)∣ds
t ∈ [0, T ] ⎪
⎩ ⎣ 0
T T

∫ ∫
+ ∣a1ρ2 (t )∣ ∣F (s, u2 , v2 ) − F (s, u1 , v1)∣ds + ∣b1ρ3 (t )∣ ∣G (s, u2 , v2 ) − G (s, u1 , v1)∣ds
0 0
η1
(η − s )α + β − 2
+ ∣νρ2 (t )∣∫ Γ(1α + β − 1) ∣G(s, u2 , v2) − G(s, u1, v1)∣ds
0
η2
(η − s ) α + β − 2
+ ∣μρ3 (t )∣ ∫ Γ(2α + β − 1) ∣F (s, u2 , v2) − F (s, u1, v1)∣ds
0
ξi
m
(ξ i − s ) β − 1
+ ∣ρ2 (t )∣∑∣μi ∣ ∫ Γ (β )
∣G (s, u2 , v2 ) − G (s, u1 , v1)∣ds (30)
i=1
0
ζj
(ζ i − s ) α − 1
+ ∣ρ3 (t )∣ n ∣σj∣
j=1
∫ Γ (α )
∣F (s, u2 , v2 ) − F (s, u1 , v1)∣ds
0
T t ⎤⎫
(t − s ) α − 1 ⎪
∫
+ ∣b0ω1ρ2 (t )∣ (T − s)∣G (s, u2 , v2 ) − G (s, u1 , v1)∣ds + ∫ ∣F (s, u2 , v2 ) − F (s, u1 , v1)∣ds⎥⎥⎬ .
⎦⎪
Γ (α )
0 0 ⎭

Using (A1 ) and the relation ∣u2 − u1∣ + ∣v2 − v1∣ ≤ t α − 2‖u2 − u1‖X + t β − 2‖v2 − v1‖Y in (30) yields
‖1(u2 , v2 ) − 1(u1 , v1)‖X ≤ (L1N1 + L2N2 )[‖u2 − u1‖X + ‖v2 − v1‖Y ]. (31)

In a similar manner, one can get

‖2(u2 , v2 ) − 2(u1 , v1)‖Y ≤ (L1N1∗ + L2N2∗)[‖u2 − u1‖X + ‖v2 − v1‖Y ]. (32)

Thus, it follows from (31) and (32) that

‖ (u2 , v2 ) −  (u1 , v1)‖X × Y ≤ [L1(N1 + N1∗) + L2 (N2 + N2∗)][‖u2 − u1‖X + ‖v2 − v1‖Y ],
768  Ahmed Alsaedi et al.

which, in view of condition (24), implies that  is a contraction. Hence, by Banach’s fixed point theorem,
the operator  has a unique fixed point, which is indeed a unique solution of the problem (1)–(2) on [0, T ].
This completes the proof. □

In the following result, we present the sufficient conditions ensuring the existence of solutions for the
problem (1)–(2). We apply Leray-Schauder alternative [32] to prove this result.

Theorem 3.2. Assume that

(A1) F , G : [0, T ] ×  ×  →  are continuous functions and there exist real constants ki, γi ≥ 0, (i = 1, 2)
and k 0 > 0 and γ0 > 0 such that, for all t ∈ [0, T ] and u , v ∈  ,
∣F (t , u , v )∣ ≤ k 0 + k1∣u∣ + k2∣v∣ , ∣G (t , u , v )∣ ≤ γ0 + γ1∣u∣ + γ2∣v∣ .

(A2) k (N1 + N1∗) + γ (N2 + N2∗) < 1, where k = max{k1, k2 } , γ = max{γ1, γ2 } .

Then the system (1)–(2) has at least one solution on [0, T ].

Proof. Let us first note that continuity of the operator  follows from that of the functions F and G. Let
 ⊂ X × Y be bounded such that ∣F (t , u , v )∣ ≤ KF , ∣G (t , u , v )∣ ≤ KG , ∀(u , v ) ∈  , for positive constants KF
and KG . Then, for any (u , v ) ∈  , we have

t 2 − α∣1(u , v )(t )∣
⎡ η α+β−1 n ζ jα ⎤
T2 T2 ⎥
≤ ∣λ1∣δ1 + ∣λ2 ω1∣δ2 + KF ⎢∣a0∣δ1 + ∣a1∣Tδ2 + ∣μ∣δ3 2 + δ3 ∑ ∣σj∣ +
⎢⎣ 2 Γ (α + β ) j=1
Γ(α + 1) Γ(α + 1) ⎥⎦
⎡ η α+β−1 m
ξi β ⎤
T2
+ KG⎢∣ν∣δ2 1 + ∣b1∣δ3T + δ2 ∑∣μi ∣ + ∣b0ω1∣δ2 ⎥
⎢⎣ Γ (α + β ) i=1
Γ(β + 1) 2 ⎥⎦
= ∣λ1∣δ1 + ∣λ2 ω1∣δ2 + KF e1 + KGe2 ,

which implies that

‖1(u , v )‖X ≤ ∣λ1∣δ1 + ∣λ2 ω1∣δ2 + KF e1 + KGe2 .

Similarly, one can show that

‖2(u , v )‖Y ≤ ∣λ2 ∣δ1∗ + ∣λ1ν1∣δ2∗ + KF e1∗ + KGe2∗.

In consequence, we get
‖ (u , v )‖X × Y ≤ ∣λ1∣δ1 + ∣λ2 ∣δ1∗ + ∣λ2 ω1∣δ2 + ∣λ1ν1∣δ2∗ + KF (e1 + e1∗) + KG (e2 + e2∗) < ∞ ,

which shows that the operator  is uniformly bounded.

Next, we show that  is equicontinuous. Let t1, t2 ∈ [0, T ] with t1 > t2. Then we have

∣t12 − α 1(u , v )(t1) − t22 − α 1(u , v )(t2 )∣

≤ ∣λ1∣∣t12 − αρ1 (t1) − t22 − αρ1 (t2 )∣ + ∣λ2 ω1∣∣t12 − αρ2 (t1) − t22 − αρ2 (t2 )∣
⎡
⎢ T2
+ KF ⎢∣a0∣∣t12 − αρ1 (t1) − t22 − αρ1 (t2 )∣ + ∣a1∣T ∣∣t12 − αρ2 (t1) − t22 − αρ2 (t2 )∣
⎢⎣ 2

∣η2 ∣α + β − 1 n ζ jα
+ ∣μ∣∣t12 − αρ3 (t1) − t22 − αρ3 (t2 )∣ + ∣t12 − αρ3 (t1) − t22 − αρ3 (t2 )∣ ∑ ∣σj∣
Γ (α + β ) j=1
Γ(α + 1)
t1 t2 ⎤
(t1 − s)α − 1 (t2 − s)α − 1 ⎥
+ t12 − α ∫ Γ (α )
ds − t22 − α ∫ Γ (α )
ds ⎥
0 0 ⎥⎦
⎡
⎢ ∣η ∣α + β − 1 2 − α
+ KG⎢∣ν∣ 1 ∣t1 ρ2 (t1) − t22 − αρ2 (t2 )∣ +∣b1∣T ∣t12 − αρ3 (t1) − t22 − αρ3 (t2 )∣
⎢⎣ Γ ( α + β )
 Semi-coupled fractional differential systems  769

m
ξi β ⎤
T 2 2−α
+ ∑∣μi ∣ ∣t12 − αρ2 (t1) − t22 − αρ2 (t2 )∣ +∣b0ω1∣ ∣t1 ρ2 (t1) − t22 − αρ2 (t2 )∣⎥
i=1
Γ(β + 1) 2 ⎥⎦
⎧
⎪⎡ T 2 ⎤⎡ ν b0T Γ(β ) ⎤
≤ ∣t1 − t2∣⎨⎢∣λ1∣ +∣a0∣ KF ⎥⎢ 1 ω2 − ω1 ⎥
⎪⎣ 2 ⎦⎣ Λ
⎩ Γ(α)(1 + a1) ⎦
⎡ η α+β−1 m
ξi β ⎤
T2 Γ(β )(b1 + 1)
+ ⎢∣λ2 ω1∣ +∣a1∣TKF + ∣ν∣KG 1 + KG ∑∣μi ∣ + KG∣b0ω1∣ ⎥
⎢⎣ Γ (α + β ) i=1
Γ(β + 1) 2 ⎥⎦ ΛΓ(α)(a1 + 1)
⎡n ζ jα ∣η ∣α + β − 1 ⎤⎡
1 b0T Γ(β ) ⎤⎫
⎪
+ ⎢ ∑ ∣σj∣ KG + ∣μ∣ 2 KF + ∣b1∣TKG ⎥⎢ ω2 − ω 1 ⎥⎬
⎢⎣ j = 1 Γ(α + 1) Γ (α + β ) ⎥⎦⎣ ∣Λ∣ Γ(α)(1 + a1) ⎦⎭
⎪

2KF t12 − α ∣t 2 − t22∣

+ ∣t1 − t2 ∣α + KF 1 →0 as t2 → t1,
Γ(α + 1) Γ(α + 1)

independent of (u , v ) ∈  . Also, we have

∣t12 − β 2(u , v )(t1) − t22 − β 2(u , v )(t2 )∣

⎧
⎪⎡ T 2 ⎤⎡ ω 1 a0T ⎤
≤ ∣t1 − t2∣⎨⎢∣λ2 ∣ + ∣b0∣KG ⎥⎢ ν2 − ν1 ⎥
⎪⎣ ⎦
2 ⎣ Λ
⎩ (1 + a1) ⎦
⎡ η α+β−1 n ζ jα ⎤⎡ 1 ⎤
T2
+ ⎢∣λ1ν1∣ + ∣a0ν1∣KF + μKF 2 + ∣b1∣TKG + KF ∑ ∣σj∣ ⎥⎢ ⎥
⎢⎣ 2 Γ(α + β ) j=1
Γ(α) ⎥⎦⎣ ∣Λ∣ ⎦
⎡ η α+β−1 m
ξ i β ⎤⎡ 1 a0T ⎤⎫
⎪
+ ⎢∣a1∣TKF + ∣ν∣KG 1 + KG ∑∣μi ∣ ⎥⎢ ν2 − ν1 ⎥⎬
⎢⎣ Γ(α + β ) i=1
Γ ( β + 1) ⎥⎦⎣ ∣Λ∣ (1 + a1) ⎦⎭
⎪

2KGt12 − β ∣t 2 − t22∣
+ ∣t1 − t2 ∣β + KG 1 →0 as t2 → t1,
Γ(β + 1) Γ(β + 1)

independently of (u , v ) ∈  . Thus, the operator  is equicontinuous. Thus, we deduce that the operator  is
completely continuous.
Finally, we consider the set
V = {(u , v ) ∈ X × Y ∣(u , v ) = m (u , v ) , 0 ≤ m ≤ 1}

and show that it is bounded. Let (u , v ) ∈ V with (u , v ) = m (u , v ), u = m1(u , v ), and v = m2(u , v ). Then
we have
‖u‖X ≤ ∣λ1∣δ1 + ∣λ2 ω1∣δ2 + (N1k1 + N2γ1)‖u‖X + (N1k2 + N2γ2 )‖v‖Y + k 0e1 + γ0e2
≤ ∣λ1∣δ1 + ∣λ2 ω1∣δ2 + (N1k + N2γ )‖u‖X + (N1k + N2γ )‖v‖Y + k0e1 + γ0e2 ,
‖v‖Y ≤ ∣λ2 ∣δ1∗ + ∣λ1ν1∣δ2∗ + (N1∗k1 + N2∗γ1)‖u‖X + (k2N1∗ + N2∗γ2 )‖v‖Y + k 0e1∗ + γ0e2∗
≤ ∣λ2 ∣δ1∗ + ∣λ1ν1∣δ2∗ + (N1∗k + N2∗γ )‖u‖X + (kN1∗ + N2∗γ )‖v‖Y + k0e1∗ + γ0e2∗,

which imply that

‖u‖X + ‖v‖Y ≤ ∣λ1∣δ1 + ∣λ2 ω1∣δ2 + ∣λ2 ∣δ1∗ + ∣λ1ν1∣δ2∗ + [k (N1 + N1∗) + γ (N2 + N2∗)]‖u‖X
+ [k (N1 + N1∗) + γ (N2 + N2∗)]‖v‖Y + k 0(e1 + e1*) + γ0(e2 + e2∗) .

Thus,
∣λ1∣δ1 + ∣λ2 ω1∣δ2 + ∣λ2 ∣δ1∗ + ∣λ1ν1∣δ2∗ + k 0(e1 + e1*) + γ0(e2 + e2∗)
‖(u , v )‖X × Y ≤ .
1 − [k (N1 + N1∗) + γ (N2 + N2∗)]

Hence, the set V is bounded. Thus, by Leray-Schauder alternative, we deduce that the operator  has at
least one fixed point, which corresponds to the fact that the problem (1)–(2) has at least one solution on
[0, T ]. The proof is completed. □
770  Ahmed Alsaedi et al.

4 Examples
This section is devoted to the illustration of the results derived in the previous section.

Example 4.1. Consider the system of fractional differential equations consisting of the equations given by
⎧ 6/5 1
⎪ D u(t ) = (u(t ) + tan−1 v (t )) + cos t , t ∈ [0, 1],
⎪ 4 1600 + t
⎨ (33)
⎪ D7 / 4v (t ) = 1
⎪ (sin u(t ) + v (t )) + t sin t , t ∈ [0, 1],
⎩ 2500 + t

supplemented by the following boundary conditions

⎧ −4 / 5 + 1
⎪D u(0 ) + D−4 / 5u(1−) = −4,
⎪ 4
⎪ 1/5 + 3 1/5 − ⎛1⎞ ⎛1⎞ ⎛2⎞
⎪ D u(0 ) − D u(1 ) = −2I 1 / 5v ⎜ ⎟ − 6v ⎜ ⎟ − 4v ⎜ ⎟ ,
⎨ 2 ⎝4⎠ ⎝2⎠ ⎝3⎠ (34)
⎪ D−1 / 4v (0+) − D−1 / 4v (1−) = 2,
⎪
⎪ 3/4 + 3 3/4 − 3/4
⎛ 1 ⎞ 5 ⎛ 11 ⎞ 2 ⎛ 3 ⎞
⎪ D v (0 ) − 2 D v (1 ) = 3I u⎝⎜ 3 ⎠⎟ + 3 u⎝⎜ 15 ⎠⎟ + 5 u⎝⎜ 4 ⎠⎟ .
⎩

Here α = 6 / 5, β = 7 / 4, a0 = 1 / 4 , b0 = −1 / 4, a1 = −3 / 2 , b1 = −3 / 2 , ν = −2 , μ = 3, λ1 = −4, λ2 = 2 , μ1 = −6,

μ2 = −4, σ1 = 5 / 3, σ2 = 2 / 5, η1 = 1 / 4, ξ1 = 1 / 2 , ξ2 = 2 / 3, η2 = 1 / 3, ζ1 = 11 / 15, ζ2 = 3 / 4.

Using the given data, it is found that Λ ≃ 70.9531, ν1 ≃ 2.9182 , ν2 ≃ −6.3910, ω1 ≃ −15.3511, ω2 ≃ 15.2944,
δ1 ≃ 0.7330, δ2 ≃ 0.0135, δ3 ≃ 0.3100, δ1∗ ≃ 0.8213, δ2∗ ≃ 0.0176, δ3∗ ≃ 0.0869, N1 ≃ 12.2414, N2 ≃ 3.6569,
N1∗ ≃ 1.1599, N2∗ ≃ 9.4324 , L1 = 1 / 160, L2 = 1 / 50, and
L1(N1 + N1∗) + L2 (N2 + N2∗) ≃ 0.3455 < 1.

Thus, all the conditions of Theorem 3.1 are satisfied and hence the problem (33)–(34) has a unique solution
on [0, 1].

Example 4.2. Let us consider the problem (33)–(34) with

sin2 t tan−1 u(t ) v (t )∣u(t )∣

F (t , u(t ) , v (t )) = + + , t ∈ [0, 1],
2+ 2 400 + t
t3 50(1 + ∣u(t )∣)
(35)
∣u(t )∣cos u(t ) v (t )
G (t , u(t ) , v (t )) = e−t + + , t ∈ [0, 1].
150 + t 640 1 + sin2 t
1 1 1 1 1
Clearly, ∣F (t , u(t ) , v (t ))∣ < 2
+ 40
∣u(t )∣ + 50
∣v (t )∣, G (t , u(t ) , v (t )) < 1 + 150
∣u(t )∣ + 640
∣v (t )∣, and so k 0 = 1/2 ,
k1 = 1 / 40, k2 = 1 / 50, γ0 = 1, γ1 = 1 / 150, γ2 = 1 / 640, k = max{k1, k2 } = 1 / 40, γ = max{γ1, γ2 } = 1 / 150. Moreover,
k (N1 + N1∗) + γ (N2 + N2∗) ≃ 0.4223 < 1.

Therefore, by Theorem 3.2, the problem (33)–(34) with F and G given by (35) has at least one solution
on [0, 1].

5 Conclusion
We have investigated the existence and uniqueness of solutions for a nonlinear system of Riemann-
Liouville fractional differential equations, equipped with nonseparated semi-coupled integro-multipoint
boundary conditions. We apply Banach contraction mapping principle to establish the existence of a
 Semi-coupled fractional differential systems  771

unique solution, while Leray-Schauder alternative is used to obtain the existence result for the problem at
hand. We emphasize that the novelty of our results lies on the semi-coupled boundary conditions (2) and
enrich the related literature on the topic. Our work also produces some special cases by fixing the para-
meters involved in the boundary conditions. For example, our results correspond to the ones for nonlocal
semi-coupled fractional multipoint boundary conditions by fixing ν = 0 = μ and the results for nonlocal
semi-coupled fractional integral boundary conditions follow by taking all μi = 0, i = 1, … , m and σj = 0,
j = 1, … , n.

Acknowledgment: This project was funded by the Deanship of Scientific Research (DSR), King Abdulaziz
University, Jeddah, Saudi Arabia under grant no. (KEP-PhD-40-130-41). The authors, therefore, acknowl-
edge with thanks DSR technical and financial support. The authors also thank the reviewers for their
constructive remarks on our work.

Conflict of interest: Authors state no conflict of interest.

References
[1] Z. Bai, On positive solutions of a nonlocal fractional boundary value problem, Nonlinear Anal. 72 (2010), 916–924.
[2] B. Ahmad and J. J. Nieto, Riemann-Liouville fractional integro-differential equations with fractional nonlocal integral
boundary conditions, Bound. Value Probl. 2011 (2011), 36.
[3] B. Ahmad, A. Alsaedi, A. Assolami, and R. P. Agarwal, A study of Riemann-Liouville fractional nonlocal integral boundary
value problems, Bound. Value Probl. 2013 (2013), 274.
[4] R. Luca, On a class of nonlinear singular Riemann-Liouville fractional differential equations, Results Math. 73 (2018),
no. 125, 1–15.
[5] R. P. Agarwal and R. Luca, Positive solutions for a semipositone singular Riemann-Liouville fractional differential problem,
Int. J. Nonlinear Sci. Numer. Simul. 20 (2019), 823–831.
[6] R. P. Agarwal, S. Hristova, and D. O’Regan, Exact solutions of linear Riemann-Liouville fractional differential equations with
impulses, Rocky Mountain J. Math. 50 (2020), 779–791.
[7] R. Agarwal, S. Hristova, and D. O’Regan, Existence and Ulam-type stability for nonlinear Riemann-Liouville fractional
differential equations with constant delay, Electron. J. Qual. Theory Differ. Equ. 2020 (2020), no. 67, 1–18.
[8] F. Mainardi, Some basic problems in continuum and statistical mechanics, in: A. Carpinteri, F. Mainardi (eds), Fractals and
Fractional Calculus in Continuum Mechanics, Springer, Berlin, 1997, pp. 291–348.
[9] R. L. Magin, Fractional Calculus in Bioengineering, Begell House Publishers, Connecticut, 2006.
[10] H. A. Fallahgoul, S. M. Focardi, and F. J. Fabozzi, Fractional Calculus and Fractional Processes with Applications to Financial
Economics, Theory and Application, Elsevier/Academic Press, London, 2017.
[11] A. K. Golmankhaneh and D. Baleanu, Calculus on Fractals, in: C. Cattani, H. M. Srivastava, X.-J. Yang (eds), Fractional
Dynamics, Warsaw, Poland, De Gruyter Open, Poland, 2016, pp. 307–332.
[12] I. M. Sokolov, J. Klafter, and A. Blumen, Fractional kinetics, Phys. Today 55 (2002), 48–54.
[13] N. Nyamoradi, M. Javidi, and B. Ahmad, Dynamics of SVEIS epidemic model with distinct incidence, Int. J. Biomath. 8
(2015), no. 6, 1550076.
[14] I. Petras and R. L. Magin, Simulation of drug uptake in a two compartmental fractional model for a biological system,
Commun. Nonlinear. Sci. Numer. Simul. 16 (2011), 4588–4595.
[15] W. Yukunthorn, B. Ahmad, S. K. Ntouyas, and J. Tariboon, On Caputo-Hadamard type fractional impulsive hybrid systems
with nonlinear fractional integral conditions, Nonlinear Anal. Hybrid Syst. 19 (2016), 77–92.
[16] N. Heymans and J. C. Bauwens, Fractal rheological models and fractional differential equations for viscoelastic behavior,
Rheologica Acta 33 (1994), 210–219.
[17] M. Kirane, B. Ahmad, A. Alsaedi, and M. Al-Yami, Non-existence of global solutions to a system of fractional diffusion
equations, Acta Appl. Math. 133 (2014), 235–248.
[18] S. Pati, J. R. Graef, and S. Padhi, Positive periodic solutions to a system of nonlinear differential equations with appli-
cations to Lotka-Volterra-type ecological models with discrete and distributed delays, J. Fixed Point Theory Appl. 21
(2019), 80.
[19] J. Henderson and R. Luca, Systems of Riemann-Liouville fractional equations with multi-point boundary conditions, Appl.
Math. Comput. 309 (2017), 303–323.
[20] H. T. Tuan, A. Czornik, J. J. Nieto, and M. Niezabitowski, Global attractivity for some classes of Riemann-Liouville fractional
differential systems, J. Integral Equ. Appl. 31 (2019), no. 2, 265–282.
772  Ahmed Alsaedi et al.

[21] A. Alsaedi, S. Aljoudi, and B. Ahmad, Existence of solutions for Riemann-Liouville type coupled systems of fractional
integro-differential equations and boundary conditions, Electron. J. Differ. Equ. 2016 (2016), 211.
[22] R. P. Agarwal, B. Ahmad, D. Garout, and A. Alsaedi, Existence results for coupled nonlinear fractional differential equations
equipped with nonlocal coupled flux and multi-point boundary conditions, Chaos Soliton. Fract. 102 (2017), 149–161.
[23] R. Luca, Positive solutions for a system of fractional differential equations with p-Laplacian operator and multi-point
boundary conditions, Nonlinear Anal. Model. Control 23 (2018), 771–801.
[24] B. Ahmad, A. Alsaedi, S. K. Ntouyas, and Y. Alruwaily, On a fractional integro-differential system involving Riemann-
Liouville and Caputo derivatives with coupled multi-point boundary conditions, Int. J. Differ. Equ. 15 (2020), no. 2,
209–241.
[25] A. Alsaedi, R. Luca, and B. Ahmad, Existence of positive solutions for a system of singular fractional boundary value
problems with p-Laplacian operators, Mathematics 8 (2020), 1890.
[26] S. Padhi, B. S. R. V. Prasad, and D. Mahendru, Systems of Riemann-Liouville fractional differential equations with nonlocal
boundary conditions – Existence, nonexistence, and multiplicity of solutions: Method of fixed point index, Math. Meth.
Appl. Sci. 44 (2021), no. 10, 8266–8285.
[27] X. M. Zhang, A new method for searching the integral solution of system of Riemann-Liouville fractional differential
equations with non-instantaneous impulses, J. Comput. Appl. Math. 388 (2021), 113307.
[28] B. Ahmad, N. Alghamdi, A. Alsaedi, and S. K. Ntouyas, Multi-term fractional differential equations with nonlocal boundary
conditions, Open Math. 16 (2018), 1519–1536.
[29] B. Ahmad, M. Alghanmi, A. Alsaedi, and J. J. Nieto, Existence and uniqueness results for a nonlinear coupled system
involving Caputo fractional derivatives with a new kind of coupled boundary conditions, Appl. Math. Lett. 116 (2021),
107018.
[30] A. A. Kilbas, H. M. Srivastava, and J. J. Trujillo, Theory and Applications of Fractional Differential Equations, North-Holland
Mathematics Studies, vol. 204, Elsevier, Amsterdam, 2006.
[31] Y. Zhou, Basic Theory of Fractional Differential Equations, World Scientific, Singapore, 2014.
[32] A. Granas and J. Dugundji, Fixed Point Theory, Springer-Verlag, New York, 2003.