Skip to main content

New generalized Hermite-Hadamard type inequalities and applications to special means

Abstract

In this paper, Hermite-Hadamard type inequalities involving Hadamard fractional integrals for the functions satisfying monotonicity, convexity and s-e-condition are studied. Three classes of left-type Hadamard fractional integral identities including the first-order derivative are firstly established. Some interesting Hermite-Hadamard type integral inequalities involving Hadamard fractional integrals are also presented by using the established integral identities. Finally, some applications to special means of real numbers are given.

MSC:26A33, 26A51, 26D15.

1 Preliminaries

Let f:[a,b]⊂R→R be a convex function, then it is integrable in the sense of Riemann and

f ( a + b 2 ) ≤ 1 b − a ∫ a b f(t)dt≤ f ( a ) + f ( b ) 2 .

The above inequality is the so-called classical Hermite-Hadamard type inequality which provides lower and upper estimations for the integral average of any convex function defined on a compact interval, involving the midpoint and the endpoints of the domain. This interesting inequality was firstly discovered by Hermite in 1881 in the journal Mathesis (see Mitrinović and Lacković [1]). However, this beautiful result was nowhere mentioned in the mathematical literature and was not widely known as Hermite’s result (see Pec̆arić et al. [2]). For more recent results which generalize, improve and extend this classical Hermite-Hadamard inequality, one can see [3–13] and references therein.

Meanwhile, fractional integrals and derivatives provide an excellent tool for the description of memory and hereditary properties of various materials and processes. They are greatly applied in nonlinear oscillations of earthquakes, in many physical phenomena such as seepage flow in porous media and in a fluid-dynamic traffic model. For more recent development on fractional calculus, one can see the monographs [14–21].

Because of the wide application of Hermite-Hadamard type inequalities and fractional integrals, many researchers extend their studies to Hermite-Hadamard type inequalities involving fractional integrals not limited to integer integrals. Recently, more and more Hermite-Hadamard inequalities involving fractional integrals have been obtained for different classes of functions; see [22, 23] for nondecreasing functions and convex functions, [24, 25] for m-convex functions and (s,m)-convex functions and the references therein.

It is remarkable that Wang et al. [23] adopted some idea in [22] to derive the following Hermite-Hadamard type inequalities involving Hadamard fractional integrals.

Theorem 1.1 (see Theorem 2.1, [23])

Letf:[a,b]→Rbe a positive function with0<a<bandf∈L[a,b]. Iffis a nondecreasing and convex function on[a,b], then the following inequality for fractional integrals holds:

f( a b )≤ Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f(b) + H J b − α f(a)]≤f(b).

We remark that the symbols denoting the left-sided and right-sided Hadamard fractional integrals of order α∈ R + of function f(x) are defined by

( H J a + α f)(x)= 1 Γ ( α ) ∫ a x ( ln x t ) α − 1 f(t) d t t (0<a<x≤b),

and

( H J b − α f)(x)= 1 Γ ( α ) ∫ x b ( ln t x ) α − 1 f(t) d t t (0<a≤x<b),

where Γ(⋅) is the gamma function.

Moreover, the following important right-type Hadamard fractional integral identities including the first-order derivative of f are also established.

Lemma 1.2 (see Lemma 3.1, [23])

Letf:[a,b]→Rbe a differentiable mapping on(a,b)with0<a<b. If f ′ ∈L[a,b], then the following equality for fractional integrals holds:

f ( a ) + f ( b ) 2 − Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] = ln b − ln a 2 ∫ 0 1 [ ( 1 − t ) α − t α ] e t ln a + ( 1 − t ) ln b f ′ ( e t ln a + ( 1 − t ) ln b ) d t .

In the sequence, some new Hermite-Hadamard type integral inequalities involving Hadamard fractional integrals are presented in [23].

However, other interesting left-type Hadamard fractional integral identities including the first-order derivative of f have not been reported. Thus, the first purpose of this paper is to find some possible presentation including the first-order derivative of f for the following equalities:

Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a + b 2 ) = What? Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a b ) = What? Γ ( α + 1 ) [ H J x − α f ( a ) + H J x + α f ( b ) ] − [ f ( a ) ( ln x − ln a ) α + f ( b ) ( ln b − ln x ) α ] = What?

The second purpose of this paper is to establish some interesting Hermite-Hadamard type inequalities involving Hadamard fractional integrals for the functions satisfying monotonicity, convexity and s-e-condition by utilizing the left-type Hadamard fractional integral identities. In [26], Wang et al. introduced the concept of s-e-condition to overcome some essential difficulties from the singular kernels in Hadamard fractional integrals.

Definition 1.3 A function f:I⊂(0,∞)→R is said to satisfy s-e-condition if

f ( e λ x + ( 1 − λ ) y ) ≤ λ s f ( e x ) + ( 1 − λ ) s f ( e y )

for all x,y∈I, λ∈[0,1] and for some fixed s∈(0,1].

Remark 1.4 If f:I⊂(0,∞)→R is a nondecreasing and convex function, then f satisfies

f ( e λ x + ( 1 − λ ) y ) ≤f ( λ e x + ( 1 − λ ) e y ) ≤λf ( e x ) +(1−λ)f ( e y )

for all x,y∈I, λ∈[0,1], which implies the s-e-condition above.

In the following, we recall the following two basic inequalities [27].

Lemma 1.5For0<σ≤1and0≤a<b, we have

| a σ − b σ | ≤ ( b − a ) σ .

Lemma 1.6For allλ,υ,ω>0, then for anyt>0, we have

∫ 0 t ( t − s ) υ − 1 s λ − 1 e − ω s ds≤max { 1 , 2 1 − υ } Γ(λ) ( 1 + λ υ ) ω − λ t υ − 1 .

2 Left-type Hadamard fractional integral identities

In this section, we establish some important left-type Hadamard fractional integral identities including the first-order derivative of a given function.

Lemma 2.1Letf:[a,b]→Rbe a differentiable mapping on(a,b)with0<a<b. If f ′ ∈L[a,b], then the following equality for fractional integrals holds:

Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a + b 2 ) = b − a 2 ∫ 0 1 k f ′ ( t a + ( 1 − t ) b ) d t − ln b − ln a 2 ∫ 0 1 [ ( 1 − t ) α − t α ] e t ln a + ( 1 − t ) ln b f ′ ( e t ln a + ( 1 − t ) ln b ) d t ,
(1)

where

k={ 1 , 0 ≤ t < 1 2 , − 1 , 1 2 ≤ t < 1 .

Proof Denote

I = b − a 2 ∫ 0 1 2 f ′ ( t a + ( 1 − t ) b ) d t − b − a 2 ∫ 1 2 1 f ′ ( t a + ( 1 − t ) b ) d t − ln b − ln a 2 ∫ 0 1 [ ( 1 − t ) α − t α ] e t ln a + ( 1 − t ) ln b f ′ ( e t ln a + ( 1 − t ) ln b ) d t : = I 1 + I 2 + I 3 + I 4 ,
(2)

where

I 1 = b − a 2 ∫ 0 1 2 f ′ ( t a + ( 1 − t ) b ) d t , I 2 = − b − a 2 ∫ 1 2 1 f ′ ( t a + ( 1 − t ) b ) d t , I 3 = − ln b − ln a 2 ∫ 0 1 ( 1 − t ) α e ln b − t ( ln b − ln a ) f ′ ( e ln b − t ( ln b − ln a ) ) d t , I 4 = ln b − ln a 2 ∫ 0 1 t α e ln b − t ( ln b − ln a ) f ′ ( e ln b − t ( ln b − ln a ) ) d t .

Integrating by parts, we have

I 1 = b − a 2 ∫ 0 1 2 f ′ ( t a + ( 1 − t ) b ) d t = − f ( t a + ( 1 − t ) b ) 2 | 0 1 2 = 1 2 [ f ( b ) − f ( a + b 2 ) ] ,
(3)

and

I 2 = − b − a 2 ∫ 1 2 1 f ′ ( t a + ( 1 − t ) b ) d t = f ( t a + ( 1 − t ) b ) 2 | 1 2 1 = 1 2 [ f ( a ) − f ( a + b 2 ) ] ,
(4)

and

I 3 = − ln b − ln a 2 ∫ 0 1 ( 1 − t ) α e ln b − t ( ln b − ln a ) f ′ ( e ln b − t ( ln b − ln a ) ) d t = ( 1 − t ) α f ( e ln b − t ( ln b − ln a ) ) 2 | 0 1 + α 2 ∫ 0 1 ( 1 − t ) α − 1 f ( e ln b − t ( ln b − ln a ) ) d t = − f ( b ) 2 + α 2 ( ln a − ln b ) ∫ b a ( ln u − ln a ln b − ln a ) α − 1 f ( u ) d u u = − f ( b ) 2 + α 2 ( ln b − ln a ) α ∫ a b ( ln u − ln a ) α − 1 f ( u ) d u u = − f ( b ) 2 + Γ ( α + 1 ) 2 ( ln b − ln a ) α H J b − α f ( a ) ,
(5)

and

I 4 = ln b − ln a 2 ∫ 0 1 t α e ln b − t ( ln b − ln a ) f ′ ( e ln b − t ( ln b − ln a ) ) d t = − t α f ( e ln b − t ( ln b − ln a ) ) 2 | 0 1 + α 2 ∫ 0 1 t α − 1 f ( e ln b − t ( ln b − ln a ) ) d t = − f ( a ) 2 + α 2 ( ln a − ln b ) ∫ b a ( ln b − ln u ln b − ln a ) α − 1 f ( u ) d u u = − f ( a ) 2 + α 2 ( ln b − ln a ) α ∫ a b ( ln b − ln u ) α − 1 f ( u ) d u u = − f ( a ) 2 + Γ ( α + 1 ) 2 ( ln b − ln a ) α H J a + α f ( b ) .
(6)

Submitting (3), (4), (5) and (6) into (2), it follows that

I= Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f(b) + H J b − α f(a)]−f ( a + b 2 ) .
(7)

This completes the proof. □

Lemma 2.2Letf:[a,b]→Rbe a differentiable mapping on(a,b)with0<a<b. If f ′ ∈L[a,b], then the following equality for fractional integrals holds:

Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a b ) = ln b − ln a 2 [ ∫ 0 1 k e t ln a + ( 1 − t ) ln b f ′ ( e t ln a + ( 1 − t ) ln b ) d t − ∫ 0 1 [ ( 1 − t ) α − t α ] e t ln a + ( 1 − t ) ln b f ′ ( e t ln a + ( 1 − t ) ln b ) d t ] ,
(8)

where

k={ 1 , 0 ≤ t < 1 2 , − 1 , 1 2 ≤ t < 1 .

Proof It is not difficult to verify that

I = ∫ 0 1 2 e ln b − t ( ln b − ln a ) f ′ ( e ln b − t ( ln b − ln a ) ) d t − ∫ 1 2 1 e ln b − t ( ln b − ln a ) f ′ ( e ln b − t ( ln b − ln a ) ) d t − ∫ 0 1 [ ( 1 − t ) α − t α ] e t ln a + ( 1 − t ) ln b f ′ ( e t ln a + ( 1 − t ) ln b ) d t = f ( b ) − f ( a b ) ln b − ln a + f ( a ) − f ( a b ) ln b − ln a − f ( b ) ln b − ln a + Γ ( α + 1 ) ( ln b − ln a ) α + 1 H J b − α f ( a ) − f ( a ) ln b − ln a + Γ ( α + 1 ) ( ln b − ln a ) α + 1 H J a + α f ( b ) = Γ ( α + 1 ) ( ln b − ln a ) α + 1 [ H J a + α f ( b ) + H J b − α f ( a ) ] − 2 f ( a b ) ln b − ln a .
(9)

Thus, by multiplying both sides by ln b − ln a 2 in (9), we have conclusion (8) immediately. □

Lemma 2.3Letf:[a,b]→Rbe a differentiable mapping on(a,b)with0<a<b. If f ′ ∈L[a,b], then the following equality for fractional integrals holds:

Γ ( α + 1 ) [ H J x − α f ( a ) + H J x + α f ( b ) ] − [ f ( a ) ( ln x − ln a ) α + f ( b ) ( ln b − ln x ) α ] = ( ln b − ln x ) α + 1 ∫ 0 1 ( t α − 1 ) e t ln x + ( 1 − t ) ln b f ′ ( e t ln x + ( 1 − t ) ln b ) d t − ( ln x − ln a ) α + 1 ∫ 0 1 ( t α − 1 ) e t ln x + ( 1 − t ) ln a f ′ ( e t ln x + ( 1 − t ) ln a ) d t .
(10)

Proof Integrating by parts, we have

∫ 0 1 ( t α − 1 ) e t ln x + ( 1 − t ) ln b f ′ ( e t ln x + ( 1 − t ) ln b ) d t = ( t α − 1 ) f ( e t ln x + ( 1 − t ) ln b ) ln x − ln b | 0 1 − α ln x − ln b ∫ 0 1 t α − 1 f ( e t ln x + ( 1 − t ) ln b ) d t = f ( b ) ln x − ln b − α ( ln b − ln x ) α + 1 ∫ b x ( ln b − ln u ) α − 1 f ( u ) d u u = − f ( b ) ln b − ln x + Γ ( α + 1 ) ( ln b − ln x ) α + 1 H J x + α f ( b ) ,
(11)

and

∫ 0 1 ( t α − 1 ) e t ln x + ( 1 − t ) ln a f ′ ( e t ln x + ( 1 − t ) ln a ) d t = ( t α − 1 ) f ( e t ln x + ( 1 − t ) ln a ) ln x − ln a | 0 1 − α ln x − ln a ∫ 0 1 t α − 1 f ( e t ln x + ( 1 − t ) ln a ) d t = f ( a ) ln x − ln a − α ( ln x − ln a ) α + 1 ∫ a x ( ln u − ln a ) α − 1 f ( u ) d u u = f ( a ) ln x − ln a − Γ ( α + 1 ) ( ln x − ln a ) α + 1 H J x − α f ( a ) .
(12)

Multiplying both sides of (11) and (12) by ( ln b − ln x ) α + 1 and − ( ln x − ln a ) α + 1 , respectively, we have

( ln b − ln x ) α + 1 ∫ 0 1 ( t α − 1 ) e t ln x + ( 1 − t ) ln b f ′ ( e t ln x + ( 1 − t ) ln b ) d t = − f ( b ) ( ln b − ln x ) α + Γ ( α + 1 ) H J x + α f ( b ) ,
(13)

and

− ( ln x − ln a ) α + 1 ∫ 0 1 ( t α − 1 ) e t ln x + ( 1 − t ) ln a f ′ ( e t ln x + ( 1 − t ) ln a ) d t = − f ( a ) ( ln x − ln a ) α + Γ ( α + 1 ) H J x − α f ( a ) .
(14)

From equalities (13) and (14), we obtain inequality (10). □

3 Hermite-Hadamard type inequalities involving Hadamard fractional integrals

In this section, we use the important Hadamard fractional integral identities including the first-order derivative of f in Section 2 to establish many interesting Hermite-Hadamard type inequalities involving Hadamard fractional integrals of the order α∈ R + for the functions satisfying monotonicity, convexity and s-e-condition.

3.1 Using monotonicity

Theorem 3.1Letf:[a,b]→Rbe a differentiable mapping on(a,b)with0<a<b. Ifα∈(0,1], f ′ ∈L[a,b]and is nondecreasing, then the following inequality for fractional integrals holds:

| Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a + b 2 ) | ≤ 1 2 [ b − a + b ( α + 2 ) α + 1 + a b ( ln b − ln a ) 2 ( α + 1 ) ] | f ′ ( b ) | .
(15)

Proof Using Lemma 2.1 and the nondecreasing property of f ′ , we find

| Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a + b 2 ) | ≤ b − a 2 ∫ 0 1 | f ′ ( b − t ( b − a ) ) | d t + ln b − ln a 2 ∫ 0 1 | ( 1 − t ) α − t α | e ln b − t ( ln b − ln a ) | f ′ ( e ln b − t ( ln b − ln a ) ) | d t ≤ b − a 2 | f ′ ( b ) | + ln b − ln a 2 ∫ 0 1 2 [ ( 1 − t ) α − t α ] e ln b − t ( ln b − ln a ) | f ′ ( b ) | d t + ln b − ln a 2 ∫ 1 2 1 [ t α − ( 1 − t ) α ] e ln b − t ( ln b − ln a ) | f ′ ( b ) | d t = b − a 2 | f ′ ( b ) | + b ( ln b − ln a ) 2 | f ′ ( b ) | ( K 1 + K 2 ) ,
(16)

where

K 1 = ∫ 0 1 2 [ ( 1 − t ) α − t α ] e − t ( ln b − ln a ) d t , K 2 = ∫ 1 2 1 [ t α − ( 1 − t ) α ] e − t ( ln b − ln a ) d t .

Calculating K 1 and K 2 , we have

K 1 = ∫ 0 1 2 [ ( 1 − t ) α − t α ] e − t ( ln b − ln a ) d t ≤ ∫ 0 1 2 ( 1 − 2 t ) α e − t ( ln b − ln a ) d t = 1 2 ∫ 0 1 ( 1 − s ) ( α + 1 ) − 1 e − 1 2 ( ln b − ln a ) s d s ≤ 1 2 max { 1 , 2 − α } ( 1 + 1 α + 1 ) ( ln b − ln a 2 ) − 1 ≤ α + 2 ( α + 1 ) ( ln b − ln a ) ,
(17)

and

K 2 = ∫ 1 2 1 [ t α − ( 1 − t ) α ] e − t ( ln b − ln a ) d t ≤ ∫ 1 2 1 ( 2 t − 1 ) α e − t ( ln b − ln a ) d t = 1 2 ∫ 1 2 ( s − 1 ) α e − ln b − ln a 2 s d t = 1 2 e − ( ln b − ln a ) ∫ 0 1 ( 1 − τ ) α e ln b − ln a 2 τ d τ ≤ 1 2 e − ln b − ln a 2 ∫ 0 1 ( 1 − τ ) α d τ = a b 2 ( α + 1 ) ,
(18)

where Lemmas 1.5 and 1.6 are used.

Thus, if we use (17) and (18) in (16), we obtain the inequality of (15). This completes the proof. □

Theorem 3.2Letf:[a,b]→Rbe a differentiable mapping on(a,b)with0<a<b. Ifα∈(0,1], f ′ ∈L[a,b]and is nondecreasing, then the following inequality for fractional integrals holds:

| Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a + b 2 ) | ≤ [ b − a 2 + b ( ln b − ln a ) α + 1 ( 1 − 1 2 α ) ] | f ′ ( b ) | .

Proof Using Lemma 2.1 and the nondecreasing property of f ′ , one can obtain

| Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a + b 2 ) | ≤ b − a 2 | f ′ ( b ) | + b ( ln b − ln a ) 2 | f ′ ( b ) | [ ∫ 0 1 2 [ ( 1 − t ) α − t α ] d t + ∫ 1 2 1 [ t α − ( 1 − t ) α ] d t ] = [ b − a 2 + b ( ln b − ln a ) α + 1 ( 1 − 1 2 α ) ] | f ′ ( b ) | .

The proof is completed. □

Theorem 3.3Letf:[a,b]→Rbe a differentiable mapping on(a,b)with0<a<b. Ifα∈(0,1], f ′ ∈L[a,b]and is nondecreasing, then the following inequality for fractional integrals holds:

| Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a b ) | ≤ b ( ln b − ln a ) 2 [ 1 + α + 2 ( α + 1 ) ( ln b − ln a ) + a b 2 ( α + 1 ) ] | f ′ ( b ) | .

Proof Using Lemma 2.2 and the nondecreasing property of f ′ , we find

| Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a b ) | ≤ ln b − ln a 2 [ ∫ 0 1 b | f ′ ( b ) | d t + ∫ 0 1 | ( 1 − t ) α − t α | e t ln a + ( 1 − t ) ln b f ′ ( e t ln a + ( 1 − t ) ln b ) d t ] ≤ b ( ln b − ln a ) 2 [ 1 + α + 2 ( α + 1 ) ( ln b − ln a ) + a b 2 ( α + 1 ) ] | f ′ ( b ) | .

This completes the proof. □

Theorem 3.4Letf:[a,b]→Rbe a differentiable mapping on(a,b)with0<a<b. Ifα∈(0,1], f ′ ∈L[a,b]and is nondecreasing, then the following inequality for fractional integrals holds:

| Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a b ) | ≤ b ( ln b − ln a ) 2 [ 1 + 2 α + 1 ( 1 − 1 2 α ) ] | f ′ ( b ) | .

Proof Using Lemma 2.2 and the nondecreasing property of f ′ , we find

| Γ ( α + 1 ) 2 ( ln b − ln a ) α [ H J a + α f ( b ) + H J b − α f ( a ) ] − f ( a b ) | ≤ ln b − ln a 2 [ ∫ 0 1 b | f ′ ( b ) | d t + ∫ 0 1 | ( 1 − t ) α − t α | b | f ′ ( b ) | d t ] ≤ b ( ln b − ln a ) 2 [ 1 + 2 α + 1 ( 1 − 1 2 α ) ] | f ′ ( b ) | .

This completes the proof. □

3.2 Using s-e-condition

Theorem 3.5Letf:[a,b]⊂(0,∞)→Rbe a differentiable mapping on(a,b)with0<a<bsuch that f ′ ∈L(a,b). If| f ′ |satisfiess-e-condition on[a,b]for some fixeds∈(0,1]and| f ′ (x)|≤M, x∈[a,b], then the following inequality for fractional integrals withα>0holds:

| Γ ( α + 1 ) [ H J x − α f ( a ) + H J x + α f ( b ) ] − [ f ( a ) ( ln x − ln a ) α + f ( b ) ( ln b − ln x ) α ] | ≤ M b [ 1 + 2 α s + 1 − Γ ( α + 1 ) Γ ( s + 1 ) Γ ( α + s + 1 ) ] ( ln x − ln a ) α + 1 + ( ln b − ln x ) α + 1 α + s + 1 .

Proof From Lemma 2.3 and since | f ′ | satisfies s-e-condition on [a,b], we have

| Γ ( α + 1 ) [ H J x − α f ( a ) + H J x + α f ( b ) ] − [ f ( a ) ( ln x − ln a ) α + f ( b ) ( ln b − ln x ) α ] | ≤ ( ln b − ln x ) α + 1 ∫ 0 1 ( 1 − t α ) e t ln x + ( 1 − t ) ln b | f ′ ( e t ln x + ( 1 − t ) ln b ) | d t + ( ln x − ln a ) α + 1 ∫ 0 1 ( 1 − t α ) e t ln x + ( 1 − t ) ln a | f ′ ( e t ln x + ( 1 − t ) ln a ) | d t ≤ ( ln b − ln x ) α + 1 ∫ 0 1 ( 1 − t α ) x t b 1 − t ( t s | f ′ ( x ) | + ( 1 − t ) s | f ′ ( b ) | ) d t + ( ln x − ln a ) α + 1 ∫ 0 1 ( 1 − t α ) x t a 1 − t ( t s | f ′ ( x ) | + ( 1 − t ) s | f ′ ( a ) | ) d t ≤ ( ln b − ln x ) α + 1 M b ∫ 0 1 [ t s ( 1 − t α ) + ( 1 − t α ) ( 1 − t ) s ] d t + ( ln x − ln a ) α + 1 M b ∫ 0 1 [ t s ( 1 − t α ) + ( 1 − t α ) ( 1 − t ) s ] d t = M b [ ( ln x − ln a ) α + 1 + ( ln b − ln x ) α + 1 ] ∫ 0 1 [ t s ( 1 − t α ) + ( 1 − t ) s − t α ( 1 − t ) s ] d t = M b [ 2 s + 1 − 1 s + α + 1 − Γ ( α + 1 ) Γ ( s + 1 ) Γ ( α + s + 2 ) ] [ ( ln x − ln a ) α + 1 + ( ln b − ln x ) α + 1 ] = M b [ 1 + 2 α s + 1 − Γ ( α + 1 ) Γ ( s + 1 ) Γ ( α + s + 1 ) ] ( ln x − ln a ) α + 1 + ( ln b − ln x ) α + 1 α + s + 1 ,

where we use the fact that x t a 1 − t ≤x≤b and x t b 1 − t ≤b via

∫ 0 1 t s ( 1 − t α ) dt= 1 s + 1 − 1 s + α + 1 ,

and

∫ 0 1 [ ( 1 − t ) s − t α ( 1 − t ) s ] dt= 1 s + 1 −B(α+1,s+1)= 1 s + 1 − Γ ( α + 1 ) Γ ( s + 1 ) Γ ( α + s + 2 ) .

So, using the reduction formula Γ(n+1)=nΓ(n) (n>0) for the Euler gamma function, the proof is completed. □

Theorem 3.6Letf:[a,b]⊂(0,∞)→Rbe a differentiable mapping on(a,b)with0<a<bsuch that f ′ ∈L(a,b). If | f ′ | q satisfiess-e-condition on[a,b]for some fixeds∈(0,1]and| f ′ (x)|≤M, x∈[a,b], then the following inequality for fractional integrals withα>0holds:

| Γ ( α + 1 ) [ H J x − α f ( a ) + H J x + α f ( b ) ] − [ f ( a ) ( ln x − ln a ) α + f ( b ) ( ln b − ln x ) α ] | ≤ M b ( 1 − 1 α + 1 ) 1 p [ 1 + 2 α s + 1 − Γ ( α + 1 ) Γ ( s + 1 ) Γ ( α + s + 1 ) ] 1 q × ( ln x − ln a ) α + 1 + ( ln b − ln x ) α + 1 ( α + s + 1 ) 1 q ,

where 1 p + 1 q =1.

Proof Using Lemma 2.3 and the well-known Hölder inequality, and since | f ′ | q satisfies s-e-condition on [a,b], we have

| Γ ( α + 1 ) [ H J x − α f ( a ) + H J x + α f ( b ) ] − [ f ( a ) ( ln x − ln a ) α + f ( b ) ( ln b − ln x ) α ] | ≤ ( ln b − ln x ) α + 1 ∫ 0 1 ( 1 − t α ) e t ln x + ( 1 − t ) ln b | f ′ ( e t ln x + ( 1 − t ) ln b ) | d t + ( ln x − ln a ) α + 1 ∫ 0 1 ( 1 − t α ) e t ln x + ( 1 − t ) ln a | f ′ ( e t ln x + ( 1 − t ) ln a ) | d t ≤ ( ln b − ln x ) α + 1 b ∫ 0 1 ( 1 − t α ) | f ′ ( e t ln x + ( 1 − t ) ln b ) | d t + ( ln x − ln a ) α + 1 b ∫ 0 1 ( 1 − t α ) | f ′ ( e t ln x + ( 1 − t ) ln a ) | d t ≤ ( ln b − ln x ) α + 1 b ( ∫ 0 1 ( 1 − t α ) d t ) 1 p ( ∫ 0 1 ( 1 − t α ) | f ′ ( e t ln x + ( 1 − t ) ln b ) | q d t ) 1 q + ( ln x − ln a ) α + 1 b ( ∫ 0 1 ( 1 − t α ) d t ) 1 p ( ∫ 0 1 ( 1 − t α ) | f ′ ( e t ln x + ( 1 − t ) ln a ) | q d t ) 1 q ≤ ( ln b − ln x ) α + 1 b ( 1 − 1 α + 1 ) 1 p ( ∫ 0 1 ( 1 − t α ) [ t s | f ′ ( x ) | q + ( 1 − t ) s | f ′ ( b ) | q ] d t ) 1 q + ( ln x − ln a ) α + 1 b ( 1 − 1 α + 1 ) 1 p ( ∫ 0 1 ( 1 − t α ) [ t s | f ′ ( x ) | q + ( 1 − t ) s | f ′ ( a ) | q ] d t ) 1 q ≤ ( ln b − ln x ) α + 1 M b ( 1 − 1 α + 1 ) 1 p ( ∫ 0 1 ( 1 − t α ) [ t s + ( 1 − t ) s ] d t ) 1 q + ( ln x − ln a ) α + 1 M b ( 1 − 1 α + 1 ) 1 p ( ∫ 0 1 ( 1 − t α ) [ t s + ( 1 − t ) s ] d t ) 1 q = M b ( 1 − 1 α + 1 ) 1 p [ 1 + 2 α s + 1 − Γ ( α + 1 ) Γ ( s + 1 ) Γ ( α + s + 1 ) ] 1 q × ( ln x − ln a ) α + 1 + ( ln b − ln x ) α + 1 ( α + s + 1 ) 1 q .

This completes the proof. □

3.3 Using monotonicity and convexity

Noting Remark 1.4 and repeating the same procedures in Theorem 3.5 and Theorem 3.6, we can derive the following results immediately.

Theorem 3.7Letf:[a,b]⊂(0,∞)→Rbe a differentiable mapping on(a,b)with0<a<bsuch that f ′ ∈L(a,b). If| f ′ |is convex and nondecreasing on[a,b]and| f ′ (x)|≤M, x∈[a,b], then the following inequality for fractional integrals withα>0holds:

| Γ ( α + 1 ) [ H J x − α f ( a ) + H J x + α f ( b ) ] − [ f ( a ) ( ln x − ln a ) α + f ( b ) ( ln b − ln x ) α ] | ≤ α M b ( ln x − ln a ) α + 1 + ( ln b − ln x ) α + 1 α + 1 .

Theorem 3.8Letf:[a,b]⊂(0,∞)→Rbe a differentiable mapping on(a,b)with0<a<bsuch that f ′ ∈L(a,b). If | f ′ | q is convex and nondecreasing on[a,b]and| f ′ (x)|≤M, x∈[a,b], then the following inequality for fractional integrals withα>0holds:

| Γ ( α + 1 ) [ H J x − α f ( a ) + H J x + α f ( b ) ] − [ f ( a ) ( ln x − ln a ) α + f ( b ) ( ln b − ln x ) α ] | ≤ α 1 q M b ( 1 − 1 α + 1 ) 1 p ( ln x − ln a ) α + 1 + ( ln b − ln x ) α + 1 ( α + 1 ) 1 q ,

where 1 p + 1 q =1.

4 Applications to special means

Consider the following special means (see Pearce and Pec̆arić [28]) for arbitrary real numbers α, β, α≠β as follows:

  1. (i)

    H(α,β)= 2 1 α + 1 β , α,β∈R∖{0},

  2. (ii)

    A(α,β)= α + β 2 , α,β∈R,

  3. (iii)

    L(α,β)= β − α ln | β | − ln | α | , |α|≠|β|, αβ≠0,

  4. (iv)

    L n (α,β)= [ β n + 1 − α n + 1 ( n + 1 ) ( β − α ) ] 1 n , n∈Z∖{−1,0}, α,β∈R, α≠β.

Now, using the results obtained in Section 3, we give some applications to special means of real numbers.

Proposition 4.1Leta,b∈ R + ∖{0}, a<b. Then

| L ( a , b ) − A ( a , b ) | ≤ 1 2 [ 5 b 2 − a + a b ( ln b − ln a ) 4 ] ,
(19)
| L ( a , b ) − A ( a , b ) | ≤ 1 2 [ b − a + b ( ln b − ln a ) 2 ] ,
(20)
|L(a,b)− [ A ( a , b ) H ( a , b ) ] 1 2 |≤ b ( ln b − ln a ) 8 ( 4 + 6 ln b − ln a + a b ) ,
(21)

and

|L(a,b)− [ A ( a , b ) H ( a , b ) ] 1 2 |≤ 3 b ( ln b − ln a ) 4 .
(22)

Proof Applying Theorems 3.1, 3.2, 3.3 and 3.4 for f(x)=x and α=1, one can obtain the results immediately. □

Proposition 4.2Leta,b∈ R + ∖{0}, a<b. Then

| L ( a n , b n ) − A n ( a , b ) | ≤ n b n − 1 2 [ 5 b 2 − a + a b ( ln b − ln a ) 4 ] ,
(23)
| L ( a n , b n ) − A n ( a , b ) | ≤ n b n − 1 2 [ b − a + b ( ln b − ln a ) 2 ] ,
(24)
|L ( a n , b n ) − [ A ( a , b ) H ( a , b ) ] n 2 |≤ n b n ( ln b − ln a ) 8 ( 4 + 6 ln b − ln a + a b ) ,
(25)

and

|L ( a n , b n ) − [ A ( a , b ) H ( a , b ) ] n 2 |≤ 3 n b n ( ln b − ln a ) 4 .
(26)

Proof Applying Theorems 3.1, 3.2, 3.3 and 3.4 for f(x)= x n and α=1, one can obtain the results immediately. □

Proposition 4.3Leta,b∈ R + ∖{0}, a<bandn∈Z, |n|≥2. Then

|L(a,b) L n n (a,b)− A n + 1 (a,b)|≤ ( n + 1 ) b n 2 [ 5 b 2 − a + a b ( ln b − ln a ) 4 ] ,
(27)
|L(a,b) L n n (a,b)− A n + 1 (a,b)|≤ ( n + 1 ) b n 2 [ b − a + b ( ln b − ln a ) 2 ] ,
(28)
| L ( a , b ) L n n ( a , b ) − [ A ( a , b ) H ( a , b ) ] n + 1 2 | ≤ ( n + 1 ) b n + 1 ( ln b − ln a ) 8 ( 4 + 6 ln b − ln a + a b ) ,
(29)

and

|L(a,b) L n n (a,b)− [ A ( a , b ) H ( a , b ) ] n + 1 2 |≤ 3 ( n + 1 ) b n + 1 ( ln b − ln a ) 4 .
(30)

Proof Applying Theorems 3.1, 3.2, 3.3 and 3.4 for f(x)= x n + 1 and α=1, x∈R, z∈Z, |n|≥2, one can obtain the result immediately. □

Proposition 4.4Leta,b∈ R + ∖{0} (a<b), a − 1 > b − 1 . Then we have, forn∈Z, |n|≥2

(c1)

| L ( b − 1 , a − 1 ) − H − 1 ( b , a ) | ≤ 1 2 ( 5 2 a − 1 b + ln b − ln a 4 a b ) ,

(c2)

| L ( b − 1 , a − 1 ) − H − 1 ( b , a ) | ≤ 1 2 a ( b − a b + ln b − ln a 2 ) ,

(c3)

|L ( b − 1 , a − 1 ) − [ A ( a , b ) H ( a , b ) ] − 1 2 |≤ ln b − ln a 8 a ( 4 + 6 ln b − ln a + a b ) ,

(c4)

|L ( b − 1 , a − 1 ) − [ A ( a , b ) H ( a , b ) ] − 1 2 |≤ 3 ( ln b − ln a ) 4 a ,

(c5)

| L ( b − n , a − n ) − H − n ( b , a ) | ≤ n 2 a n − 1 ( 5 2 a − 1 b + ln b − ln a 4 a b ) ,

(c6)

| L ( b − n , a − n ) − H − n ( b , a ) | ≤ n 2 a n ( b − a b + ln b − ln a 2 ) ,

(c7)

|L ( b − n , a − n ) − [ A ( a , b ) H ( a , b ) ] − n 2 |≤ n ( ln b − ln a ) 8 a n ( 4 + 6 ln b − ln a + a b ) ,

(c8)

|L ( b − n , a − n ) − [ A ( a , b ) H ( a , b ) ] − n 2 |≤ 3 n ( ln b − ln a ) 4 a n ,

(c9)

|L ( b − 1 , a − 1 ) L n n ( b − 1 , a − 1 ) − H − n − 1 (b,a)|≤ n + 1 2 a n ( 5 2 a − 1 b + ln b − ln a 4 a b ) ,

(c10)

|L ( b − 1 , a − 1 ) L n n ( b − 1 , a − 1 ) − H − n − 1 (b,a)|≤ n + 1 2 a n + 1 ( b − a b + ln b − ln a 2 ) ,

(c11)

| L ( b − 1 , a − 1 ) L n n ( b − 1 , a − 1 ) − [ A ( a , b ) H ( a , b ) ] − n + 1 2 | ≤ ( n + 1 ) ( ln b − ln a ) 8 a n + 1 ( 4 + 6 ln b − ln a + a b ) ,

(c12)

|L ( b − 1 , a − 1 ) L n n ( b − 1 , a − 1 ) − [ A ( a , b ) H ( a , b ) ] − n + 1 2 |≤ 3 ( n + 1 ) ( ln b − ln a ) 4 a n + 1 .

Proof Making the substitutions a→ b − 1 , b→ a − 1 in (19)-(30), one can obtain desired inequalities respectively, where A − 1 ( a − 1 , b − 1 )=H(a,b)= 2 1 a + 1 b , b − 1 < a − 1 . □

References

  1. Mitrinović DS, Lacković IB: Hermite and convexity. Aequ. Math. 1985, 28: 229–232. 10.1007/BF02189414

    Article  MathSciNet  MATH  Google Scholar 

  2. Pec̆arić JE, Proschan F, Tong YL: Convex Functions, Partial Orderings, and Statistical Applications. Academic Press, San Diego; 1992.

    Google Scholar 

  3. Abramovich S, Barić J, Pec̆arić J: Fejer and Hermite-Hadamard type inequalities for superquadratic functions. J. Math. Anal. Appl. 2008, 344: 1048–1056. 10.1016/j.jmaa.2008.03.051

    Article  MathSciNet  MATH  Google Scholar 

  4. Cal J, Carcamob J, Escauriaza L: A general multidimensional Hermite-Hadamard type inequality. J. Math. Anal. Appl. 2009, 356: 659–663. 10.1016/j.jmaa.2009.03.044

    Article  MathSciNet  MATH  Google Scholar 

  5. Ödemir ME, Avci M, Set E: On some inequalities of Hermite-Hadamard type via m -convexity. Appl. Math. Lett. 2010, 23: 1065–1070. 10.1016/j.aml.2010.04.037

    Article  MathSciNet  MATH  Google Scholar 

  6. Ödemir ME, Avci M, Kavurmaci H:Hermite-Hadamard-type inequalities via (α,m)-convexity. Comput. Math. Appl. 2011, 61: 2614–2620. 10.1016/j.camwa.2011.02.053

    Article  MathSciNet  MATH  Google Scholar 

  7. Dragomir SS: Hermite-Hadamard’s type inequalities for operator convex functions. Appl. Math. Comput. 2011, 218: 766–772. 10.1016/j.amc.2011.01.056

    Article  MathSciNet  MATH  Google Scholar 

  8. Dragomir SS: Hermite-Hadamard’s type inequalities for convex functions of selfadjoint operators in Hilbert spaces. Linear Algebra Appl. 2012, 436: 1503–1515. 10.1016/j.laa.2011.08.050

    Article  MathSciNet  MATH  Google Scholar 

  9. Sarikaya MZ, Aktan N: On the generalization of some integral inequalities and their applications. Math. Comput. Model. 2011, 54: 2175–2182. 10.1016/j.mcm.2011.05.026

    Article  MathSciNet  MATH  Google Scholar 

  10. Xiao Z, Zhang Z, Wu Y: On weighted Hermite-Hadamard inequalities. Appl. Math. Comput. 2011, 218: 1147–1152. 10.1016/j.amc.2011.03.081

    Article  MathSciNet  MATH  Google Scholar 

  11. Bessenyei M: The Hermite-Hadamard inequality in Beckenbach’s setting. J. Math. Anal. Appl. 2010, 364: 366–383. 10.1016/j.jmaa.2009.11.015

    Article  MathSciNet  MATH  Google Scholar 

  12. Tseng K, Hwang S, Hsu K: Hadamard-type and Bullen-type inequalities for Lipschitzian functions and their applications. Comput. Math. Appl. 2012, 64: 651–660. 10.1016/j.camwa.2011.12.076

    Article  MathSciNet  MATH  Google Scholar 

  13. Niculescu CP: The Hermite-Hadamard inequality for log-convex functions. Nonlinear Anal. TMA 2012, 75: 662–669. 10.1016/j.na.2011.08.066

    Article  MathSciNet  MATH  Google Scholar 

  14. Baleanu D, Machado JAT, Luo AC-J: Fractional Dynamics and Control. Springer, New York; 2012.

    Book  Google Scholar 

  15. Diethelm K Lecture Notes in Mathematics. In The Analysis of Fractional Differential Equations. Springer, Berlin; 2010.

    Chapter  Google Scholar 

  16. Kilbas AA, Srivastava HM, Trujillo JJ: Theory and Applications of Fractional Differential Equations. Elsevier, Amsterdam; 2006.

    MATH  Google Scholar 

  17. Lakshmikantham V, Leela S, Devi JV: Theory of Fractional Dynamic Systems. Cambridge Scientific Publishers, Cambridge; 2009.

    MATH  Google Scholar 

  18. Miller KS, Ross B: An Introduction to the Fractional Calculus and Differential Equations. Wiley, New York; 1993.

    MATH  Google Scholar 

  19. Michalski, MW: Derivatives of noninteger order and their applications. Dissertationes Mathematicae CCCXXVIII. Inst. Math., Polish Acad. Sci. (1993)

    Google Scholar 

  20. Podlubny I: Fractional Differential Equations. Academic Press, San Diego; 1999.

    MATH  Google Scholar 

  21. Tarasov VE: Fractional Dynamics: Application of Fractional Calculus to Dynamics of Particles, Fields and Media. Springer, Berlin; 2011.

    Google Scholar 

  22. Sarikaya MZ, Set E, Yaldiz H, Başak N: Hermite-Hadamard’s inequalities for fractional integrals and related fractional inequalities. Math. Comput. Model. 2013, 57: 2403–2407. 10.1016/j.mcm.2011.12.048

    Article  MATH  Google Scholar 

  23. Wang, J, Li, X, Zhu, C: Refinements of Hermite-Hadamard type inequalities involving fractional integrals. Bull. Belg. Math. Soc. Simon Stevin 20 (2013, in press)

    Google Scholar 

  24. Set E: New inequalities of Ostrowski type for mappings whose derivatives are s -convex in the second sense via fractional integrals. Comput. Math. Appl. 2012, 63: 1147–1154. 10.1016/j.camwa.2011.12.023

    Article  MathSciNet  MATH  Google Scholar 

  25. Wang J, Li X, Fec̆kan M, Zhou Y: Hermite-Hadamard-type inequalities for Riemann-Liouville fractional integrals via two kinds of convexity. Appl. Anal. 2012. 10.1080/00036811.2012.727986

    Google Scholar 

  26. Wang, J, Deng, J, FÄ•ckan, M: Exploring s-e-condition and applications to some Ostrowski type inequalities via Hadamard fractional integrals. Math. Slovaca (2013, in press)

    MATH  Google Scholar 

  27. Prudnikov AP, Brychkov YA, Marichev OI: Integral and series. 1. In Elementary Functions. Nauka, Moscow; 1981.

    Google Scholar 

  28. Pearce CEM, Pec̆arić JE: Inequalities for differentiable mappings with application to special means and quadrature formula. Appl. Math. Lett. 2000, 13: 51–55.

    Article  MathSciNet  Google Scholar 

Download references

Acknowledgements

The first author acknowledges the support by the National Natural Science Foundation of China (11201091), Key Projects of Science and Technology Research in the Chinese Ministry of Education (211169) and Key Support Subject (Applied Mathematics) of Guizhou Normal College. The third author acknowledges the support by Key Projects of Science and Technology Research in the Chinese Ministry of Education (211169) and the third author acknowledges the support by the National Natural Science Foundation of China (10971173) and Specialized Research Fund for the Doctoral Program of Higher Education (20114301110001).

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to JinRong Wang.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

This work was carried out in collaboration between all authors. JRW and YZ raised these interesting problems in the research. JRW, CZ and YZ proved the theorems, interpreted the results and wrote the article. All authors defined the research theme, read and approved the manuscript.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 2.0 International License (https://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and permissions

About this article

Cite this article

Wang, J., Zhu, C. & Zhou, Y. New generalized Hermite-Hadamard type inequalities and applications to special means. J Inequal Appl 2013, 325 (2013). https://doi.org/10.1186/1029-242X-2013-325

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/1029-242X-2013-325

Keywords