On Hermite-Hadamard Inequalities for ℎ-Preinvex Functions

Author(s): Muhammad Aslam Noor, Khalida Inayat Noor, Muhammad Uzair Awan and
Jueyou Li
Source: Filomat , 2014, Vol. 28, No. 7 (2014), pp. 1463-1474
Published by: University of Nis, Faculty of Sciences and Mathematics

Stable URL: https://www.jstor.org/stable/24897643

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

is collaborating with JSTOR to digitize, preserve and extend access to

Filomat

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
Filomat 28:7 (2014), 1463–1474 Published by Faculty of Sciences and Mathematics,
DOI 10.2298/FIL1407463N University of Niš, Serbia
Available at: http://www.pmf.ni.ac.rs/filomat

On Hermite-Hadamard Inequalities for h-Preinvex Functions

Muhammad Aslam Noora , Khalida Inayat Noora , Muhammad Uzair Awana , Jueyou Lib
a Mathematics Department, COMSATS Institute of Information Technology, Park Road, Islamabad, Pakistan.
b Department of Mathematics, Chongqing Normal University, Chongqing, 400047, P. R. China.

Abstract. The objective of this paper is to obtain some Hermite-Hadamard type inequalities for h-preinvex
functions. Firstly, a new kind of generalized h-convex functions, termed h-preinvex functions, is introduced
through relaxing the concept of h-convexity introduced by Varosanec. Some Hermite-Hadamard type
inequalities for h-preinvex functions are established under certain conditions. Our results can be viewed
as generalization of several previously known results. Results proved in this paper may stimulate further
research in different areas of pure and applied sciences.

1. Introduction

In recent years, several extensions and generalizations have been considered for classical convexity. A
significant generalization of convex functions is that of preinvex functions introduced by Weir and Mond
[33]. It is well known that the preinvex functions and invex sets may not be convex functions and convex
sets. For the applications, properties and other aspects of the preinvex functions, see [1, 3, 11, 15, 16, 24, 33].

Various refinements of the Hermite-Hadamard inequalities for the convex functions and their variant forms
are being obtained in the literature by many researchers (see [2–4, 6–10, 14, 15, 17, 19, 21, 23, 25, 29–31]).
In [32], Varosanec introduced the concept of h-convex functions, which include classical convex functions,
s-convex functions [5], Godunova-Levin functions [12] and P-functions [10] as special cases. Sarikaya et al.
[31] proved some Hermite-Hadamard inequalities for h-convex functions. Noor [19, 21] has established sev-
eral Hermite-Hadamard inequalities for preinvex, log-preinvex functions and product of two log-preinvex
functions. For some recent developments, see [26–28].

Motivated and inspired by the recent activities in this area, we obtain some new Hermite-Hadamard
type inequalities for h-preinvex functions through relaxing the concept of h-convex functions. This is
the main motivation of this paper. In particular, our results include the previously known results for
h-convex functions as special cases. The interested readers are encouraged to find the novel and innovative
applications of these results in other areas.

2010 Mathematics Subject Classification. 26A33; 26A51; 26D15

Keywords. Convex functions, h-convex, preinvex, h-preinvex, Hermite-Hadamard inequality.
Received: 01 July 2013; Accepted: 03 November 2013
Communicated by Predrag Stanimirović
Email addresses: [email protected] (Muhammad Aslam Noor), [email protected] (Khalida Inayat Noor),
[email protected] (Muhammad Uzair Awan), [email protected] (Jueyou Li)

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1464

2. Preliminaries

Let K be a nonempty closed set in Rn . Let f : K → R be a continuous function and let η(., .) : K × K → Rn
be a continuous bifunction. First of all, we recall some known results and concepts.
Definition 2.1 ([33]). A set K is said to be invex set with respect to η(., .), if

u + tη(v, u) ∈ K, ∀u, v ∈ K, t ∈ [0, 1]. (1)

The invex set K is also called η-connected set.

Remark 2.2 ([1]). We would like to mention that Definition 2.1 of an invex set has a clear geometric interpretation.
This definition essentially says that there is a path starting from a point u which is contained in K. We do not require
that the point v should be one of the end points of the path. This observation plays an important role in our analysis.
Note that, if we demand that v should be an end point of the path for every pair of points u, v ∈ K, then η(v, u) = v − u,
and consequently invexity reduces to convexity. Thus, it is true that every convex set is also an invex set with respect
to η(v, u) = v − u, but the converse is not necessarily true, see [18, 34] and the references therein. For the sake of
simplicity, we always assume that K = [u, u + η(v, u)], unless otherwise specified.

Definition 2.3 ([33]). A function f is said to be preinvex with respect to arbitrary bifunction η(., .), if

f (u + tη(v, u)) ≤ (1 − t) f (u) + t f (v), ∀u, v ∈ K, t ∈ [0, 1]. (2)

The function f is said to be preincave if and only if − f is preinvex. For the basic properties and applications
of the preinvex functions in variational inequalities, see [1, 11, 16, 18, 20, 22, 24, 33, 34].
For η(v, u) = v − u in (2), the preinvex functions becomes convex functions in the classical sense.
Definition 2.4. A function f : K → R is said to be convex in the classical sense, if

f (u + t(v − u)) ≤ (1 − t) f (u) + t f (v), ∀u, v ∈ K, t ∈ [0, 1].

From Definitions 2.3 and 2.4 it is obvious that every convex function is a preinvex function. However it is
known [33] that preinvex functions may not be convex functions.
Definition 2.5 ([32]). Let I, J be intervals in R, (0, 1) ⊆ J, and let h : J → R be a non-negative function. We say
that a non-negative function f : I → R is called h-convex function, or that f belongs to the class SX(h, I), if we have

f ((1 − t)u + tv) ≤ h(1 − t) f (u) + h(t) f (v), ∀u, v ∈ I, t ∈ (0, 1). (3)

If the above inequality is reversed, then f is said to be h-concave function.

Remark 2.6. It is worth pointing out that not all convex functions belong to the class of h-convex ones and this
inconvenience can be avoided omitting the assumption that f is nonnegative, see [4]. So, in the further text we
assume that h and f are real functions without assumption of non-negativity. Obviously, if h(t) = t, then all convex
(concave) functions belong to the class of h-convex (concave) functions; if h(t) = ts , s ∈ (0, 1), then all s-convex
(concave) functions in the second sense belong to the class of h-convex (concave) functions. Additional, this large
class of h-convex functions contains Godunova-Levin functions and P-functions, if h(t) = t−1 , h(t) = 1, respectively.

Definition 2.7. Let h : J → R where (0, 1) ⊆ J be an interval in R, and let K be an invex set with respect to η(., .).
A function f : K → R is called h-preinvex with respect to η(., .), if

f (u + tη(v, u)) ≤ h(1 − t) f (u) + h(t) f (v), u, v ∈ K, t ∈ (0, 1).

If above inequality is reversed, then f is said to be h-preconcave with respect to bifunction η(., .).
Now, we discuss some special cases of h-preinvex functions.
I. If h(t) = t, then Definition 2.7 reduces to Definition 2.3.
II. If h(t) = ts with s ∈ (0, 1), then we have the definition of s-preinvex functions.

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1465

Definition 2.8. A function f : K → R is said to be s-preinvex with respect to η(., .), if

f (u + tη(v, u)) ≤ (1 − t)s f (u) + (t)s f (v), u, v ∈ K, t ∈ [0, 1], s ∈ (0, 1).

III. If h(t) = t−1 , then we have the definition of Q-preinvex functions.

Definition 2.9. A function f : K → R is said to be Q-preinvex with respect to η(., .), if

1 1
f (u + tη(v, u)) ≤ f (u) + f (v), u, v ∈ K, t ∈ (0, 1).
1−t t
IV. If h(t) = 1, then we have the definition of P-preinvex functions.
Definition 2.10. A function f : K → R is said to be P-preinvex with respect to η(., .), if

f (u + tη(v, u)) ≤ f (u) + f (v), u, v ∈ K, t ∈ [0, 1].

Now we give an example of h-preinvex function. Which also illustrates the fact that an h-preinvex functions
may not be preinvex functions.
Example 2.11. Let us consider the function f (u) = −|u| with

v − u, i f
 uv < 0,
η(v, u) = 

u − v, i f
 uv > 0.

If function h(t) = ts , t ∈ (0, 1), where s ≥ 1, then by Definition 2.7, f is h-preinvex function with respect to η(., .) on
R \ {0}. However, if h(t) = ts , t ∈ (0, 1) where s < 1, by letting x = 1, y = 1, t = 1/2, s = 1/2, we have
√
f (u + tη(v, u)) = f (1) = −1 > − 2 = ts f (u) + (1 − t)s f (v),

which shows that f is not h-preinvex with respect to the same η(., .).
To prove some results in the paper, we need the well-known Condition C introduced by Mohan and Neogy
in [18].
Condition C. Let K ⊂ R be an invex set with respect to bifunction η(., .). Then for any x, y ∈ K and t ∈ [0, 1],

η(y, y + tη(x, y)) = −tη(x, y),

η(x, y + tη(x, y)) = (1 − t)η(x, y).

Note that for every x, y ∈ K, t1 , t2 ∈ [0, 1] and from Condition C, we have

η(y + t2 η(x, y), y + t1 η(x, y)) = (t2 − t1 )η(x, y).

It is worth mentioning that Condition C has played a crucial and significant role in the development of the
variational-like inequalities and optimization problems, see [11, 18, 22] and the references therein.
From now onwards I = [a, a + η(b, a)] will be the interval unless otherwise specified.

3. Main Results

In this section, we prove our main results.

Using the technique of Varosanec [32], we prove that the product of two h-preinvex functions is again a
h-preinvex function.
Theorem 3.1. Let f and w be two h-preinvex functions. Then their product f w is also h-preinvex function provided
if f and w are similarly ordered functions and h(t) + h(1 − t) ≤ 1.

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1466

Proof. Since f and w are h-preinvex functions, then

f (x + tη(y, x))w(x + tη(y, x))
≤ [h(1 − t) f (x) + h(t) f (y)][h(1 − t)w(x) + h(t)w(y)]
= [h(1 − t)]2 f (x)w(x) + h(t)h(1 − t)[ f (x)w(y) + f (y)w(x)] + [h(t)]2 f (y)w(y)
≤ [h(1 − t)]2 f (x)w(x) + h(t)h(1 − t)[ f (x)w(x) + f (y)w(y)] + [h(t)]2 f (y)w(y)
= [h(1 − t) f (x)w(x) + h(t) f (y)w(y)][h(t) + h(1 − t)]
≤ h(1 − t) f (x)w(x) + h(t) f (y)w(y),
where we have used the fact that h(t) + h(1 − t) ≤ 1. This shows that the product of two h-preinvex functions
is also h-preinvex.
We prove the following result which will be helpful in proving our next result of Fejer type inequality for
h-preinvex function.
Lemma 3.2. Let f be h-preinvex function, then we have
f (2a + η(b, a) − x) ≤ [h(t) + h(1 − t)] f (a) + f (b) − f (x).
 
(4)
Proof. Given x = a + tη(b, a) ∈ I, then we have
f (2a + η(b, a) − x) = f (a + (1 − t)η(b, a))
≤ h(t) f (a) + h(1 − t) f (b)
= [h(t) + h(1 − t)] f (a) + f (b) − h(1 − t) f (a) + h(t) f (b)
   

≤ [h(t) + h(1 − t)] f (a) + f (b) − f (a + tη(b, a))

 

= [h(t) + h(1 − t)] f (a) + f (b) − f (x).

 

This proves the required result.

Theorem 3.3. Let f : I → (0, ∞) be a h-preinvex function with a < a + η(b, a), h( 12 ) , 0 and w : [a, a + η(b, a)] → R
is a non-negative, integrable function and symmetric about a + 21 η(b, a), then, using Condition C, we have

! a+η(b,a) a+η(b,a) a+η(b,a)

2a + η(b, a) f (a) + f (b)
Z Z Z
1
 f w(x)dx ≤ f (x)w(x)dx ≤ (h(t) + h(1 − t)) w(x)dx.
2h 1 2 2
2 a a a

Proof. Since f is h-preinvex function, we have

! a+η(b,a)
2a + η(b, a)
Z
1
 f w(x)dx
2h 1 2
2 a
a+η(b,a)
2a + η(b, a)
Z !
1
=   f w(x)dx
2h 1 2
2 a
a+η(b,a)
2a + η(b, a) − x + x
Z !
1
=   f w(x)dx
2h 1 2
2 a
a+η(b,a)
Z
1 1 
   
≤   h f (2a + η(b, a) − x) + f (x) w(x)dx
2h 1 2
2 a

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1467
a+η(b,a)
Z a+η(b,a)
Z
1 1
= f (2a + η(b, a) − x)w(2a + η(b, a) − x)dx + f (x)w(x)dx
2 2
a a
a+η(b,a)
Z
= f (x)w(x)dx
a
a+η(b,a)
Z a+η(b,a)
Z
1 1
= f (2a + η(b, a) − x)w(2a + η(b, a) − x)dx + f (x)w(x)dx
2 2
a a
a+η(b,a)
Z a+η(b,a)
Z
1 1
= f (2a + η(b, a) − x)w(x)dx + f (x)w(x)dx
2 2
a a
a+η(b,a)
Z a+η(b,a)
Z
1 1
≤ [(h(t) + h(1 − t))[ f (a) + f (b)] − f (x)]w(x)dx + f (x)w(x)dx
2 2
a a
a+η(b,a)
f (a) + f (b)
Z
≤ (h(t) + h(1 − t)) w(x)dx.
2
a

This completes the proof.

Theorem 3.4. Let f : I → (0, ∞) and w : I → (0, ∞) be h1 -preinvex and h2 -preinvex functions respectively with
a < a + η(b, a), h1 ( 21 ) , 0, h2 ( 12 ) , 0. If the bifunction η(., .) satisfies Condition C, then
a+η(b,a)
2a + η(b, a) 2a + η(b, a)
! ! Z
1 1
 f w − f (x)w(x)dx
η(b, a)
 
2h1 1
h2 12 2 2
2 a
Z1 Z1
≤ M(a, b) h1 (t)h2 (1 − t)dt + N(a, b) h1 (t)h2 (t)dt.
0 0

where
M(a, b) = f (a)w(a) + f (b)w(b) (5)
and
N(a, b) = f (a)w(b) + f (b)w(a). (6)
Proof. Since f and w are h1 -preinvex and h2 -preinvex functions respectively and η satisfies Condition C, we
have
2a + η(b, a) 2a + η(b, a)
! !
f w
2 2
1
= f (a + (1 − t)η(b, a) + η(a + tη(b, a), a + (1 − t)η(b, a)))
2
1
× w(a + (1 − t)η(b, a) + η(a + tη(b, a), a + (1 − t)η(b, a)))
2
1 1
   
≤ h1 h2 [ f (a + tη(b, a)) + f (a + (1 − t)η(b, a))][w(a + tη(b, a)) + w(a + (1 − t)η(b, a))]
2 2

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1468
1 1
   
≤ h1 h2 [ f (a + tη(b, a))w(a + tη(b, a)) + f (a + (1 − t)η(b, a))w(a + (1 − t)η(b, a))]+
2 2
1 1
   
h1 h2 {[h1 (t)h2 (1 − t) + h1 (1 − t)h2 (t)]M(a, b) + [h1 (t)h2 (t) + h1 (1 − t)h2 (1 − t)]N(a, b)} .
2 2
Integrating above inequality with respect to t on [0,1], we have
    a+η(b,a)
2a + η(b, a) 2a + η(b, a) 2h1 12 h2 21 Z
! !
f w − f (x)w(x)dx
2 2 η(b, a)
a
Z1 Z1
 
1 1 
     
≤ 2h1 h2 M(a, b) h1 (t)h2 (1 − t)dt + N(a, b) h1 (t)h2 (t)dt .

2 2  
0 0

From the above inequality, we can obtain the required result.

Theorem 3.5. Let f : I → (0, ∞) and w : I → (0, ∞) be h1 -preinvex and h2 -preinvex functions respectively with
a < a + η(b, a), then we have

a+η(b,a)
Z Z1 Z1
1
f (x)w(x)dx ≤ M(a, b) h1 (t)h2 (t)dt + N(a, b) h1 (t)h2 (1 − t)dt,
η(b, a)
a 0 0

where M(a, b) and N(a, b) are given by (5) and (6).

Proof. Let f, w be nonnegative h1 -preinvex and ,h2 -preinvex function respectively, then for all t ∈ [0, 1], we
have

f (a + tη(b, a))w(a + tη(b, a))

≤ [h1 (1 − t) f (a) + h1 (t) f (b)][h2 (1 − t)w(a) + h2 tw(b)]
= h1 (1 − t)h2 (1 − t) f (a)w(a) + h1 (t)h2 (1 − t) f (b)w(a) + h1 (1 − t)h2 (t) f (a)w(b) + h1 (t)h2 (t) f (b)w(b).

Integrating above inequality with respect to t on [0, 1], we have

a+η(b,a)
Z
1
f (x)w(x)dx
η(b, a)
a
Z1 Z1
≤ f (a)w(a) h1 (1 − t)h2 (1 − t)dt + f (a)w(b) h1 (1 − t)h2 (t)dt
0 0
Z1 Z1
+ f (b)w(a) h1 (t)h2 (1 − t)dt + f (b)w(b) h1 (t)h2 (t)dt
0 0
Z1 Z1
= [ f (a)w(a) + f (b)w(b)] h1 (t)h2 (t)dt + [ f (a)w(b) + f (b)w(a))] h1 (t)h2 (1 − t)dt
0 0
Z1 Z1
= M(a, b) h1 (t)h2 (t)dt + N(a, b) h1 (t)h2 (1 − t)dt.
0 0

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1469

Thus we have
a+η(b,a)
Z Z1 Z1
1
f (x)w(x)dx ≤ M(a, b) h1 (t)h2 (t)dt + N(a, b) h1 (t)h2 (1 − t)dt.
η(b, a)
a 0 0

This completes the proof.

Next, we prove some Hermite-Hadamard type inequalities for differentiable h-preinvex functions.
We need the following result, which is an extension of a result proved by Dragomir et al. [6].
Lemma 3.6 ([3]). Let f : I → (0, ∞) be a differentiable mapping a, a + η(b, a) ∈ I with a < a + η(b, a). If
f 0 ∈ L1 [a, a + η(b, a)], then
a+η(b,a)  1 
f (a) + f (a + η(b, a)) η(b, a) 
Z Z
1

f (x)dx − =  (1 − 2t) f 0 (a + tη(b, a))dt .

η(b, a) 2 2   
a 0

Proof. Consider

Z1 1 Z1
f (a + tη(b, a)) f (a + tη(b, a))
(1 − 2t) f (a + tη(b, a))dt =
0
(1 − 2t) + 2 dt
η(b, a) 0 η(b, a)
0 0
Z1
f (a + η(b, a)) f (a) 2
=− − + f (a + tη(b, a))dt
η(b, a) η(b, a) η(b, a)
0
a+η(b,a)
f (a) + f (a + η(b, a))
Z
2
=− + f (x)dx
η(b, a) [η(b, a)]2
a
a+η(b,a)
 
+ + η(b,
Z
f (a) f (a a))
 
2  1
= + f (x)dx .
 
−
η(b, a)  2 η(b, a) 
a

Suitable rearrangements complete the proof.

Theorem 3.7. Let f : I → R be a differentiable mapping on I◦ , a, a + η(b, a) ∈ I◦ with a < a + η(b, a). If | f 0 | is a
h-preinvex on [a, a + η(b, a)], then
a+η(b,a) Z1
f (a) + f (a + η(b, a)) η(b, a)  0
Z
1
| f (a)| + | f 0 (b)|

− f (x)dx ≤ |1 − 2t|h(t)dt.
2 η(b, a) 2
a 0

Proof. Using Lemma 3.6, it follows that

a+η(b,a) Z1
f (a) + f (a + η(b, a)) η(b, a)
Z
1
− f (x)dx = (1 − 2t) f 0 (a + tη(b, a))dt
2 η(b, a) 2
a 0
Z1
η(b, a)
≤ |1 − 2t|| f 0 (a + tη(b, a))|dt
2
0

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1470
Z1
η(b, a)
|1 − 2t| h(1 − t)| f 0 (a)| + h(t)| f 0 (b)| dt
 
≤
2
0
Z1
η(b, a)  0
| f (a)| + | f 0 (b)|

≤ |1 − 2t|h(t)dt.
2
0

This completes the proof.

Theorem 3.8. Let f : I → R be a differentiable mapping on I◦ , a, a + η(b, a) ∈ I◦ with a < a + η(b, a). If | f 0 |q is a
h-preinvex on [a, a + η(b, a)] where q ≥ 1 and p1 + 1q = 1, then we have

a+η(b,a)  p−1
Z1
 p
f (a) + f (a + η(b, a)) η(b, a)  0
Z
1

h p p i 
− f (x)dx ≤  | f (a)| p−1 + | f (b)| p−1
0
h(t)dt

2 η(b, a) 1 
2(p + 1) p 
 
a 0

Proof. Using Lemma 3.6, we have

a+η(b,a) Z1
f (a) + f (a + η(b, a)) η(b, a)
Z
1
− f (x)dx = (1 − 2t) f 0 (a + tη(b, a))dt
2 η(b, a) 2
a 0
Z1
η(b, a)
≤ |1 − 2t|| f 0 (a + tη(b, a))|dt.
2
0

Using Hölder’s integral inequality, we have

a+η(b,a)  1  1p  1  1q
f (a) + f (a + η(b, a)) η(b, a) 
Z Z Z
1
  
− f (x)dx ≤  |1 − 2t|p dt  | f 0 (a + tη(b, a))|q dt ,
  
(7)
2 η(b, a) 2     
a 0 0

where 1
p + 1
q = 1. Now using h-preinvexity of f , we have

Z1 Z1
| f (a + tη(b, a))| ≤
0 q
[h(1 − t)| f 0 (a)|q + h(t)| f 0 (b)|q ]dt, (8)
0 0

where
Z1
1
|1 − 2t|p dt = . (9)
p+1
0

Using (7) (8) and (9), we have the required result.

Theorem 3.9. Let f : I → R be a differentiable mapping on I◦ , a, a + η(b, a) ∈ I◦ with a < a + η(b, a). If | f 0 |q is a
h-preinvex on [a, a + η(b, a)] where q ≥ 1, then we have

a+η(b,a)  1q
Z1

1
f (a) + f (a + η(b, a)) η(b, a)  1 1− q
Z
1
 
 | f (a)| + | f 0 (b)|q  .
 0 q 
− f (x)dx ≤ h(t)dt

2 η(b, a) 2 2  
a 0

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1471

Proof. Using Lemma 3.6, we have

a+η(b,a) Z1
f (a) + f (a + η(b, a)) η(b, a)
Z
1
− f (x)dx = (1 − 2t) f 0 (a + tη(b, a))dt
2 η(b, a) 2
a 0
Z1
η(b, a)
≤ |1 − 2t|| f 0 (a + tη(b, a))|dt.
2
0

Due to the power-mean inequality, we have

1− 1q  1q
Z1
 1  1
Z  Z 
|1 − 2t|| f 0 (a + tη(b, a))|dt ≤  |1 − 2t|dt  |1 − 2t|| f 0 (a + tη(b, a))|q dt .
   
  (10)
 
0 0 0

By the h-preinvexity of f , we have

Z1 Z1
|1 − 2t|| f (a + tη(b, a))| dt ≤
0 q
|1 − 2t|[h(1 − t)| f 0 (a)|q + h(t)| f 0 (b)|q ]dt
0 0
Z1
= [| f (a)| + | f (b)| ]
0 q 0 q
|1 − 2t|h(t)dt, (11)
0

where
Z1
1
|1 − 2t|dt = (12)
2
0

Using (10), (11) and (12), we have the required result.

Now using the technique of [14], we prove the following result which play a key role in proving our next
results.
Lemma 3.10. Let f : I → (0, ∞) be a differentiable mapping a, a + η(b, a) ∈ I with a < a + η(b, a). If f 0 ∈
L1 [a, a + η(b, a)], then
1
a+η(b,a) Z2 Z1
2a + η(b, a)
! Z
1 1
f − f (x)dx = t f (a + tη(b, a))dt + (t − 1) f 0 (a + tη(b, a))dt
0
η(b, a) 2 [η(b, a)]2
a 0 1
2

Proof. Consider
1
Z2 Z1
t f (a + tη(b, a))dt +
0
(t − 1) f 0 (a + tη(b, a))dt
0 1
2
1
1 Z2 1 Z1
f (a + tη(b, a)) 2 f (a + tη(b, a)) f (a + tη(b, a)) f (a + tη(b, a))
= t − dt + (t − 1) − dt
η(b, a) 0 η(b, a) η(b, a) 1 η(b, a)
2
0 1
2

a+η(b,a)
2a + η(b, a)
! Z
1 1
= f − f (x)dx.
η(b, a) 2 [η(b, a)]2
a

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1472

This completes the proof.

Theorem 3.11. Let f : I → R be a differentiable mapping on I◦ , a, a + η(b, a) ∈ I◦ with a < a + η(b, a). If | f 0 | is a
h-preinvex on [a, a + η(b, a)], then we have

a+η(b,a)
2a + η(b, a)
Z !
1
f (x)dx − f
η(b, a) 2
a
  1   1 


Z2
 Z1 

Z2
 Z1 

≤ η(b, a) | f (a)|  th(1 − t)dt + (1 − t)h(1 − t)dt + | f (b)|  th(t)dt + (1 − t)h(t)dt .

 0 
 
 0 

    
0 1 1 0
2 2

Proof. Using Lemma 3.10 we have

a+η(b,a)
2a + η(b, a)
Z !
1
f (x)dx − f
[η(b, a)] 2
a
 1 
Z2
 Z1 

≤ η(b, a)  t f 0 (a + tη(b, a))dt + (1 − t) f 0 (a + tη(b, a))dt
 
0 1
2
 1 
Z2
 Z1 

≤ η(b, a)  t[h(1 − t)| f (a)| + h(t)| f (b)|]dt + (1 − t)[h(1 − t)| f (a)| + h(t)| f (b)|]dt

 0 0 0 0
 
0 1
2
  1   1 
 Z2 Z 1  Z2 Z 1 
    
= η(b, a) | f (a)|  th(1 − t)dt + (1 − t)h(1 − t)dt + | f (b)|  th(t)dt + (1 − t)h(t)dt .
 0   0 
    
0 1 1 0
2 2

This proves the result.

Theorem 3.12. Let f : I → R be a differentiable mapping on I◦ , a, a + η(b, a) ∈ I◦ with a < a + η(b, a). If | f 0 |q is a
h-preinvex on [a, a + η(b, a)] where q ≥ 1 and p1 + 1q = 1, then we have

a+η(b,a)
f (a) + f (a + η(b, a))
Z
1
− f (x)dx
2 η(b, a)
a

1
 1q   1q 
! 1p  2   1  
η(b, a)
Z   Z 
1     
 0 q
+ 0 q
+ 0 q
+ 0 q
  
≤ − t)| f h(t)| f − t)| f h(t)| f

[h(1 (a)| (b)| ]dt [h(1 (a)| (b)| ]dt
   
2(p + 1) 
 
2   
 
 
 
  
0 
 
1  

2

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1473

Proof. Using Lemma 3.10 and Hölder’s inequality, we have

a+η(b,a)
2a + η(b, a)
Z !
1
f (x)dx − f
[η(b, a)] 2
a
 1 
Z2
 Z1 

≤ η(b, a)  t f (a + tη(b, a))dt + (1 − t) f (a + tη(b, a))dt

 0 0
 
0 1
2
  1p  1q   1p  1q 
Z12
 1 
 Z2  Z1  Z1  
        
≤ η(b, a)  t dt

 
 p   | f 0 (a + tη(b, a))|q dt +  (1 − t)p dt
   
 | f 0 (a + tη(b, a))|q dt 
   (13)
        
 0 0 1 1 
2 2

Using h-preinvexity of f , we have

1 1
Z2 Z2
| f 0 (a + tη(b, a))|q dt ≤ [h(1 − t)| f 0 (a)|q + h(t)| f 0 (b)|q ]dt. (14)
0 0

Z1 Z1
| f (a + tη(b, a))| dt ≤
0 q
[h(1 − t)| f 0 (a)|q + h(t)| f 0 (b)|q ]dt. (15)
1 1
2 2

And
1
Z2 Z1
1
t dt =
p
(1 − t)p dt = . (16)
2p+1 (p + 1)
0 1
2

Using (13), (14), (15) and (16), we have the required result.

Remark 3.13. Now we point out some special cases which are included in our main results.
1. If η(b, a) = b − a and h(t) = t, h(t) = ts , h(t) = t−1 and h(t) = 1, then our results reduce to the results for
classical convexity, s-convexity, Q-convexity and P-convexity respectively.
2. If h(t) = t, then our results reduce to the results for classical preinvexity.
3. If h(t) = t, h(t) = ts , h(t) = t−1 and h(t) = 1, then our results reduce to the results for classical preinvexity,
s-preinvexity, Q-preinvexity and P-preinvexity respectively, which also appears to be new.

Acknowledgements

The authors are thankful to the Editor and anonymous referees for their constructive comments and
valuable suggestions. The authors also would like to thank Dr. S. M. Junaid Zaidi, Rector, COMSATS
Institute of Information Technology, Pakistan, for providing excellent research and academic environment.
This research is supported by HEC NRPU project No: 20 − 1966/R&D/11 − 2553.

References
[1] T. Antczak, Mean value in invexity analysis, Nonl. Anal. 60(2005), 1473-1484.
[2] Rui-Fang Bai, Feng Qi, Bo-Yan Xi, Hermite-Hadamard type inequalities for the m− and (α, m)-logarithmically convex functions,
Filomat, 27 (1), 1-7.

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms
 M. A. Noor et al / Filomat 28:7 (2014), 1463–1474 1474

[3] A. Barani, A. G. Ghazanfari, S. S. Dragomir, Hermite-Hadamard inequality for functions whose derivatives absolute values are
preinvex, Journal of Inequalities and Applications 2012, 2012:247.
[4] M. Bombardelli, S. Varosanec, Properties of h-convex functions related to the Hermite-Hadamard-Fejer inequalities, Compt.
Math. Appl. 58(2009) 1869-1877.
[5] W. W. Breckner, Stetigkeitsaussagen fiir eine Klasse verallgemeinerter convexer funktionen in topologischen linearen Raumen.
Pupl. Inst. Math. 23 (1978), 13-20.
[6] S. S. Dragomir, R. P. Agarwal, Two inequalities for differentiable mappings and applications to special means of real numbers
and to Trapezoidal formula, Appl. Math. Lett. 11(5) (1998),91-95.
[7] S. S. Dragomir, S. Fitzpatrick, The Hadamard inequalities for s-convex functions in the second sense, Demonstratio Math.
32(4)(1999) 687-696.
[8] S. S. Dragomir, A. Mcandrew, Refinment of the Hermite-Hadamard inequality for convex functions, J. Inequal. Oure Appl. Math.,
6(2005), 1-6.
[9] S. S. Dragomir, C. E. M. Pearce, Selected topics on Hermite-Hadamard inequalities and applications, Victoria University, 2000.
[10] S. S. Dragomir, J. Pecaric, L. E. Persson, Some inequalities of Hadamard type. Soochow J. Math. 21 (1995), 335-341.
[11] A. Farajzadeh, M. A. Noor, K. I. Noor, Vector nonsmooth variational-like inequalities and optimization problems, Nonlinear
Analysi, 71, (2009), 3471-3476.
[12] E. K. Godunova, V. I. Levin, Neravenstva dlja funkcii sirokogo klassa, soderzascego vypuklye, monotonnye i nekotorye drugie
vidy funkii. Vycislitel. Mat. i. Fiz. Mezvuzov. Sb. Nauc. Trudov, MGPI, Moskva. (1985) 138-142, (in Russian).
[13] H. Hudzik, L. Maligranda, Some remarks on s-convex functions, Aequationes Math. 48(1994) 100-111.
[14] U. S. Kirmaci, M. K. Bakula, M. E. Ozdemir, J. Pecaric, Hadamard-type inequalities for s-convex functions, Appl. Math. and
Comput. 193(2007), 26-35.
[15] M. Matloka, On some Hadamard-type inequalities for (h1 , h2 )-preinvex functions on the co-ordinates, Journal of Inequalities and
Applications 2013, 2013:227.
[16] S. K. Mishra, M. A. Noor, On vector variational-like inequality problems, J. Math. Anal. Appl. 311(2005), 69-75.
[17] D. S. Mitrinovic, J. E. Pecaric, A. M. Fink, Classical and new inequalities in analysis, Kluwer Academic, Dordrecht, 1993.
[18] S. R. Mohan, S. K. Neogy, On invex sets and preinvex functions, J. Math. Anal. Appl. 189(1995), 901-908.
[19] M. A. Noor, Hermite-Hadamard integral inequalities for log-preinvex functions, J. Math. Anal. Approx. Theory, 2(2007), 126-131.
[20] M. A. Noor, Invex equilibrium problems, J. Math. Anal. Appl, 302(2005), 463-475.
[21] M. A. Noor, On Hermite-Hadamard integral inequalities for involving two log-preinvex functions, J. Inequal. Pure and Appl.
Math., 3(2007), 75-81.
[22] M. A. Noor, Variational like inequalities, Optimization, 30(1994), 323-330.
[23] M. A. Noor, M. U. Awan, K. I. Noor, On some inequalities for relative semi-convex functions, Journal of Inequalities and
Applications 2013, 2013:332.
[24] M. A. Noor, K. I. Noor, Hemi-equilibrium-like problems, Nonlinear Analysis, 64(2006), 2631–2642.
[25] M. A. Noor, K. I. Noor, M. U. Awan, Hermite-Hadamard inequalities for relative semi-convex functions and applications, Filomat
28(2) (2014), 221-230.
[26] M. A. Noor, K. I. Noor, M. U. Awan, Generalized convexity and integral inequalities, Appl. Math. Inf. Sci. 9(1)(2015), 233–243.
[27] M. A. Noor, K. I. Noor, M. U. Awan, Integral inequalities for coordinated Harmonically convex functions, Complex Var. Ellipt.
Equat. 2014: http://dx.doi.org/10.1080/17476933.2014.976814
[28] M. A. Noor, K. I. Noor, M. U. Awan, S. Khan, HermiteHadamard type inequalities for differentiable hϕ -preinvex functions, Arab.
J. Math. DOI 10.1007/s40065-014-0124-3.
[29] B. G. Pachpatte, Mathematical inequalities, North-Holland Library, Vol. 67, Elsevier Science, Amsterdam, Holland, 2005.
[30] B. G. Pachpatte, On some inequalities for convex functions, RGMIA Res. Rep. Coll. 6(E)(2003).
[31] M. Z. Sarikaya, A. Saglam, H. Yildrim, On some Hadamard-type inequalities for h-convex functons, J. Math. Inequal. 2(2008)
335-341.
[32] S. Varosanec, On h-convexity, J. Math. Anal. Appl. 326(2007), 26-35.
[33] T. Weir, B. Mond, Preinvex functions in multiobjective optimization, J. Math. Anal. Appl. 136(1988), 29-38.
[34] X. M. Yang, X. Q. Yang, K. L. Teo, Generalized invexity and generalized invariant monotonicity, J. Optim. Theory. Appl., 117(2003),
607-625.

This content downloaded from

185.89.98.114 on Wed, 05 Feb 2025 10:46:47 UTC
All use subject to https://about.jstor.org/terms