Open Access

Some generalizations of Hermite–Hadamard type inequalities

SpringerPlus20165:1661

https://doi.org/10.1186/s40064-016-3301-3

Received: 11 May 2016

Accepted: 11 September 2016

Published: 26 September 2016

Abstract

Some generalizations and refinements of Hermite–Hadamard type inequalities related to \(\eta\)-convex functions are investigated. Also applications for trapezoid and mid-point type inequalities are given.

Keywords

\(\eta\)-convex functionIntegral inequalitiesHermite–Hadamard inequality

Mathematics Subject Classification

26A5126D1552A01

Introduction and preliminaries

This paper generalizes some well-known results for Hermite–Hadamard integral inequality by generalizing the convex function factor of the integrand to be an \(\eta\)-convex function. The obtained results have as particular cases those previously obtained for convex functions in the integrand.

The following inequality is well known in the literature as the Hermite–Hadamard integral inequality (Pecaric et al. 1991):
$$f\left( {\frac{a + b}{2}} \right) \le \frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx \le \frac{f\left( a \right) + f\left( b \right)}{2}.$$
(1)
where \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) be a convex function. For more results about (1), see Alomari et al. (2010), Dragomir (1992), Kirmaci (2004), Pearce and Pecaric (2000), Rostamian Delavar and Dragomir (2016), Rostamian Delavar et al. (to appear), Wasowicz and Witkowski (2012), Yang (2001), Yang et al. (2004) and references therein.

Let \(I\) be an interval in real line \({\mathbb{R}}\). Consider \(\eta :A \times A \to B\) for appropriate \(A,B \subseteq {\mathbb{R}}\).

Definition 1 (Gordji et al. 2016)

A function \(f:I \to {\mathbb{R}}\) is called convex with respect to \(\eta\) (briefly \(\eta\)-convex), if
$$f\left( {tx + \left( {1 - t} \right)y} \right)\,\le\,f\left( y \right) + t\eta \left( {f\left( x \right),\,f\left( y \right)} \right),$$
(2)
for all \(x,y \in I\) and \(t \in \left[ {0,1} \right]\).

In fact above definition geometrically says that if a function is \(\eta\)-convex on \(I\), then its graph between any \(x,y \in I\) is on or under the path starting from \(\left( {y,f\left( y \right)} \right)\) and ending at \((x,f\left( y \right) + \eta \left( {f\left( x \right),f\left( y \right)} \right)\). If \(f\left( x \right)\) should be the end point of the path for every \(x,y \in I\), then we have \(\eta \left( {x,y} \right) = x - y\) and the function reduces to a convex one.

There exists \(\eta\)-convex functions for some bifunctions \(\eta\) that are not convex. We have the following simple examples:

Example 2 (Gordji et al. 2015)

a. Consider a function \(f:{\mathbb{R}} \to {\mathbb{R}}\) defined by
$$f\left( x \right) = \left\{ {\begin{array}{*{20}l} { - x,} \hfill &\quad {x \ge 0;} \hfill \\ {x,} \hfill &\quad {x < 0.} \hfill \\ \end{array} } \right.$$
and define a bifunction \(\eta\) as \(\eta \left( {x,y} \right) = - x - y\), for all \(x,y \in {\mathbb{R}}^{ - } = \left( { - \infty ,0} \right].\) It is not hard to check that \(f\) is a \(\eta\)-convex function but not a convex one.
b. Define the function \(f:{\mathbb{R}}^{ + } \to {\mathbb{R}}^{ + }\) by
$$f\left( x \right) = \left\{ {\begin{array}{*{20}l} {x,} \hfill &\quad {0 \le x \le 1;} \hfill \\ {1,} \hfill &\quad {x > 1.} \hfill \\ \end{array} } \right.$$
and define the bifunction \(\eta :{\mathbb{R}}^{ + } \times {\mathbb{R}}^{ + } \to {\mathbb{R}}^{ + }\) by
$$\eta \left( {x,y} \right) = \left\{ {\begin{array}{*{20}l} {x + y,} \hfill &\quad {x \le y;} \hfill \\ {2\left( {x + y} \right),} \hfill &\quad {x > y.} \hfill \\ \end{array} } \right.$$

Then \(f\) is \(\eta\)-convex but is not convex.

The following theorem is an important result:

Theorem 3

(Gordji et al. 2016) Suppose that \(f:I \to {\mathbb{R}}\) is a \(\eta\)-convex function and \(\eta\) is bounded from above on \(f\left( I \right) \times f\left( I \right)\) . Then \(f\) satisfies a Lipschitz condition on any closed interval \(\left[ {a,b} \right]\) contained in the interior \(I^{ \circ }\) of \(I\) . Hence, \(f\) is absolutely continuous on \(\left[ {a,b} \right]\) and continuous on \(I^{ \circ }\).

Remark 4

As a consequence of Theorem 3, an \(\eta\)-convex function \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) where \(\eta\) is bounded from above on \(f\left( {\left[ {a,b} \right]} \right) \times f\left( {\left[ {a,b} \right]} \right)\) is integrable.

The following simple lemma is required.

Lemma 5

Suppose that \(a,b \in {\mathbb{R}}\) . Then

(i) \(\hbox{min} \{ a,b\} \le \frac{a + b}{2}\).

(ii) if \(f\), \(g\) are integrable on \(\left[ {a,b} \right]\) then, \(\mathop \int \limits_{a}^{b} \hbox{min} \{ f,g\} = \hbox{min} \left\{ {\mathop \int \nolimits_{a}^{b} f,\mathop \int \nolimits_{a}^{b} g} \right\}\).

Proof

Assertions are consequence of this fact:
$$\hbox{min} \{ a,b\} = \frac{{a + b - \left| {a - b} \right|}}{2}.$$

We have a basic lemma:

Lemma 6

Let \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) be a \(\eta\)-convex function. Then for any \(t \in \left[ {0,1} \right]\) we have the inequalities
$$\begin{aligned} \frac{1}{2}\left[ {f\left( {ta + \left( {1 - t} \right)b} \right) + f\left( {\left( {1 - t} \right)a + tb} \right)} \right] & \le \hbox{min} \left\{ {f\left( b \right) + \frac{1}{2}\eta \left( {f\left( a \right),f\left( b \right)} \right),f\left( a \right) + \frac{1}{2}\eta \left( {f\left( b \right),f\left( a \right)} \right)} \right\} \\ & \le \frac{1}{2}\left[ {f\left( a \right) + f\left( b \right)} \right] + \frac{1}{4}\left[ {\eta \left( {f\left( a \right),f\left( b \right)} \right) + \eta \left( {f\left( b \right),f\left( a \right)} \right)} \right], \\ \end{aligned}$$
(3)
$$\frac{1}{2}\left[ {f\left( {ta + \left( {1 - t} \right)b} \right) + f\left( {\left( {1 - t} \right)a + tb} \right)} \right] \le \frac{1}{2}\left[ {f\left( a \right) + f\left( b \right)} \right] + t\frac{1}{2}\left[ {\eta \left( {f\left( a \right),f\left( b \right)} \right) + \eta \left( {f\left( b \right),f\left( a \right)} \right)} \right],$$
(4)
$$f({ta + ( {1 - t} )b} ) \le \frac{1}{2}[ {f(a) + f(b)] + {\frac{1}{2}} [t\eta ( {f(a),f(b)}) + ({1 - t})\eta ( {f( b ),f( a )} )} ]$$
(5)
and
$$\begin{aligned} f\left( {\frac{a + b}{2}} \right) & \le \hbox{min} \left\{ f\left( {ta + \left( {1 - t} \right)b} \right) +\, \frac{1}{2}\eta \left( {f\left( {\left( {1 - t} \right)a + tb} \right),f\left( {ta + \left( {1 - t} \right)b} \right)} \right),f\left( {\left( {1 - t} \right)a + tb} \right) + \frac{1}{2}\eta \left( {f\left( {ta + \left( {1 - t} \right)b} \right),f\left( {\left( {1 - t} \right)a + tb} \right)} \right)\right\} \\ & \le \frac{1}{2}\left[ {f\left( {\left( {1 - t} \right)a + tb} \right) + f\left( {ta + \left( {1 - t} \right)b} \right)} \right] \\ & \quad + \frac{1}{4}\eta \left( {f\left( {\left( {1 - t} \right)a + tb} \right),f\left( {ta + \left( {1 - t} \right)b} \right)} \right) \\ & \quad + \frac{1}{4}\eta \left( {f\left( {ta + \left( {1 - t} \right)b} \right),f\left( {\left( {1 - t} \right)a + tb} \right)} \right). \\ \end{aligned}$$
(6)

Proof

If in (2) we put \(t\) instead of \(1 - t\) and then add that inequality with (2) we have:
$$\frac{1}{2}\left[ {f\left( {ta + \left( {1 - t} \right)b} \right) + f\left( {\left( {1 - t} \right)a + tb} \right)} \right] \le f\left( b \right) + \frac{1}{2}\eta \left( {f\left( a \right),f\left( b \right)} \right)$$
(7)
for all \(t \in \left[ {0,1} \right].\)

If in (7) we replace \(a\) with \(b\) and add the result with (7), then we have (3).

Now, if in (2) we put \(a\) instead of \(b\) and then add that inequality with (2) we get:
$$f\left( {ta + \left( {1 - t} \right)b} \right) + f\left( {tb + \left( {1 - t} \right)a} \right) \le f\left( b \right) + f\left( a \right) + t\left[ {\eta \left( {f\left( a \right),f\left( b \right)} \right) + \eta \left( {f\left( b \right),f\left( a \right)} \right)} \right]$$
for all \(t \in \left[ {0,1} \right],\) which is equivalent to (4).
If we change \(a\) with \(b\), and \(t\) with \(1 - t\) in (2) and then add that inequality with (2) we get:
$$2f\left( {ta + \left( {1 - t} \right)b} \right) \le f\left( b \right) + f\left( a \right) + t\eta \left( {f\left( a \right),f\left( b \right)} \right) + \left( {1 - t} \right)\eta \left( {f\left( b \right),f\left( a \right)} \right)$$
for all \(t \in \left[ {0,1} \right]\) and the inequality (5) is proved.
Finally since we have
$$f\left( {\frac{a + b}{2}} \right) = f\left( {\frac{{ta + \left( {1 - t} \right)b + tb + \left( {1 - t} \right)a}}{2}} \right)$$
and
$$f\left( {\frac{a + b}{2}} \right) = f\left( {\frac{{tb + \left( {1 - t} \right)a + ta + \left( {1 - t} \right)b}}{2}} \right),$$
then by using (2) we can obtain (6) □

Hermite–Hadamard type inequalities

In this section we obtain some Hermite–Hadamard type integral inequalities which improve right and left side of (1) respectively.

Theorem 1

Let \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) be a \(\eta\)-convex function with \(\eta\) bounded from above on \(f\left( {\left[ {a,b} \right]} \right) \times f\left( {\left[ {a,b} \right]} \right)\) . Then we have inequalities
$$\begin{aligned} \frac{1}{2}\mathop \int \limits_{0}^{1} [f\left( {ta + \left( {1 - t} \right)b} \right) + f\left( {\left( {1 - t} \right)a + tb} \right)] & \le \hbox{min} \left\{ {f\left( b \right) + \frac{1}{2}\eta \left( {f\left( a \right),f\left( b \right)} \right),f\left( a \right) + \frac{1}{2}\eta \left( {f\left( b \right),f\left( a \right)} \right)} \right\} \\ & \le \frac{1}{2}\left[ {f\left( a \right) + f\left( b \right)} \right] + \frac{1}{4}\left[ {\eta \left( {f\left( a \right),f\left( b \right)} \right) + \eta \left( {f\left( b \right),f\left( a \right)} \right)} \right], \\ \end{aligned}$$
(8)
$$\begin{aligned} \frac{1}{2}\mathop \int \nolimits_{0}^{1} [f\left( {ta + \left( {1 - t} \right)b} \right) + f\left( {\left( {1 - t} \right)a + tb} \right)] \le \frac{1}{2}\left[ {f\left( a \right) + f\left( b \right)} \right] + \frac{1}{2}\left[ {\eta \left( {f\left( a \right),f\left( b \right)} \right) + \eta \left( {f\left( b \right),f\left( a \right)} \right)} \right]\mathop \int \nolimits_{0}^{1} tdt \hfill \\ \hfill \\ \end{aligned}$$
(9)
and
$$\mathop \int \nolimits_{0}^{1} f\left( {ta + \left( {1 - t} \right)b} \right)dt \le \frac{1}{2}\left[ {f\left( a \right) + f\left( b \right)} \right] + \frac{1}{2}\eta \left( {f\left( a \right),f\left( b \right)} \right)\mathop \int \nolimits_{0}^{1} tdt + \frac{1}{2}\eta \left( {f\left( b \right),f\left( a \right)} \right)\mathop \int \nolimits_{0}^{1} \left( {1 - t} \right)dt.$$
(10)

Proof

Since \(\eta\) is bounded from above on \(f\left( {\left[ {a,b} \right]} \right) \times f\left( {\left[ {a,b} \right]} \right)\), the note after Theorem 3, guarantees existence of above integrals. The inequalities (8)–(10) follow by Lemma 6 on integrating over \(t \in \left[ {0,1} \right].\)

Remark 2

If \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) is a \(\eta\)-convex function and \(\eta\) is bounded from above on \(f\left( {\left[ {a,b} \right]} \right) \times f\left( {\left[ {a,b} \right]} \right)\), then by Theorem 1 we have
$$\begin{aligned} \frac{1}{2}\mathop \int \nolimits_{a}^{b} [f\left( x \right) + f\left( {a + b - x} \right)]dx & \le \hbox{min} \left\{ {f\left( b \right) + \frac{1}{2}\eta \left( {f\left( a \right),f\left( b \right)} \right),f\left( a \right) + \frac{1}{2}\eta \left( {f\left( b \right),f\left( a \right)} \right)} \right\}\left( {b - a} \right) \\ & \le \frac{1}{2}\left[ {f\left( a \right) + f\left( b \right)} \right]\left( {b - a} \right) + \frac{1}{4}\left[ {\eta \left( {f\left( a \right),f\left( b \right)} \right) + \eta \left( {f\left( b \right),f\left( a \right)} \right)} \right]\left( {b - a} \right), \\ \end{aligned}$$
(11)
$$\frac{1}{2}\mathop \int \nolimits_{a}^{b} [f\left( x \right) + f\left( {a + b - x} \right)]dx \le \frac{1}{2}\left[ {f\left( a \right) + f\left( b \right)\left] {\left( {b - a} \right) + \frac{1}{2}} \right[\frac{{\eta \left( {f\left( a \right),f\left( b \right)} \right) + \eta \left( {f\left( b \right),f\left( a \right)} \right)}}{b - a}} \right]\mathop \int \nolimits_{a}^{b} \left( {x - a} \right)dx,$$
(12)
and
$$\mathop \int \nolimits_{a}^{b} f\left( x \right) \le \frac{1}{2}\left[ {f\left( a \right) + f\left( b \right)} \right]\left( {b - a} \right) + \frac{1}{2}\frac{{\eta \left( {f\left( a \right),f\left( b \right)} \right)}}{b - a}\mathop \int \nolimits_{a}^{b} \left( {x - a} \right)dx + \frac{1}{2}\frac{{\eta \left( {f\left( b \right),f\left( a \right)} \right)}}{b - a}\mathop \int \nolimits_{a}^{b} \left( {b - x} \right)dx.$$
(13)
All of inequalities (11)–(13) are different views for right side of generalized Hermite-Hadamard inequalities and finally can be stated as a unique form of
$$\frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx \le \frac{1}{2}\left[ {f\left( a \right) + f\left( b \right)} \right] + \frac{1}{4}\left[ {\eta \left( {f\left( a \right),f\left( b \right)} \right) + \eta \left( {f\left( b \right),f\left( a \right)} \right)} \right].$$
(14)

If we suppose that \(\eta \left( {x,y} \right) = x - y\), then we recapture right side of (1).

Also we can obtain the following result:

Theorem 3

Let \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) be a \(\eta\)-convex function with \(\eta\) bounded from above on \(f\left( {\left[ {a,b} \right]} \right) \times f\left( {\left[ {a,b} \right]} \right)\) . Then we have the inequalities:
$$\begin{aligned} f\left( {\frac{a + b}{2}} \right) & \le \mathop \int \nolimits_{0}^{1} \hbox{min} \left\{ f\left( {ta + \left( {1 - t} \right)b} \right) + \frac{1}{2}\eta \left( {f\left( {\left( {1 - t} \right)a + tb} \right),f\left( {ta + \left( {1 - t} \right)b} \right)} \right),f\left( {\left( {1 - t} \right)a + tb} \right) + \frac{1}{2}\eta \left( {f\left( {ta + \left( {1 - t} \right)b} \right),f\left( {\left( {1 - t} \right)a + tb} \right)} \right)\right\} dt \\ & \le \hbox{min} \left\{ \mathop \int \nolimits_{0}^{1} f\left( {ta + \left( {1 - t} \right)b} \right)dt + \frac{1}{2}\mathop \int \nolimits_{0}^{1} \eta \left( {f\left( {\left( {1 - t} \right)a + tb} \right),f\left( {ta + \left( {1 - t} \right)b} \right)} \right)dt,\mathop \int \nolimits_{0}^{1} f\left( {\left( {1 - t} \right)a + tb} \right)dt + \frac{1}{2}\mathop \int \nolimits_{0}^{1} \eta \left( {f\left( {ta + \left( {1 - t} \right)b} \right),f\left( {\left( {1 - t} \right)a + tb} \right)} \right)dt\right\} \\ & \le \mathop \int \nolimits_{0}^{1} \frac{{f\left( {ta + \left( {1 - t} \right)b} \right) + f\left( {\left( {1 - t} \right)a + tb} \right)}}{2}dt + \frac{1}{4}\mathop \int \nolimits_{0}^{1} \left[\eta \left( {f\left( {\left( {1 - t} \right)a + tb} \right),f\left( {ta + \left( {1 - t} \right)b} \right)} \right) +\,\eta \left( {f\left( {ta + \left( {1 - t} \right)b} \right),f\left( {\left( {1 - t} \right)a + tb} \right)} \right)\right]dt. \\ \end{aligned}$$
(15)

Proof

From (6) we have
$$\begin{aligned} f\left( {\frac{a + b}{2}} \right) \le \hbox{min} \left\{ f\left( {ta + \left( {1 - t} \right)b} \right) + \frac{1}{2}\eta \left( {f\left( {\left( {1 - t} \right)a + tb} \right),f\left( {ta + \left( {1 - t} \right)b} \right)} \right),f\left( {\left( {1 - t} \right)a + tb} \right) \quad +\,\frac{1}{2}\eta \left( {f\left( {ta + \left( {1 - t} \right)b} \right),f\left( {\left( {1 - t} \right)a + tb} \right)} \right)\right\} , \\ \end{aligned}$$
for any \(t \in \left[ {0,1} \right].\) Integrating over \(t\) we get the first inequality in (15). Now Using properties of Lemma 5 along with integrating rules gives
$$\begin{aligned}& \mathop \int \nolimits_{0}^{1} \hbox{min} \left\{ f\left( {ta + \left( {1 - t} \right)b} \right) + \frac{1}{2}\eta \left( {f\left( {\left( {1 - t} \right)a + tb} \right),f\left( {ta + \left( {1 - t} \right)b} \right)} \right), f\left( {\left( {1 - t} \right)a + tb} \right) + \frac{1}{2}\eta \left( {f\left( {ta + \left( {1 - t} \right)b} \right),f\left( {\left( {1 - t} \right)a + tb} \right)} \right)\right\} dt \hfill \\ &\le \hbox{min} \left \{ \mathop \int \nolimits_{0}^{1} f\left( {ta + \left( {1 - t} \right)b} \right)dt + \frac{1}{2}\mathop \int \nolimits_{0}^{1} \eta \left( {f\left( {\left( {1 - t} \right)a + tb} \right),f\left( {ta + \left( {1 - t} \right)b} \right)} \right)dt, \mathop \int \nolimits_{0}^{1} f\left( {\left( {1 - t} \right)a + tb} \right)dt + \frac{1}{2}\mathop \int \nolimits_{0}^{1} \eta \left( {f\left( {ta + \left( {1 - t} \right)b} \right),f\left( {\left( {1 - t} \right)a + tb} \right)} \right)dt\right\} \hfill \\& \le \mathop \int \nolimits_{0}^{1} \frac{{f\left( {ta + \left( {1 - t} \right)b} \right) + f\left( {\left( {1 - t} \right)a + tb} \right)}}{2}dt + \frac{1}{4}\mathop \int \nolimits_{0}^{1} \left[ {\eta \left( {f\left( {\left( {1 - t} \right)a + tb} \right),f\left( {ta + \left( {1 - t} \right)b} \right)} \right)} \right. \hfill \\& + \left. {\eta \left( {f\left( {ta + \left( {1 - t} \right)b} \right),f\left( {\left( {1 - t} \right)a + tb} \right)} \right)} \right]dt. \hfill \\ \end{aligned}$$

Remark 4

If \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) is a \(\eta\)-convex function and \(\eta\) is bounded from above on \(f\left( {\left[ {a,b} \right]} \right) \times f\left( {\left[ {a,b} \right]} \right)\), then by Theorem 3 we have
$$\begin{aligned} f\left( {\frac{a + b}{2}} \right)\left( {b - a} \right) & \le \mathop \int \nolimits_{a}^{b} \hbox{min} \left\{ {f\left( {a + b - x} \right) + \frac{1}{2}\eta \left( {f\left( x \right),f\left( {a + b - x} \right)} \right),} \right.\left. {f\left( x \right) + \frac{1}{2}\eta \left( {f\left( {a + b - x} \right),f\left( x \right)} \right)} \right\}dx \\ & \le \hbox{min} \left\{ {\mathop \int \nolimits_{a}^{b} f\left( {a + b - x} \right)dx + \frac{1}{2}\mathop \int \nolimits_{a}^{b} \eta \left( {f\left( x \right),f\left( {a + b - x} \right)} \right)dx,} \right.\left. {\mathop \int \nolimits_{a}^{b} f\left( x \right)dx + \frac{1}{2}\mathop \int \nolimits_{a}^{b} \eta \left( {f\left( {a + b - x} \right),f\left( x \right)} \right)dx} \right\} \\ & \le \mathop \int \nolimits_{a}^{b} \frac{{f\left( {a + b - x} \right) + f\left( x \right)}}{2}dx + \frac{1}{4}\mathop \int \nolimits_{a}^{b} [\eta \left( {f\left( x \right),f\left( {a + b - x} \right)} \right) \\ & \quad + \eta \left( {f\left( {a + b - x} \right),f\left( x \right)} \right)]dx = \mathop \int \nolimits_{a}^{b} f\left( x \right)dx + \frac{1}{2}\mathop \int \nolimits_{a}^{b} \eta \left( {f\left( x \right),f\left( {a + b - x} \right)} \right)dx, \\ \end{aligned}$$
(16)
which gives a refinement for left side of (1). If we suppose that \(\eta \left( {x,y} \right) = x - y\), then we recapture left side of (1).

Trapezoid and mid-point type inequalities

An interesting question in (1), is estimating the difference between left and middle terms and between right and middle terms. In this section we investigate about this question, when the absolute value of the derivative of a function is \(\eta\)-convex. We need Lemma 2.1 in Kirmaci (2004):

Lemma 1

Suppose that \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) is a differentiable mapping, \(g:\left[ {a,b} \right] \to {\mathbb{R}}^{ + }\) is a continuous mapping and \(f^{\prime}\) is integrable on \(\left[ {a,b} \right]\). Then
$$\frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx - f\left( {\frac{a + b}{2}} \right) = \left( {b - a} \right)\left[ {\mathop \int \nolimits_{0}^{1/2} tf^{\prime}\left( {ta + \left( {1 - t} \right)b} \right)dt + \mathop \int \nolimits_{1/2}^{1} (t - 1} )f^{\prime}\left( {ta + \left( {1 - t} \right)b} \right)dt\right],$$

Remark 2

In Lemma 1, if we use the change of variable \(x = tb + \left( {1 - t} \right)a\), then
$$\frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx - f\left( {\frac{a + b}{2}} \right) = \left( {b - a} \right)\left[ {\mathop \int \nolimits_{0}^{1/2} ( - t})f^{\prime}( {tb + ( {1 - t} )a} )dt + \mathop \int \nolimits_{1/2}^{1} (1 - t)f^{\prime}( {tb + ( {1 - t} )a} )dt\right],$$

Using Lemma 1, we can prove the following theorem to estimate the difference between the middle and left terms in (1).

Theorem 3

Suppose that \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) is a differentiable mapping and \(\left| {f^{\prime}} \right|\) is an \(\eta\)-convex mapping on \(\left[ {a,b} \right]\) with a bounded \(\eta\) from above. Then
$$\left| {\frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx - f\left( {\frac{a + b}{2}} \right)} \right| \le \frac{1}{8}\left( {b - a} \right)K,$$
where
$$K = \hbox{min} \left\{ \left| {f^{\prime}\left( b \right)} \right| + \frac{{\left| {\eta \left( {f^{\prime}\left( a \right),f^{\prime}\left( b \right)} \right)} \right|}}{2},\left| {f^{\prime}\left( a \right)} \right| + \frac{{\left| {\eta \left( {f^{\prime}\left( b \right),f^{\prime}\left( a \right)} \right)} \right|}}{2}\right\}$$

Proof

From \(\eta\)-convexity of \(\left| {f^{\prime}} \right|\), Theorem 3 and Lemma 1 it follows that
$$\begin{aligned} \left| {\frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx - f\left( {\frac{a + b}{2}} \right)} \right| & \le \left( {b - a} \right)\left\{ \mathop \int \nolimits_{0}^{1/2} t\left( {\left| {f^{\prime}\left( b \right)\left| { + t} \right|\eta \left( {f^{\prime}\left( a \right),f^{\prime}\left( b \right)} \right)} \right|} \right)dt + \mathop \int \nolimits_{1/2}^{1} (1 - t)\left( {\left| {f^{\prime}\left( b \right)\left| { + t} \right|\eta \left( {f^{\prime}\left( a \right),f^{\prime}\left( b \right)} \right)} \right|} \right)dt\right\} = \frac{1}{8}\left( {b - a} \right)\left[ {2\left| {f^{\prime}\left( b \right)\left| + \right|\eta \left( {f^{\prime}\left( a \right),f^{\prime}\left( b \right)} \right)} \right|} \right] = I \\ \end{aligned}$$
On the other hand according to Remark 2 we have
$$\begin{aligned} \left| {\frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx - f\left( {\frac{a + b}{2}} \right)} \right| & \le \left( {b - a} \right)\left\{ \mathop \int \nolimits_{0}^{1/2} ( - t)\left( {\left| {f^{\prime}\left( a \right)\left| { + t} \right|\eta \left( {f^{\prime}\left( b \right),f^{\prime}\left( a \right)} \right)} \right|} \right)dt + \mathop \int \nolimits_{1/2}^{1} (t - 1)\left( {\left| {f^{\prime}\left( a \right)\left| { + t} \right|\eta \left( {f^{\prime}\left( b \right),f^{\prime}\left( a \right)} \right)} \right|} \right)dt\right\} = \frac{1}{8}\left( {b - a} \right)\left[ {2\left| {f^{\prime}\left( a \right)\left| + \right|\eta \left( {f^{\prime}\left( b \right),f^{\prime}\left( a \right)} \right)} \right|} \right] = J \\ \end{aligned}$$
Then we can deduce the result from
$$\left| {\frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx - f\left( {\frac{a + b}{2}} \right)} \right| \le \hbox{min} \{ I,J\} .$$

Remark 4

If in the proof of Theorem 3 we consider \(\eta \left( {x,y} \right) = x - y\) for all \(x,y \in \left[ {a,b} \right]\), we approach to Theorem 2.2 in Kirmaci (2004).

The following is Lemma 2.1 in Dragomir and Agarwal (1998).

Lemma 5

Suppose that \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) is a differentiable function and \(f^{\prime}\) is an integrable function on \(\left[ {a,b} \right]\) . Then
$$\frac{f\left( a \right) + f\left( b \right)}{2} - \frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx = \frac{1}{b - a}\mathop \int \nolimits_{a}^{b} (x - \frac{a + b}{2})f^{\prime}\left( x \right)dx.$$
(17)

Using Lemma 5, we can prove the following theorem to estimate the difference between the middle and right terms in ( 1 ).

Theorem 6

Suppose that \(f:\left[ {a,b} \right] \to {\mathbb{R}}\) is a differentiable function and \(\left| {f^{\prime}} \right|\) is an \(\eta\)-convex function where \(\eta\) is bounded from above on \(\left[ {a,b} \right]\) . Then
$$\left| {\frac{f\left( a \right) + f\left( b \right)}{2} - \frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx} \right| \le \frac{1}{8}\left( {b - a} \right)K,$$
where
$$K = \hbox{min} \left\{ \left| {f^{\prime}\left( b \right)} \right| + \frac{{\left| {\eta \left( {f^{\prime}\left( a \right),f^{\prime}\left( b \right)} \right)} \right|}}{2},\left| {f^{\prime}\left( a \right)} \right| + \frac{{\left| {\eta \left( {f^{\prime}\left( b \right),f^{\prime}\left( a \right)} \right)} \right|}}{2}\right\} .$$

Proof

Using Lemma 5 and the change of the variable \(x = ta + \left( {1 - t} \right)b\), \(t \in \left[ {0,1} \right]\) in right hand of (7) along with the fact that \(\left| {f^{\prime}} \right|\) is \(\eta\)-convex imply that
$$\begin{aligned} \left| {\frac{f\left( a \right) + f\left( b \right)}{2} - \frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx} \right| & \le \frac{1}{b - a}\left|\mathop \int \nolimits_{a}^{b} (x - \frac{a + b}{2})f^{\prime}\left( x \right)dx\right| \\ & = \frac{{\left( {b - a} \right)}}{2}\left|\mathop \int \nolimits_{0}^{1} (1 - 2t)f^{\prime}\left( {ta + \left( {1 - t} \right)b} \right)dt\right| \\ & \le \frac{{\left( {b - a} \right)}}{2}\mathop \int \nolimits_{0}^{1} \left|\left( {1 - 2t} \right)\right|\left| {f^{\prime}\left( {ta + \left( {1 - t} \right)b} \right)} \right|dt \\ & \le \frac{{\left( {b - a} \right)}}{2}\mathop \int \nolimits_{0}^{1} |\left( {1 - 2t} \right)|\left[ {\left| {f^{\prime}\left( b \right)\left| { + t} \right|\eta \left( {f^{\prime}\left( a \right),f^{\prime}\left( b \right)} \right)} \right|} \right]dt \\ & = \frac{{\left( {b - a} \right)}}{4}\left[ {2\left| {f^{\prime}\left( b \right)\left| + \right|\eta \left( {f^{\prime}\left( a \right),f^{\prime}\left( b \right)} \right)} \right|} \right]. \\ \end{aligned}$$
(18)
Similarly if we use the change of variable \(x = tb + \left( {1 - t} \right)a\), \(t \in \left[ {0,1} \right]\) we have
$$\left| {\frac{f\left( a \right) + f\left( b \right)}{2} - \frac{1}{b - a}\mathop \int \nolimits_{a}^{b} f\left( x \right)dx} \right| \le \frac{{\left( {b - a} \right)}}{4}\left[ {2\left| {f^{\prime}\left( a \right)\left| + \right|\eta \left( {f^{\prime}\left( b \right),f^{\prime}\left( a \right)} \right)} \right|} \right].$$

Remark 7

Theorem 6 reduces to Theorem 2.2 in Dragomir and Agarwal (1998), if we consider \(\eta \left( {x,y} \right) = x - y\) for all \(x,y \in \left[ {a,b} \right]\).

Conclusions

The convexity of a function is a basis for many inequalities in mathematics and is applicable for nonlinear programming and optimization theory. It should be noticed that in new problems related to convexity, generalized notions about convex functions are required to obtain applicable results. One of this generalizations may be notion of \(\eta\)-convex functions which can generalizes many inequalities related to convex functions such as the famous Hermite-Hadamard inequality along with estimating the difference between left and middle terms and between right and middle terms of this inequality. Also refinement of Hermite-Hadamard inequality is another application of \(\eta\)-convex functions.

Declarations

Authors’ contributions

All authors contributed equally in this article. All authors read and approved the final manuscript.

Acknowledgements

The authors are very grateful to Prof. S.S. Dragomir for his valuable suggestions about properties of \(\eta\)-convex functions. We are also thankful to the anonymous referees for the useful comments. Author M. De la Sen is grateful to the Basque Government by its support through Grant IT987-16 (internal code 182) and to the Spanish Ministry of Economy and Competitiveness by its support through Grant DPI2015-64766-R including the partial support by FEDER (European Research Funds of Regional Development).

Competing interests

The authors declare that they have no competing interests.

Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Authors’ Affiliations

(1)
Department of Mathematics, Faculty of Basic Sciences, University of Bojnord
(2)
Institute of Research and Development of Processes, University of Basque Country

References

  1. Alomari M, Darus M, Kirmaci US (2010) Refinements of Hadamard-type inequalities for quasi-convex functions with applications to trapezoidal formula and to special means. Comput Math Appl 59:225–232MathSciNetView ArticleMATHGoogle Scholar
  2. Dragomir SS (1992) Two mappings in connection to Hadamards inequalities. J Math Anal Appl 167:49–56MathSciNetView ArticleMATHGoogle Scholar
  3. Dragomir SS, Agarwal RP (1998) Two inequalities for differentiable mappings and applications to special means of real numbers and to trapezoidal formula. Appl Math Lett 11:91–95MathSciNetView ArticleMATHGoogle Scholar
  4. Gordji ME, Dragomir SS, Delavar MR (2015) An inequality related to η-convex functions (II). Int J Nonlinear Anal Appl 6(2):26–32Google Scholar
  5. Gordji ME, Delavar MR, De La Sen M (2016) On φ-convex functions. J Math Inequal 10(1):173–183View ArticleGoogle Scholar
  6. Kirmaci US (2004) Inequalities for differentiable mappings and applications to special means of real numbers and to midpoint formula. Appl Math Comput 147(1):137–146MathSciNetView ArticleMATHGoogle Scholar
  7. Pearce CEM, Pecaric JE (2000) Inequalities for differentiable mappings with application to special means and quadrature formula. Appl Math Lett 13:51–55MathSciNetView ArticleMATHGoogle Scholar
  8. Pecaric JE, Proechan F, Tong YL (1991) Convex functions, partial ordering and statistical applications. Academic Press, New YorkGoogle Scholar
  9. Rostamian Delavar M, Dragomir SS (2016) On η-convexity. Math Inequal Appl (preprint)Google Scholar
  10. Rostamian Delavar M, Dragomir SS, Gordji ME. An inequality related to η-convex functions. U P B Sci ​Bull Ser A​ (to apear)​Google Scholar
  11. Wasowicz S, Witkowski A (2012) On some inequality of Hermite–Hamard type. Opusc Math 32(3):591–600MathSciNetView ArticleGoogle Scholar
  12. Yang GS (2001) Inequalities of Hadamard type for Lipschitzian mappings. J Math Anal Appl 260:230–238MathSciNetView ArticleMATHGoogle Scholar
  13. Yang GS, Hwang DY, Tseng KL (2004) Some inequalities for differentiable convex and concave mappings. Comput Math Appl 47:207–216MathSciNetView ArticleMATHGoogle Scholar

Copyright

© The Author(s) 2016