• No results found

On interval valued invex mappings and optimality conditions for interval valued optimization problems

N/A
N/A
Protected

Academic year: 2020

Share "On interval valued invex mappings and optimality conditions for interval valued optimization problems"

Copied!
19
0
0

Loading.... (view fulltext now)

Full text

(1)

R E S E A R C H

Open Access

On interval-valued invex mappings and

optimality conditions for interval-valued

optimization problems

Lifeng Li

1*

, Sanyang Liu

2

and Jianke Zhang

1

*Correspondence:

[email protected]

1School of Science, Xi’an University

of Posts and Telecommunications, Xi’an, China

Full list of author information is available at the end of the article

Abstract

In this paper, we first introduce the concept of interval-valued invex mappings by using gH-differentiability and compare it with interval-valued weakly invex mappings. We can observe that interval-valued invex mappings are more general than

interval-valued weakly invex mappings. In addition, the sufficient optimality condition for interval-valued objective functions is derived under invexity.

Keywords: interval-valued optimization; interval-valued invex mappings; sufficiency

1 Introduction

Convexity plays a vital role in many aspects of mathematical programming including, for example, sufficient optimality conditions and duality theorems. In inequality constrained optimization, the Kuhn-Tucker conditions are sufficient for optimality if the functions in-volved are convex. However, application of the Kuhn-Tucker conditions as sufficient con-ditions for optimality is not restricted to convex problems, and various generalizations of convexity have been made in order to explore the extent of this applicability.

An invex function, introduced by Hanson [], is one of the generalized convex functions. He considered a differentiable functionf :RnRfor which there exists a vector-valued functionη:Rn×RnRnsuch that, for allx,yRn, the inequality

f(x) –f(y)≥ ∇f(y)(x,y) ()

holds. Hanson [] proved that if, instead of the usual convexity conditions, the objective function and each of the constraints of a nonlinear constrained optimization problem are all invex for the sameη, then both the sufficiency of Kuhn-Tucker conditions and weak and strong Wolfe duality still hold. Later, Craven [] named functions satisfying () invex (with respect toη).

Ben-Israel and Mond [] considered the preinvex functionf with respect toη(not nec-essarily differentiable) for which there exists a vector-valued functionη:Rn×RnRn such that, for allx,yRn, the inequality

fy+λη(x,y)λf(x) + ( –λ)f(y) ()

(2)

holds. Moreover, they found that differentiable functions satisfying () satisfy (). Further properties and applications of preinvexity and its some generalizations for some more general problems were studied by Antczak [, ], Bectoret al.[], Mohan and Neogy [], Sunejaet al.[], and others.

However, the majority of real world optimization problems often involve data uncer-tainty or imprecision owing to measurement errors or some unexpected things. Interval-valued optimization [] is an important model to deal with the problems with data uncer-tainty. Many approaches to interval-valued optimization problems have been explored in considerable details (see, for example, [–]). Recently, Wu has extended the concept of convexity for a real-valued function to LU-convexity for an interval-valued function, then he has established the Kuhn-Tucker conditions [, ] for an optimization problem with an interval-valued objective function under the assumption of LU-convexity. In [], Wu studied the Kuhn-Tucker optimality conditions in multiobjective programming problems with an interval-valued objective function. Similar to the concept of non-dominated solu-tion in vector optimizasolu-tion problems, Wu has proposed a solusolu-tion concept in optimizasolu-tion problems with an interval-valued objective function based on a partial ordering on the set of all closed intervals. Then, the interval-valued Wolfe duality theory [] and Lagrangian duality theory [] for interval-valued optimization problems have been proposed. Wu [] studied the duality theory for interval-valued linear programming problems. Chalco-Canoet al.[] gave Kuhn-Tucker type optimality conditions, which are obtained using gH-derivative of interval-valued functions. Also, they discussed the relationship between the approach presented with other well-known approaches given by Wu []. However, these methods given by Chalco-Canoet al.[] cannot solve a kind of optimization prob-lems with interval-valued objective functions, which are not LU-convex but invex. For example, the interval-valued functions such asf(x) = [x– sinx,x–sinx+ ] are not

LU-convex but invex with respect to

η(x,y) =

sinx–siny

cosy

,sinx–siny cosy

t

(the concept of invex can be seen in Definition ). Zhanget al.[] proposed the Kuhn-Tucker optimality conditions for an optimization problem with an interval-valued objec-tive function under the assumptions of preinvexity and weak invexity. The definition of interval-valued invexity in this paper is more general than that of weak invexity given in [] (see Theorem  and Example ).

This paper aims at extending the Kuhn-Tucker optimality conditions to nonconvex opti-mization problem. First, we extend the concept of invexity using gH-derivative of interval-valued functions. The concept of invexity by using gH-differentiability of interval-interval-valued functions is more general than the concept of invexity by using weak differentiability (see Theorem  and Example ). Second, we present several properties of invex interval-valued functions. Finally, the Kuhn-Tucker optimality conditions are proposed for an interval-valued objective function under the assumptions of invexity.

2 Preliminaries

(3)

Definition  LetA= [aL,aU] andB= [bL,bU] be inI. We define (i) A+B={a+b:aAandbB}= [aL+bL,aU+bU]; (ii) –A={–a:aA}= [–aU, –aL];

(iii) A×B={ab:aAandbB}= [minab,maxab], where

minab=min{aLbL,aLbU,aUbL,aUbU}andmaxab=max{aLbL,aLbU,aUbL,aUbU}.

Then it is easy to conclude that

AB=A+ (–B) =aLbU,aUbL,

kA={ka:aA}=

[kaL,kaU] ifk,

|k|[–aU, –aL] ifk< ,

()

wherekis a real number.

Hausdorff metricbetween two closed intervalsAandBdefined as

dH(A,B) =max aLbL,aUbU.

Definition  LetA= [aL,aU] andB= [bL,bU] inI. We writeABifaLbLandaUbU, ABifABandA=B,i.e., the following (a) or (a), or (a) is satisfied.

(a) aL<bLandaUbU; (a) aLbLandaU<bU; (a) aL<bLandaU<bU.

LetA,BI, if there existsCI such thatA=B+C, thenCis called the Hukuhara difference ofAandBand written asC=AB; when we say that the H-differenceC exists, it means thataLbLaUbUandC= [aLbL,aUbU].

Proposition  Let A= [aL,aU]and B= [bL,bU]be two closed intervals inI.If aLbL aUbU,then the H-difference C exists and C= [aLbL,aUbU].

It follows from Proposition  that the H-difference is unique, but it does not always exist. To address this issue, a generalization of the Hukuhara difference is proposed in [].

Definition ([]) LetA= [aL,aU] andB= [bL,bU] be two closed intervals, the gH-differ-ence ofAandBis defined by

aL,aUg

bL,bU=minaLbL,aUbU,maxaLbL,aUbU.

For example, [, ]g[, ] = [, ], [, ]g[, ] = [–, ]. Andab= [a,a]g[b,b] = [a–b,ab] =ab.

Proposition ([])

(i) For every pairA,BI,AgBalways exists andAgBI. (ii) AgBif and only ifAB.

(4)

written asf(x) = [fL(x),fU(x)], wherefLandfUare two real-valued functions defined on Rnand satisfyfL(x)fU(x) for everyxRn. Based on the above concept, Wu [] has introduced the concepts of limit, continuity and two kinds of differentiation of interval-valued functions.

Letf be an interval-valued function defined onRnandA= [aL,aU] be an interval inR,

cRn. If for every>  there existsδ>  such that, for  <x–c<δ, we havedH(f(x),A) < , then

lim

xcf(x) =A.

Proposition ([]) Let f be an interval-valued function defined on Rnand A= [aL,aU]be an interval in R.Thenlimxcf(x) =A if and only iflimxcfL(x) =aLandlimxcfU(x) =aU.

Proposition ([]) Let f be an interval-valued function defined on Rn.Then f is contin-uous at cRnif and only if both fLand fUare continuous at c.

Proposition ([]) Let X be an open set in R.An interval-valued function f :XIwith f(x) = [fL(x),fU(x)]is calledweakly differentiableat x

if the real-valued functions fLand

fUare differentiable at x(in the usual sense).

Definition ([]) LetXbe an open set inR. An interval-valued functionf :XIis called H-differentiable atxif there exists a closed intervalA(x)∈Isuch that the limits

lim h→+

f(x+h)f(x)

h

and

lim h→+

f(x)f(x–h)

h

both exist and equalA(x). In this case,A(x) is called the H-derivative off atx.

The following concept is particularization of the fuzzy concepts presented in [] to the interval case. These are defined by using the usual Hukuhara difference.

Definition ([]) LetT= (a,b) and lett∈T. Givenf :TI, we say thatfis strongly

generalized differentiable (G-differentiable) attif there exists an elementf(t)∈Isuch

that for allh>  sufficiently small,

(i) ∃f(x+h)f(x),f(x)f(x–h)and

lim h→

f(x+h)f(x)

h =hlim→

f(x)f(x–h)

h =f

(x

),

or

(ii) ∃f(x)f(x+h),f(x–h)f(x)and

lim h→

f(x)f(x+h)

–h =hlim→

f(x–h)f(x)

–h =f

(x

(5)

or

(iii) ∃f(x+h)f(x),f(x–h)f(x)and

lim h→

f(x+h)f(x)

h =hlim→

f(x–h)f(x)

–h =f

(x

),

or

(iv) ∃f(x)f(x+h),f(x)f(x–h)and

lim h→

f(x)f(x+h)

–h =hlim→

f(x)f(x–h)

h =f

(x

).

Based on the gH-difference, Stefanini [] proposed the following differentiation.

Definition ([]) Letx∈(a,b) andhbe such thatx+h∈(a,b), then the gH-derivative

of a functionf: (a,b)Iatxis defined as

f(x) =lim

h→

f(x+h)gf(x)

h . ()

If f(x)∈I satisfying () exists, we say that f is generalized Hukuhara differentiable

(gH-differentiable for short) atx.

The next two results express the gH-derivative in terms of the endpoints of the interval-valued function.

Theorem ([]) Let f : (a,b)Ibe such that f(x) = [fL(x),fU(x)].If fL(x)and fU(x)are differentiable functions at t∈(a,b),then f(x)is gH-differentiable at t,and

f(x) =min fL(x),fU(x),max fL(x),fU(x).

Theorem ([]) Let f: (a,b)Ibe such that f(x) = [fL(x),fU(x)].Then f(x)is gH-dif-ferentiable at t∈(a,b)if and only if one of the following cases holds:

(a) fL(x)andfU(x)are differentiable att;

(b) the lateral derivatives(fL)(t),(fL)+(t)and(fU)(t),(fU)+(t)exist and satisfy (fL)(t) = (fU)+(t)and(fL)+(t) = (fU)(t).

Letf be an interval-valued function defined onXRn, comparing above definitions, the following statements hold.

Proposition 

(i) Iff is H-differentiable atx∈X,then it is G-differentiable atx,the converse is not

true.

(ii) Iff is G-differentiable atx∈X,then it is gH-differentiable atx,the converse is not

true.

(iii) Iff is weakly differentiable atx,then it is gH-differentiable atx,the converse is not

true.

Definition ([]) Letf(x) be an interval-valued function defined on, whereis an open subset ofRn. LetDx

(6)

to theith variablexi. Assume thatfL(x) andfU(x) have continuous partial derivatives so thatDxifL(x) andDxifU(x) are continuous. Fori= , , . . . ,n, define

Dxif(x) =

minDxif

L(x),Dx

if

U(x),maxDx

if

L(x),Dx

if

U(x),

we will say thatf(x) is differentiable atx, and we write

f(x) =Dxf(x),Dxf(x), . . . ,Dxnf(x)

t .

We call∇f(x) the gradient of the interval-valued function atx.

Example  Letf :RI be defined byf(x) = [x

+x, x+ x+ ]. SofL(x) =x+x

andfU(x) = x+ x+ .Dxf

L(x) = x

,Dxf

L(x) = x

,Dxf

U(x) = x

,Dxf

U(x) = x

.

Thus,

Dxf(x) =

[x, x] ifx≥,

[x, x] ifx< ,

()

Dxf(x) =

[x, x] ifx≥,

[x, x] ifx< .

()

Thus,

f(x) = ⎧ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎩

([x, x], [x, x])t ifx≥,x≥,

([x, x], [x, x])t ifx≥,x< ,

([x, x], [x, x])t ifx< ,x≥,

([x, x], [x, x])t ifx< ,x< .

()

Further,

Lf(x) = ⎧ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎩

(x, x)t ifx≥,x≥,

(x, x)t ifx≥,x< ,

(x, x)t ifx< ,x≥,

(x, x)t ifx< ,x< ,

()

Uf(x) = ⎧ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎩

(x, x)t ifx≥,x≥,

(x, x)t ifx≥,x< ,

(x, x)t ifx< ,x≥,

(x, x)t ifx< ,x< .

()

3 Preinvexity and invexity of interval-valued functions

(7)

In what follows, we show the connection between preinvex and invex interval-valued mappings. Here, we recall the definition of preinvex interval-valued mappings.

Definition  LetyXRn. Then we say thatXis invex atywith respect toη:X×X Rnif for eachxX,λ[, ],y+λη(x,y)X.Xis said to be an invex set with respect to ηifXis invex at eachyX.

Definition ([]) LetKRnbe an invex set with respect toη:K×KRn,f(x) = [fL(x),fU(x)] be an interval-valued function defined onK. We say thatf ispreinvexatx∗ with respect toηif

fx+ληx∗,xλfx∗+ ( –λ)f(x)

for eachλ∈[, ] and eachxK.

Theorem  Let K be an invex subset of Rnwith respect toη:K×KRnand f be an interval-valued function defined on K.Then f is preinvex at xif and only if fLand fUare preinvex at xwith respect to the sameη:K×KRn,i.e.,

fLx+ληx∗,xλfLx∗+ ( –λ)fL(x), ()

fUx+ληx∗,xλfUx∗+ ( –λ)fU(x) ()

for eachλ∈[, ]and each xK.

Definition ([]) LetKRnbe an invex set with respect toη:K×KRn,f(x) = [fL(x),fU(x)] be an interval-valued function defined onK. We say thatf isinvexatxif

the real-valued functionsfLandfUare invex atx,i.e.,

fL(x) –fLx∗≥ηx,x∗t∇fLx∗, ()

fU(x) –fUx∗≥ηx,x∗t∇fUx∗ ()

for eachxK.

Remark  Since the definition of interval-valued invex functions defined in [] consid-ered the end-point functions, we call them weakly invex functions in this paper.

Based on the gH-differentiability, we give the definition of interval-valued invex func-tions as follows.

Definition  A gH-differentiable interval-valued mappingf :XIis said to be invex on the invex setXRnwith respect toηif for anyx,yX,λ∈[, ],

f(x)gf

xηx,x∗t∇fx∗ ()

(8)

Example  Consider the interval-valued mapping f(x) = [, ]x, xR. Then f(x) is

gH-differentiable onRby Theorem , and

f(x) =

[x, x], x≥, [x, x], x< .

Letη(x,y) =xy, thus

η(x,y)tf(y) = ⎧ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎩

[y(x–y), y(xy)], y≥,xy≥, [y(x–y), y(xy)], y< ,xy< , [y(x–y), y(xy)], y≥,xy< , [y(x–y), y(xy)], y< ,xy≥,

()

f(x)gf(y) =

[xy, x– y], xy,

[x– y,xy], x<y. ()

We can observe thatf(x) is invex with respect toη(x,y) =xyby Definition .

The following theorem shows the relationship between interval-valued invex functions and weakly invex functions.

Theorem  Let K be an invex subset of Rn with respect toη:K×KRnand f(x) = [fL(x),fU(x)]be an interval-valued function defined on K.If f is weakly invex,then it is invex,but the converse is not true in general.

Proof Sincef is weakly invex atx∗, we have that real-valued functionsfLandfUare invex atx∗,i.e.,

fL(x) –fLx∗≥ηx,x∗t∇fLx∗, ()

fU(x) –fUx∗≥ηx,x∗t∇fUx∗ ()

for eachλ∈[, ] and eachxK.

() On the condition ofη(x,x∗)tfL(x∗)≤η(x,x∗)tfU(x∗), we have

ηx,x∗t∇fx∗=ηx,x∗t∇fLx∗,ηx,x∗t∇fUx∗.

Iff(x)gf(x∗) = [fL(x) –fL(x∗),fU(x) –fU(x∗)], then from () and () we have

f(x)gf

xηx,x∗t∇fx∗.

Iff(x)gf(x∗) = [fU(x) –fU(x∗),fL(x) –fL(x∗)], then

fL(x) –fLx∗≥fU(x) –fUx∗≥ηx,x∗t∇fUx∗≥ηx,x∗t∇fLx∗.

We have

f(x)gf

(9)

() On the condition ofη(x,x∗)tfL(x) >η(x,x)tfU(x), we have

ηx,x∗t∇fx∗=ηx,x∗t∇fUx∗,ηx,x∗t∇fLx∗.

Iff(x)gf(x∗) = [fU(x) –fU(x∗),fL(x) –fL(x∗)], then from () and () we have

f(x)gf

xηx,x∗t∇fx∗.

Iff(x)gf(x∗) = [fL(x) –fL(x∗),fU(x) –fU(x∗)], then

fU(x) –fUx∗≥fL(x) –fLx∗≥ηx,x∗t∇fLx∗≥ηx,x∗t∇fUx∗.

Thus

f(x)gf

xηx,x∗t∇fx∗.

Example  Considering the interval-valued functionf(x) = [x– sinx,x–sinx+ ],

xR, we can prove that bothfL(x) andfU(x) are weakly invex with respect to

η(x,y) =

sinx–siny

cosy

,sinx–siny cosy

t .

Thenf(x) = [x– sinx,x–sinx+ ],xRis invex with respect to the sameη(x,y) by

Theorem .

Example  Considering the interval-valued functionf(x) = [–|x|,|x|],xR,

η(x,y) =

xy, xy≥, x+y, xy< .

From Theorem , it follows thatf(x) is gH-differentiable onR, andf(y) = [–, ], thus

η(x,y)tf(y) = ⎧ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎩

[–(x–y),xy], xy≥ oryx≤, [x–y, –(xy)], yx≥ orxy≤, [–(x+y),x+y], xy<  andx+y> , [x+y, –(x+y)], xy<  andx+y< ,

f(x)gf(y) = ⎧ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎨ ⎪ ⎪ ⎪ ⎪ ⎪ ⎪ ⎩

[–(x–y),xy], xy≥ oryx≤, [x–y, –(xy)], yx≥ orxy≤, [–(x+y),x+y], x>  >yandx+y> , [x+y, –(x+y)], x>  >yandx+y< , [–(x+y),x+y], x<  <y.

Then

f(x)gf(y)η(x,y)f(y)t.

(10)

Letx= ,y= –,η(x,y) = ,

fLy+λη(x,y)=  –λ> – +λ=λfL(y) + ( –λ)fL(x).

Thus,f(x) is not preinvex sincefL(x) is not preinvex with respect toη.

The following theorem given in [] illustrates the relations between weakly invex interval-valued and preinvex interval-valued functions.

Theorem  Let KRn be an invex set with respect to η: K × K Rn; if f(x) = [fL(x),fU(x)]is a weakly continuously differentiable and preinvex interval-valued func-tion defined on K,then f is also a weakly invex interval-valued function with respect to the sameηdefined on K.

We can prove the following result.

Theorem  Let KRn be an invex set with respect to η: K × KRn. If f(x) = [fL(x),fU(x)]is a weakly differentiable and preinvex interval-valued function defined on K,then f is also an interval-valued invex function with respect to the sameηdefined on K.

Proof Sincef is interval-valued preinvex and weakly differentiable, we have

fLy+λη(x,y)fL(y)≤λfL(x) –fL(y),

fUy+λη(x,y)fU(y)≤λfU(x) –fU(y),

which forλ∈(, ] implies

fL(y+λη(x,y)) –fL(y)

λf

L(x) –fL(y),

fU(y+λη(x,y)) –fU(y)

λf

U(x) –fU(y).

By taking limits forλ→+, sincef is weakly differentiable, we get

η(x,y)tfL(y)≤fL(x) –fL(y), ()

η(x,y)tfU(y)≤fU(x) –fU(y). ()

On the other hand, from Definition  and Theorem , we have

f(x)gf(y) =

[fL(x) –fL(y),fU(x) –fU(y)], fL(x) –fL(y)≤fU(x) –fU(y),

[fU(x) –fU(y),fL(x) –fL(y)], fL(x) –fL(y) >fU(x) –fU(y) ()

and

η(x,y)tf(y)

=η(x,y)tmin fL(y),fU(y),max fL(y),fU(y) ()

=

[η(x,y)tfL(y),η(x,y)tfU(y)], η(x,y)tfL(y)η(x,y)tfU(y),

(11)

From ()-() it follows that

f(x)gf(y)η(x,y)tf(y).

Thus,f is an invex interval-valued function.

The following example given in [] shows that a weakly invex interval-valued function may not be a preinvex interval-valued function, from Theorem  we can conclude that the converse of Theorem  is not true.

Example  The interval-valued functionf(x) = [, ]·ex,xRis invex with respect to η= –, but not preinvex with respect to the same functionη.

However, Mohan and Neogy [] have proved that a differentiable invex real-valued func-tion is also preinvex under the following condifunc-tion.

Condition C We say that the functionη:RnRnsatisfies Condition C if for anyx,yX,

ηx,y+λη(x,y)= –λη(x,y), ηy,y+λη(x,y)= ( –λ)η(x,y).

We can conclude that a continuously weakly differentiable invex interval-valued func-tionf :KIis also a preinvex interval-valued function onKif the functionηsatisfies Condition C.

Theorem  Suppose that K is an invex set of Rn with respect toη:K×KRnand f :KI is a continuously weakly differentiable interval-valued function on an open set containing K. If f is invex on K with respect toηandηsatisfies ConditionC,then f is preinvex with respect toηon K.

Proof Suppose thatx,x∈X. Let  <λ<  be given and look atx=x+λη(x,x).

Note thatxX, by invexity of˜f, we have

f(x)gf(x)η(x,x)tf(x)

and

f(x)gf(x)η(x,x)tf(x),

i.e.,

min fL(x) –fL(x),fU(x) –fU(x)

,max fL(x) –fL(x),fU(x) –fU(x)

η(x,x)t

min ∇fL(x),∇fU(x),max ∇fL(x),∇fU(x)

and

min fL(x) –fL(x),fU(x) –fU(x)

,max fL(x) –fL(x),fU(x) –fU(x)

η(x,x)t

(12)

() On the condition offL(x

) –fL(x)≤fU(x) –fU(x),η(x,x)tfL(x)≤η(x,x)tfU(x),

fL(x

) –fL(x)≤fU(x) –fU(x), andη(x,x)tfL(x)≤η(x,x)tfU(x), we have

f(x)gf(x) =

fL(x) –fL(x),fU(x) –fU(x)

,

f(x)gf(x) =

fL(x) –fL(x),fU(x) –fU(x)

and

η(x,x)tf(x) =

η(x,x)tfL(x),η(x,x)tfU(x)

,

η(x,x)tf(x) =

η(x,x)tfL(x),η(x,x)tfU(x)

.

Thus,

fL(x) –fL(x)≥η(x,x)tfL(x),

fU(x) –fU(x)≥η(x,x)tfU(x)

and

fL(x) –fL(x)≥η(x,x)tfL(x),

fU(x) –fU(x)≥η(x,x)tfU(x).

Therefore,

λfL(x) + ( –λ)fL(x) –fL(x)≥

λη(x,x) + ( –λ)η(x,x)

t

fL(x).

However, by Condition C,λη(x,x) + ( –λ)η(x,x) = . Hence,

fLx+λη(x,x)

λfL(x) + ( –λ)fL(x).

By a similar way,

fUx+λη(x,x)

λfU(x) + ( –λ)fU(x).

Thusf is preinvex with respect toηby Theorem . () On the condition offL(x

) –fL(x)≤fU(x) –fU(x),η(x,x)tfL(x)≤η(x,x)tfU(x),

fL(x

) –fL(x)≤fU(x) –fU(x), andη(x,x)tfL(x) >η(x,x)tfU(x), we have

f(x)gf(x) =

fL(x) –fL(x),fU(x) –fU(x)

,

f(x)gf(x) =

fL(x) –fL(x),fU(x) –fU(x)

and

η(x,x)tf(x) =

η(x,x)tfL(x),η(x,x)tfU(x)

,

η(x,x)tf(x) =

η(x,x)tfU(x),η(xx)tfL(x)

(13)

Thus,

fL(x) –fL(x)≥η(x,x)tfL(x),

fU(x) –fU(x)≥η(x,x)tfU(x)

and

fL(x) –fL(x)≥η(x,x)tfU(x),

fU(x) –fU(x)≥η(x,x)tfL(x).

Supposemin{η(x,x)tfL(x),η(x,x)tfU(x)}=η(x,x)tfL(x). Therefore,

λfL(x) + ( –λ)fL(x) –fL(x)≥λη(x,x)tfL(x) + ( –λ)η(x,x)tfU(x)

λη(x,x)tfL(x) + ( –λ)η(x,x)tfL(x)

=λη(x,x) + ( –λ)η(x,x)

t

fL(x).

However, by Condition C,λη(x,x) + ( –λ)η(x,x) = . Hence,

fLx+λη(x,x)

λfL(x) + ( –λ)fL(x).

By a similar way,

fUx+λη(x,x)

λfU(x) + ( –λ)fU(x).

Thusf is preinvex with respect toηby Theorem .

() On the condition offL(x) –fL(x)≤fU(x) –fU(x),η(x,x)tfL(x)≤η(x,x)tfU(x),

fL(x) –fL(x) >fU(x) –fU(x), andη(x,x)tfL(x)≤η(x,x)tfU(x), we have

f(x)gf(x) =

fL(x) –fL(x),fU(x) –fU(x)

,

f(x)gf(x) =

fU(x) –fU(x),fL(x) –fL(x)

and

η(x,x)tf(x) =

η(x,x)tfL(x),η(x,x)tfU(x)

,

η(x,x)tf(x) =

η(x,x)tfL(x),η(x,x)tfU(x)

.

Thus,

fL(x) –fL(x)≥η(x,x)tfL(x),

fU(x) –fU(x)≥η(x,x)tfU(x)

and

fL(x) –fL(x)≥η(x,x)tfU(x),

(14)

Supposemin{η(x,x)tfL(x),η(x,x)tfU(x)}=η(x,x)tfL(x). Thus,

λfL(x) + ( –λ)fL(x) –fL(x)≥λη(x,x)tfU(x) + ( –λ)η(x,x)tfL(x)

λη(x,x)tfL(x) + ( –λ)η(x,x)tfL(x)

=λη(x,x) + ( –λ)η(x,x)

t

fL(x).

However, by Condition C,λη(x,x) + ( –λ)η(x,x) = . Hence,

fLx+λη(x,x)

λfL(x) + ( –λ)fL(x).

By a similar way,

fUx+λη(x,x)

λfU(x) + ( –λ)fU(x).

Thusf is preinvex with respect toηby Theorem .

() On the condition offL(x) –fL(x)≤fU(x) –fU(x),η(x,x)tfL(x)≤η(x,x)tfU(x),

fL(x

) –fL(x) >fU(x) –fU(x), andη(x,x)tfL(x) >η(x,x)tfU(x), we have

f(x)gf(x) =

fL(x) –fL(x),fU(x) –fU(x)

,

f(x)gf(x) =

fU(x) –fU(x),fL(x) –fL(x)

and

η(x,x)tf(x) =

η(x,x)tfL(x),η(x,x)tfU(x)

,

η(x,x)tf(x) =

η(x,x)tfU(x),η(x,x)tfL(x)

.

Thus,

fL(x) –fL(x)≥η(x,x)tfL(x),

fU(x) –fU(x)≥η(x,x)tfU(x)

and

fL(x) –fL(x)≥η(x,x)tfL(x),

fU(x) –fU(x)≥η(x,x)tfU(x).

Thus,

λfL(x) + ( –λ)fL(x) –fL(x)≥

λη(x,x) + ( –λ)η(x,x)

t

fL(x).

However, by Condition C,λη(x,x) + ( –λ)η(x,x) = . Hence,

fLx+λη(x,x)

(15)

By a similar way,

fUx+λη(x,x)

λfU(x) + ( –λ)fU(x).

As a consequence,f is preinvex with respect toηby Theorem .

We can prove the result on the other four conditions by a similar way.

4 The Kuhn-Tucker optimality conditions with interval-valued objective functions

An interval-valued objective minimization problem is

(IVOP) minf(x) =fL(x),fU(x)

subject togi(x)≤, i= , , . . . ,m.

LetP={xRn:g

i(x)≤,i= , , . . . ,m}be a feasible set of (IVOP).

Definition  Letx∗be a feasible solution of the primal problem (IVOP). We say thatx∗ is a non-dominated solution of problem (IVOP) if and only if there exists noxPsuch thatf(x)≺f(x∗). In this case,f(x∗) is called the non-dominated objective value off.

Theorem  Let f(x)be gH-differentiable on XRnand interval-valued invex with respect toη:X×XRn,and gi(x) (i= , , . . . ,m)be invex with respect to the sameη.If there exist x∗∈P and vi(i= , , . . . ,m)such that

fx∗L+ n

i=

vigi

x∗= , ()

n

i=

vigix∗= , ()

vi≥, ()

then xis a non-dominated solution of problem(IVOP).

Proof For anyx∗∈P satisfyinggi(x∗)≤,i= , , . . . ,m, sincef(x) is gH-differentiable onXRnand interval-valued invex with respect toη, then [min{fL(x) –fL(x),fU(x) – fU(x)},max{fL(x) –fL(x),fU(x) –fU(x)}]η(x,x)tf(x).

(i) On the condition ofmin{fL(x) –fL(x),fU(x) –fU(x)}=fL(x) –fL(x),η(x,x)t×

f(x∗) = [η(x,x∗)t{∇f(x∗)}L,η(x,x∗)t{∇f(x∗)}U],

fL(x) –fLx∗≥ηx,x∗t ∇fx∗L

= –ηx,x∗t n

i=

vigi

x

≥– m

i=

(16)

= m

i=

–vigi(x) +vigix

= m

i=

–vigi(x)

≥.

Thus,f(x)≺f(x∗) does not hold.

(ii) On the condition ofmin{fL(x) –fL(x∗),fU(x) –fU(x∗)}=fU(x) –fU(x∗),η(x,x∗)t×

f(x∗) = [η(x,x∗)t{∇f(x)}L,η(x,x)t{∇f(x)}U],

fU(x) –fUx∗≥ηx,x∗t ∇fx∗L

= –ηx,x∗t n

i=

vigix

≥– m

i=

vigi(x) –gix

= m

i=

–vigi(x) +vigix

= m

i=

–vigi(x)

≥.

Thus,f(x)≺f(x∗) does not hold.

Similarly, we can prove the other two conditions.

The proof of the following theorem is similar to the one of Theorem .

Theorem  Let f(x)be gH-differentiable on XRnand interval-valued invex with respect toη:X×XRn,and g

i(x) (i= , , . . . ,m)be invex with respect to the sameη.If there exist x∗∈P and vi(i= , , . . . ,m)such that

fx∗U+ n

i=

vigix∗= , ()

n

i=

vigi

x∗= , ()

vi≥, ()

then xis a non-dominated solution of problem(IVOP).

Example 

minimizef(x) =x+x+ ,x+ 

subject tog(x) =x– ≤,

(17)

From Theorem  we have∇f(x) = [x, x+ ].

(i) It is easy to see that the problem satisfies the assumptions of Theorem . Then

x∗, x∗+ L+v–v= ,

v

x∗– = ,

v

–x∗– = .

We obtainv=v= , andx∗=  is a non-dominated solution.

(ii) It is easy to see that the problem satisfies the assumptions of Theorem . Then

x∗, x∗+ U+v–v= ,

v

x∗– = ,

v

–x∗– = .

We obtainv=v= , andx∗= –is also a non-dominated solution.

Example  Consider the following interval-valued programming problem:

minimizef(x) = [, ]sinx

subject tog(x) = (sinx– )– ≤,

x

,π

.

Note that functionsf andgare invex with respect to

η(x,y) =sinx–siny cosy .

And∇f(x) = [sinxcosx, sinxcosx],∇g(x) = (sinx– )cosx.

It is easy to see that the problem satisfies the assumptions of Theorem . Then

sinx∗cosx∗+ vsinx∗– cosx∗= ,

vsinx∗– –  

= .

After some algebraic calculations, we obtainx∗=sin–( – 

√),v=  √

 – . Therefore, x∗=sin–( – 

√) is a non-dominated solution.

The following example also shows the advantages of our method in respect to [].

Example  Consider the following interval-valued programming problem:

minimizef(x) = [x– sinx,x–sinx+ ]

(18)

g(x) = x+ x– ≤,

g(x) = –sinx≤,

g(x) = –sinx≤.

Then f is gH-differentiable and weakly differentiable. SincefLis not LU-convex and fL+fUis not LU-convex, methods in [] cannot be used.

Note that functionsf andgi(i= , , , ) are invex with respect to

η(x,y) =

sinx–siny

cosy

,sinx–siny cosy

t .

It is easy to see that the problem satisfies the assumptions of Theorem . Then

 +v(cosx+ ) + v–vcosx= ,

–cosx+ vcosx+ v–vcosx= ,

v(sinx+ sinx+x– ) = ,

v(x+ x– ) = ,

v(–sinx) = ,

v(–sinx) = .

After some algebraic calculations, we obtainx∗= (,sin– 

)

t,v= ( , ,

 , )

t. Therefore,

x∗is a non-dominated solution.

5 Conclusion

The concept of convex interval-valued mappings has been studied in the literature by many researchers. The aim of this paper is to introduce the concept of invex interval-valued mappings with gH-differentiable functions. Then we discussed the relationships between interval-valued invex mappings and interval-valued weakly invex mappings. Fi-nally, the sufficient optimality condition for interval-valued objective functions has been derived under invexity.

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

All authors have equal contributions. All authors read and approved the final manuscript.

Author details

1School of Science, Xi’an University of Posts and Telecommunications, Xi’an, China.2Department of Mathematics, School

of Science, Xidian University, Xi’an, China.

Acknowledgements

This work was partially supported by the National Natural Science Foundation of China (Grant Nos. 60974082, 11301415, 11401469), Foundation from Xi’an University of Posts and Telecommunications for Young Teachers (Grant No. ZL2014-34) and the Natural Science Foundation of Shaanxi Province (Grant No. 2013JQ1020).

(19)

References

1. Hanson, MA: On sufficiency of the Kuhn-Tucker conditions. J. Math. Anal. Appl.80, 545-550 (1981) 2. Craven, BD: Invex functions and constrained local minima. Bull. Aust. Math. Soc.24, 357-366 (1981) 3. Ben-Israel, A, Mond, B: What is invexity? J. Aust. Math. Soc. Ser. B, Appl. Math28, 1-9 (1986) 4. Antczak, T: (p,r)-Invex sets and functions. J. Math. Anal. Appl.263, 355-379 (2001) 5. Antczak, T: Relationships between pre-invex concepts. Nonlinear Anal.60, 349-367 (2005)

6. Bector, CR, Suneja, SK, Lalitha, CS: Generalized B-vex functions and generalized B-vex programming. J. Optim. Theory Appl.76(3), 561-576 (1993)

7. Mohan, SR, Neogy, SK: On invex sets and preinvex functions. J. Math. Anal. Appl.189, 901-908 (1995)

8. Suneja, SK, Singh, C, Bector, CR: Generalizations of pre-invex functions and B-vex functions. J. Optim. Theory Appl.76, 577-587 (1993)

9. Ben-Israel, A, Robers, PD: A decomposition method for interval linear programming. Manag. Sci.16, 374-387 (1970) 10. Chanas, S, Kuchta, D: Multiobjective programming in optimization of interval objective functions - a generalized

approach. Eur. J. Oper. Res.94, 594-598 (1996)

11. Liu, ST: Polynomial geometric programming with interval exponents and coefficients. Eur. J. Oper. Res.186, 7-27 (2008)

12. Jiang, C, Han, X, Liu, GP: A nonlinear interval number programming method for uncertain optimization problems. Eur. J. Oper. Res.188, 1-13 (2008)

13. Wu, HC: The Karush-Kuhn-Tucker optimality conditions in an optimization problem with interval-valued objective function. Eur. J. Oper. Res.176, 46-59 (2007)

14. Wu, HC: On interval-valued nonlinear programming problems. J. Math. Anal. Appl.338, 299-316 (2008) 15. Wu, HC: The Karush-Kuhn-Tucker optimality conditions in multiobjective programming problems with

interval-valued objective function. Eur. J. Oper. Res.196, 49-60 (2009)

16. Wu, HC: Wolfe duality for interval-valued optimization. J. Optim. Theory Appl.138, 497-509 (2008)

17. Wu, HC: Duality theory for optimization problems with interval-valued objective functions. J. Optim. Theory Appl. 144, 615-628 (2010)

18. Wu, HC: Duality theory in interval-valued linear programming problems. J. Optim. Theory Appl.150, 298-316 (2011) 19. Chalco-Cano, Y, Lodwick, WA, Rufian-Lizana, A: Optimality conditions of type KKT for optimization problem with

interval-valued objective function via generalized derivative. Fuzzy Optim. Decis. Mak.12, 305-322 (2013) 20. Zhang, J, Liu, S, Li, L, Feng, Q: The KKT optimality conditions in a class of generalized convex optimization problems

with an interval-valued objective function. Optim. Lett.8, 607-631 (2014)

21. Stefanini, L: Generalized Hukuhara differentiability of interval-valued functions and interval differential equations. Nonlinear Anal.71, 1311-1328 (2009)

22. Bede, B, Gal, SG: Generalizations of the differentiability of fuzzy number valued functions with applications to fuzzy differential equation. Fuzzy Sets Syst.151, 581-599 (2005)

23. Chalco-Cano, Y, Roman-Flores, H, Jimenez-Gamero, MD: Generalized derivative and pi-derivative for set-valued functions. Inf. Sci.181, 2177-2188 (2011)

24. Li, L, Liu, S, Zhang, J: Univex interval-valued mapping with differentiability and its application in nonlinear programming. J. Appl. Math.2013, Article ID 383692 (2013)

References

Related documents