(1)1

1991 - 8690

2008/
ISSN 1991- 8690

thiqaruni.org
-

:

,

.
.

.

:


,

.

)Linear
( Programming

( (Goal Programming


(Adam,1982),(Holzman,1981) .

.

,

).
(Srivastava,1991

) (

. (Kothari,1992) .


,



. (Philips ,1987) .



.

-1
:
-1.1:



.
) Azarm , ( , (,1995
( 1996 ) , (Ignizio ,1976

2008/

(1)1

,

d d
) (


d=d+ -d : +d .
: -d ) .
(Markland,1983

:

G1:7x1+10x2=1000
G2:x1= 10
G3:x2=10
S.t.x1+2x2<=60
x1,x2 0

:
7x1+10x2

MinZ=d1+d2+d3
+d1=1000
10:

=x1 +d2

x2 +d3=10
x1+2x2<=60
x1,x2 0 , d1,d2,d3 , free

:
-1.2 :
) (Karim,2000),(Ignizio ,1976
-1
.
-2
.
-3
.
-4

M )Big-
(M .

:
1

=MinZ
7x1+10x2 + d1+-d1-=1000
x1 + d2+-d2-= 10:
x2 + d3+-d3-=10
x1+2x2<=60
d1+,d1-,d2+,d2-,d3+,d3-, 0
x1,x2

:
=d1=580,d250,d3=10,x1=60,MINZ=540

-2 MPS
:
2.1 :



.

(1)1

2008/

: C

(xNB = (xm+1,, xn
n

: B
xB

mxm
)(non singular

.
Gass,1985 ( , ) 2005 ,
)
2.2
:
)(Gass ,1985),(Taha ,2000
) (1.1

:

: A
xNB .
: b
.
:
= CBB-1

) (1.1
:
Z CB XB C XNB = 0
B XB + A XNB = b

:
k

: gi .
i =1
n

a x
ij

j =1

: .
i =1

MPS

) (2.1 ,
)) 1.1 :
: CB

(xB= (d1,,dk,dk+1,,xm
m

XNB = 0
... Z CB XB = 0
B XB = b

2008/

(1)1

b
B-1b

XB
0
I

XNB
A-C
B-1A

(XB= (d1,,dk,Xk+1,,Xm
(XNB = (Xm+1, ,Xn

) (2.1

= CB B-1
=

b
B

Xm+1,
.,X
n
bB-1
Ab- C

Xk+1,
.,Xm
0

d1,

dk
0

B-1 A

-1

Idxd

) (1 :
--------I

( (m-d)x(m-d

--------0

Z
d1
.
.
.
dk
----Xk+1

.
.
.
Xm

=CBB-1

2
.
3
MPS
-1
.
- 2
.

 (1)1

-3 ) ( n
)) m )(d
A,b,B,C,CB . B-1
-4 Z

=CBB

-1

=b Z

B-1 A, B-1b

. A C

2008/

m=4

d=3

0 0 7

1 0 1
= , cB = (1, 1,1,0), c
0 1 0

0 0 1

0
B=
0

(0,0) ,

-5
A C-:
: A C 0
,
.

=b

: A C > 0



MPS
.
)(1.2
G1:7x1+10x2=1000
G2:x1= 10
G3:x2=10
S.t.x1+2x2<=60
x1,x2 0

:
G1:7x1+10x2+d1=1000
G2:x1+d2= 10
G3:x2+d3=10
S.t.x1+2x2<=60
x1,x2 0 , d1,d2,d3 , free

Minf=d1+d2+d3
7x1+10x2 =1000 + d1
x1
=10 + d2
x2
=10 + d3
x1+2x2+s1=60

n=6

1000
10
10
60

MPS -:
)(2.1

(1)1

d1=580,d2
=-50,d3=10,x1=60,MINZ=540
.
MPS


MPS
.

MPS
:
2
G1:3x1+5x2=15
G2:x1+x2=9
S.T. x1+x2<=7

G1:7x1+8x2=100
G2:x1=50
G3:x2=40
S.T. 2x1+x2<=80
G1:2x1+3x2=18
G2:2x1+x2=12
G3:x1+x2=8
S.T. x1<=6
x2<=3

5
G1:100x1+150x2=2000
G2:4x1+2x2=45
G3:2x1+2x2=35
S.T. x1+2x2<=20

6
G1:x1+x2=10
G2:5x1+3x2=56
S.T. 2x1+3x2<=12

:

2
2
1
3
1
4
1
5
1
6

2008/
-3:
3.1:

:

1
2
3
4
5
6

7
4
6
7
7
4

MPS

1
2
3
4
5
6

1
2
1
1
1
1

-3.2:








:
-1 MPS


.
-2 MPS
.
-3 ,

.
-4



.
-5
.

2008/

(1)1

13- Srivastava .U.K.,"quantitative techniques

for managerial decisions",Wiley
Eastern
Limited ,New
Delhi,(1991).
14-. Taha H. A , "operations research , an
interoduction", fifth edition,( 2000).

6

.

", , ,"
(1995), .
.
" , -2
,"
),
.(2005
3-Adam E.and Edert R."Production
Operations Management",third
edition ,prentice Hall,Adivision of
simon and Schuster Inc.,U.S.A,(1982)
4-Karim A.,"Multi-opjective
Optimization techniques",
http://www.glue.umd.edu./,(2000)
5-Azarm S.,"Multi objective optimum
design", http://www.glue.umd.edu./,
(1996)
6- Gass S.I. On the solution of linear
programming problems with free
variables, computer and operations
research vol-12 no.3 , (1985)
7-Holzman A..G. ,"Mathematical
programming for operations
researchers
and computer scientist
",Marcel Dekker
Inc.,New york ,
(1981).
8-Ignizio J.P.,"Goal programming and
extension ",D.C. Heath ,Lexington,
M.A.,(1976).
9- Ignizio J.P.,"linear programming in single
and multi objective systems",Prentice
Hall,Englewood cliffs,N.J.,(1982).
10-Kothari C.R,"An introduction to
operational research ",Vikas
publishing House PVT LT D,(1992).
11-Markland R.E.,"Topics in Management
Science ",Jon Wiley and Sons,New
York,(1983).
12-Philips D.T.,Ravindran A.,and Solberg
J.J,"operations research :principles and
practice ", Jon Wiley and Sons,New
York,(1987).

2008/

(1)1

Abstract
The classical method is one of the common important methods
in the field of operation research to solve linear goal programming
problems.
This method represents the deviation variable as difference between
two nonnegative variables in the objective function and the goals ,
The problems are then solved by using linear programming
methods.
The aim of this paper is to develop such method .The
developed method MPS is used then to solve linear goal
programming problems. It is found that MPS method is better than
the classical method, since MPS method reduces the solution tables
which makes the calculations easier and reduces the time required
to solve the problems under study .

Solving linear Goal Programming

by Using the Matrices
Rasheed Basheer Reheima
University of Thi Qar-College of Siences