ISSN: 2277-3754

ISO 9001:2008 Certified

International Journal of Engineering and Innovative Technology (IJEIT)
Volume 4, Issue 8, February 2015

Optimum Solution of Integer Programming

Problem by New Approach
Putta Baburao; Pranay N. Khobragade and N.W.Khobragade
Department of Mathematics, RTM Nagpur University, Nagpur -440033

Abstract- In this paper, new alternative simplex method for Step (7). Test the integerality of the optimum solution
the solution of IPP is introduced. This method is easy to solve (a) If the optimum solution includes all integer
Integer programming problem. This is powerful method to get values, an optimum basic feasible integer
improved solution. It reduces number of iterations and save solution has been obtained.
valuable time by skipping calculations of net evaluation.
(b) If the optimum solution does not include all –
Key words: Integer programming problem, optimal integer values then proceed onto next step.
solution, simplex method, alternative method. Step (8). Examine the constraint equations corresponding
to the current optimum solution. Let these equations be
I. INTRODUCTION represented by
n

 y
Integer programming problem is a special class of
L.P.P. where all or some variables are constrained to ij x j  bi [ i  0 1 2  m]
j 0
assume non – negative integer values. This type of
problem is of particular importance in business and Where n denotes the number of variables and m the
industry where discrete nature of the variables are number of equations.
involved in many decision – making situations. Choose the largest fraction of bi s i.e. find max {bi} f .
i
Khobragade et al. [1-3, 6-14] suggested an alternative
approach to solve linear programming problem. Let it be [bk ] f or write it simply as f k 0 .
In this paper, an attempt has been made to solve integer Step (9). Express each of the negative fractions if any, in
programming problem (IPP) by new method which is an the kth row of the optimum simplex table as the sum of a
alternative simplex method. This method is different from negative integer and a non – negative fraction.
Khobragade et al. [1-3, 6-14] Method. n
Step (10). Find the Gomorian constraint f
j 0
kj x j  fk0
II. ALL I.P.P. ALGORITHM
The iterative procedure for the solution of an all – and append the equation
integer programming problem is as follow: n
Step (1). Convert the minimization I.P.P. into that of
(1)
Gsla   f k 0   f kj x j
maximization, if it is in the minimization form. Ignore the j 0
integerality condition. to the current set of equation constraints.
Step (2). Introduce stack/or surplus variables, if necessary Step (11). Starting with this new set of equation
to convert the inequations into equations and obtain the constraints, find the new optimum solution by dual
optimum solution of the given I.P.P. by using new (1)
simplex algorithm (so that G sla is the initial leaving
simplex algorithm [1]
C j  xij
basic variable).
Step (3). Select max xij ≥0, for entering Step (12). If this new optimum solution for the modified
,
vector. I.P.P. is an integer solution, it is also feasible and
optimum for the given I.P.P. Otherwise return to step (4)
Step (4). Choose greatest coefficient of decision and repeat the process until an optimum feasible integer
variables. solution has been obtained.
(i) If greatest coefficient is unique, then element
corresponding to this row and column becomes pivotal III. SOLVED PROBLEMS
(leading) element. PROBLEM- 1
(ii) If greatest coefficient is not unique, then use tie Maximize Z  7 x1  9 x2
breaking technique. Subject to the constraints:
Step (5). Use usual simplex method for this table and go x1  3x2  6 ,
to next step.
7 x1  x2  35 , x1 , x2  0 and are integers.
Step (6). Ignore corresponding row and column. Proceed
SOLUTION: Introducing the slack variables x3  0 and
to step 4 for remaining elements and repeat the same
procedure until an optimal solution is obtained or there is x4  0 in the equations of the constraints, the
an indication for unbounded solution.

218
 ISSN: 2277-3754
ISO 9001:2008 Certified
International Journal of Engineering and Innovative Technology (IJEIT)
Volume 4, Issue 8, February 2015
inequations becomes equations. An initial basic feasible x1  4 , x2  3 and max Z  55 .
solution is
xB  [ x3 , x4 ]  [6, 35 ] PROBLEM- 2
Starting Table: The owner of a ready – made garments store makes
7 9 0 0 two types of shirts known as zee shirt and Button – Down
CB yB xB x1 x2 S1 S2 shirts. He makes a profit of Re. 1 and Rs. 4 per shirt on
0 S1 6 –1 3 1 0 Zee Shirts and Button – Down shirts respectively. He has
0 S2 35 7 1 0 1 two tailors [Tailor A and Tailor B] at his disposal to stitch
these shirts. Tailor A and Tailor B can devote at the most
0 S1 11 0 22/7 1 1/7 7 hours and 15 hours per day respectively. Both these
7 5 1 1/7 0 1/7 shirts are to be stitched by both the Tailors. Tailor A and
x1
Tailor B spend two hours and five hours respectively in
  stitching a Zee Shirt and four hours and three hours
9 x2 7/2 0 1 7/22 1/22
respectively in stitching of a Button – Down Shirt. How
7 x1 9/2 1 0 -1/2 3/22 many shirts of both the types should be stitched in order
to maximize daily profits ?
Applying Gomory Constraint Technique SOLUTION: [Formulation of the problem]
Suppose the owner of ready made garments decide to
7 9 0 0 0 make x1 Zee Shirts and x 2 Button – Down shirts. Then
CB yB xB x1 x2 S1 S2 G1 the availability of time to Tailors has the following
restrictions:
9 x2 7/2 0 1 7/22 1/22 0
2 x1  4 x2  7 , 5 x1  3x2  15 , x1 , x2  0 and x1 is
7 x1 9/2 1 0 -1/22 3/22 0
an integer.
1
0 G -1/2 0 0 -7/22 -1/22 1 The problem of the owner is to find values of x1 and x 2
 
to maximize the profit Z  x1  4x2 .
9 x2 3 0 1 0 0 1
Introduce the slack variables x3  0 and x4  0 in the
7 x1 32/7 1 0 0 1/7 -1/7
constraints of the given problem, an initial basic feasible
0 S1 11/7 0 0 1 1/7 -22/7 solution is xB  [ x3 , x4 ]  [7, 15 ]
Starting Table:
7 9 0 0 0 0 1 4 0 0
S1 S2 G1 G2 CB yB xB x1 x2 S1 S2
CB yB xB x1 x2
0 S1 7 2 4 1 0
9 3 0 1 0 0 1 0 0 S2 15 5 3 0 1
x2
 
7 x1 32/7 1 0 0 1/7 -1/7 0 0 S1 1 0 14/5 1 -2/5

1 x1 3 1 3/5 0 1/5
0 S1 11/7 0 0 1 1/7 - 0
22/7
0 G2 -4/7 0 0 0 -1/7 -6/7 1 4 5/14 0 1 5/14 -1/7
x2
1 x1 39/14 1 0 -3/14 2/7
9 x2 3 0 1 0 0 1 0

Second Iteration: [Modified Table]

7 x1 4 1 0 0 0 -1 1
CB yB xB x1 x2 S1 S2 G1
0 S1 1 0 0 1 0 -4 1 4 5/14 0 1 5/14 -1/7 0
x2
0 S2 4 0 0 0 1 6 -7 1 x1 39/14 1 0 -3/14 2/7 0
This table shows that the optimum feasible solution has 0 G1 - 0 0 - -2/7 1
been obtained in integers. Hence an integer optimum 11/14 11/14
solution to the given I.P.P. is  
219
 ISSN: 2277-3754
ISO 9001:2008 Certified
International Journal of Engineering and Innovative Technology (IJEIT)
Volume 4, Issue 8, February 2015
4 x2 3/4 0 1 0 0 -1/2 1 x1 25/6 1 0 0 1 -5/6
1 x1 2 1 0 0 0 1 0 S1 8/3 0 0 1 3 -10/3
Apply G C T
0 S2 11/4 0 0 11/4 1
CB yB xB x1 x2 S1 S2 G1 G2

This gives us an optimum integer solution is: x1  2 , 1 x2 1 0 1 0 -1 1 0

x 2  3 / 4 and max. Z  5 . 1 x1 25/6 1 0 0 1 -5/6 0

PROBLEM- 3 0 S1 8/3 0 0 1 3 - 0
A manufacturer of baby – dolls makes two types of 10/3
dolls : Doll X and Doll Y. Processing of these two dolls is 0 G2 -1/6 0 0 0 0 -1/6 1
done on two machines, A and B. Doll X requires two
1 x2 0 0 1 0 -1 0 6
hours on machine A and six hours, on machine B. Doll Y
requires five hours on machine A and also five hours on 1 x1 5 1 0 0 1 0 -5
machine B. There are sixteen hours of time per day
0 S1 6 0 0 1 3 0 -20
available on machine A and thirty hours on machine B.
0 G1 1 0 0 0 0 1 -6
The profit gained on both the dolls is same i.e. one rupee
per doll. What should be the daily production of each of This gives the optimum integer solution is
the two dolls?
SOLUTION. [Formulation of the Problem] :
x1  5 , x2  0 and max. Z  5 .
Suppose the manufacturer decides to produce Thus the manufacturer should produce 3 dolls of type X,
2 dolls of type Y in order to get the maximum profit of
x1 Dolls of type X and x 2 dolls of type Y. Rs. 5.
Then availability of time on two machines has the
following restrictions. PROBLEM- 4
2 x1  5 x2  16 , 6 x1  5 x2  30 and x1 , x2  0 . Max z  3x1  x 2  3 x3
The manufacturer wishes to determine the value of x1 Subject to the constraints:
and x 2 so as to maximize the profit Z  x1  x2 .  x1  2 x 2  x3  4 , 4 x 2  3 x3  2 ,
Introduce the slack variable x3  0 and x4  0 in the x 1 3 x 2  2 x3  3 and x1 , x 2 , x3  0
Constraints of the given problem, an initial basic feasible SOLUTION. Starting table
solution is xB  [ x3 , x4 ]  [16 , 30 ] . CB yB xB x1 x2 x3 S1 S2 S3
Starting Table: 0 S1 4 -1 2 1 1 0 0
0 S2 2 0 4 -3 0 1 0
Apply G C T 0 S3 3 1 -3 2 0 0 1

CB yB xB x1 x2 S1 S2
0 S1 3 -1 0 5/2 1 -1/2 0
0 S1 16 2 5 1 0 1 x2 1/2 0 1 - 0 1/4 0
0 S2 30 6 5 0 1 3/4
0 S3 9/2 1 0 - 0 3/4 1
1/4
1 x2 16/5 2/5 1 1/5 0
0 S2 14 4 0 -1 1 3 6/5 -2/5 0 1 2/5 -1/5 0
x3
1 x2 9/5 0 1 3/10 -1/10 1 7/5 - 1 0 3/10 2/3 0
x2
1 7/2 1 0 -1/4 1/4 3/10
x1 S3 24/5 9/10 0 0 1/10 7/10 1
0
CB yB xB x1 x2 S1 S2 G1
1 x2 9/5 0 1 3/10 -1/10 0 3 x3 10/3 0 0 1 4/9 1/9 4/9

1 x1 7/2 1 0 -1/4 1/4 0

1 x2 3 0 1 0 1/3 1/3 1/3
1
0 G -4/5 0 0 -3/10 -9/10 1
3 x1 16/3 1 0 0 1/9 7/9 10/9
1 x2 1 0 1 0 -1 1

220
 ISSN: 2277-3754
ISO 9001:2008 Certified
International Journal of Engineering and Innovative Technology (IJEIT)
Volume 4, Issue 8, February 2015
Apply G C T 0 S1 5/3 0 0 0 1 2/3 0 -
CB yB xB x1 x 2 x3 S1 S2 S3 G1 10/3
Again apply GCT :
CB y B xB x1 x 2 x3 S S2 S G1 G2
3 x3 10/3 0 0 1 4/9 1/9 4/9 0
1 3

1 x2 3 0 1 0 1/3 1/3 1/3 0 4 2 1 0 0 0 0 0 1 0

x1
3 x1 16/3 1 0 0 1/9 7/9 10/9 0
0 G1
-1/3 0 0 0 -1/9 - -10/9 1 6 x2 7/6 0 1 0 0 1/6 0 1/6 0
7/9
2 x3 35/6 0 0 1 0 -1/6 1 5/6 0
3 x3 2 0 0 1 0 -3 -4 4

1 x2 2 0 1 0 0 -2 -3 3 0 S1 5/3 0 0 0 1 2/3 0 - 0
10/3
3 x1 5 1 0 0 0 0 0 1
0 G2 -5/6 0 0 0 0 -5/6 0 -5/6 1
0 S2 3 0 0 0 1 7 10 -9
4 x1 2 1 0 0 0 0 0 1 0
Solution is, x1 =5 , x 2 =2 , x 3 =2, Max z= 23
PROBLEM- 5 6 x2 1 0 1 0 0 0 0 1/6 1/5
Max z  4 x1  6 x 2  2 x3
Subject to the constraints: 2 x3 6 0 0 1 0 0 1 5/6 -1/5
4 x1  4 x 2  5 ,  x1  6 x2  5 ,
 x 1  x 2  x3  5 and x1 , x 2 , x3  0 0 S1 1 0 0 0 1 0 0 - 4/5
10/3
SOLUTION. Stating table 0 S2 1 0 0 0 0 1 0 1 -6/5
CB yB xB x1 x2 x3 S1 S2 S3

0 S1 5 4 -4 0 1 0 0 Solution is , x1 =2 , x 2 =1 , x 3 =6 and Max Z = 26

0 S2 5 -1 6 0 0 1 0
0 S3 5 -1 1 1 0 0 1
IV. CONCLUSION

0 S1 25/3 10/3 0 0 1 2/3 0 An alternative simplex method have been derived to

6 5/6 -1/6 1 0 0 1/6 0 obtain the solution of Integer programming problem. The
x2 proposed algorithms have simplicity and ease of
0 S3 25/6 -5/6 0 1 0 - 1 understanding. From the above examples, authors
1/6 observed that this method reduces number of iterations
and improves the optimum solutions in most of the cases.
4 x1 5/2 1 0 0 3/10 1/5 0 This method save valuable time as there is no need to
6 x2 5/4 0 1 0 1/20 1/5 0 calculate the net evaluation Zj-Cj.
0 S3 25/4 0 0 1 1/4 0 1
REFERENCES
Apply G C T [1]. Mrs Lokhande K. G., Khobragade N. W., Khot P. G.:
CB yB xB x1 x 2 x3 S1 S2 S3 G1 Simplex Method: An Alternative Approach, International
Journal of Engineering and Innovative Technology,
Volume 3, Issue 1, P: 426-428 (2013).
4 x1 5/2 1 0 0 3/10 1/5 0 0
[2]. Khobragade N. W. and Khot P. G.: Alternative Approach
6 x2 5/4 0 1 0 1/20 1/5 0 0 to the Simplex Method-I, Bulletin of Pure and applied
Sciences, Vol. 23(E) (No.1); P. 35-40 (2004).
0 S3 25/4 0 0 1 1/4 0 1 0
0 G1 -1/2 0 0 0 -3/10 -1/5 0 1 [3]. Khobragade N. W. and Khot P. G.: Alternative Approach
to the Simplex Method-II, Acta Ciencia Indica, Vol.xxx
IM, No.3, 651, India (2005).
4 x1 2 1 0 0 0 0 0 1
[4]. Sharma S. D.: Operation Research, Kedar Nath Ram Nath,
6 x2 7/6 0 1 0 0 1/6 0 -1/6 132, R. G. Road, Meerut-250001 (U.P.), India.
0 S3 35/6 0 0 1 0 -1/6 1 5/6

221
 ISSN: 2277-3754
ISO 9001:2008 Certified
International Journal of Engineering and Innovative Technology (IJEIT)
Volume 4, Issue 8, February 2015
[5]. Gass S. I.: Linear Programming, 3/e, McGraw-Hill awarded Ph.D Degree and six students submitted their thesis in
Kogakusha, Tokyo (1969). University for award of Ph.D Degree under their guidance.
[6]. Ghadle, K.P; Pawar, T.S and Khobragade, N.W (2013):
Solution of Linear Programming Problem by New
Approach, Int. J. of Engg. And Information Technology,
vol. 3, Issue 6, pp.301-307
[7]. Khobragade, N.W, Lamba, N.K and Khot, P. G (2009):
“Alternative Approach to Revised Simplex Method”,
Int. J. of Pure and Appl. Maths. vol. 52, No.5, 693-699.
[8]. Khobragade, N.W, Lamba, N.K and Khot, P. G (2012):
“Alternative Approach to Wolfe’s Modified Simplex Pranay Khobragade student of M.E (Final) in Information Technology
Method for Quadratic Programming Problems”, Int. J.
Latest Trends in Maths. vol. 2, No. 1, pp. 19-24. of R.A.I.T College of Engg, Nerul, New Mumbai.

[9]. Mrs. Vaidya N.V and Khobragade, N.W (2012):

“Optimum solution to the simplex method, an alternative
approach”, Int. Journal of Latest Trends in Maths,
(accepted), UK.
[10].Mrs.Vaidya, N.V and Khobragade, N.W (2013): Solution
of Game problems using New Approach, Int. J. of Engg.
And Information Technology, vol. 3, Issue 5, pp.181-186.
[11].Mrs Lokhande, K.G; Khobragade, N.W, and Khot, P. G
(2013): “Alternative Approach to Linear Fractional
Programming”, Int. J. of Engg. And Information Putta Baburao for being M.Sc, Mphil in Maths, he has been teaching
Technology, vol. 3, Issue 4, pp.369-372. since 2001 for 14 years at P B Siddhartha College of Arts and Sci,
[12].Khobragade, N.W, Lamba, N.K and Khot, P. G (2013): Vijaywada (A.P).
“Solution of LPP by KKL Method”, Int. J. of Engg. And
Information Technology, vol. 3, Issue 4, pp.334-340.
[13].Khobragade, N.W, Lamba, N.K and Khot, P. G (2013):
“Solution of Game Theory Problems by KKL Method”,
Int. J. of Engg. And Information Technology, vol. 3, Issue
4, pp.350-355.
[14].Mrs. N.V Vaidya and Khobragade, N.W (2014):
“Approximation algorithm for optimal solution to the
linear programming problem”, Int. Journal of Maths in
Operational Research, Vol.6, No.2, pp 139-154.
AUTHOR BIOGRAPHY

Dr. N.W. Khobragade for being M.Sc in statistics and Maths, he

attained Ph.D in both subjects. He has been teaching since 1986 for
28 years at PGTD of Maths, RTM Nagpur University, Nagpur and
successfully handled different capacities.

At present he is working as Professor. Achieved excellent experiences

in Research for 15 years in the area of Boundary value problems
(Thermoelasticity in particular) and Operations Research. Published
more than 200 research papers in reputed journals. Sixteen students

222