International Journal of Computer Science and Business Informatics

[Link]

An Optimal Solution to the Linear

Programming Problem using Lingo
Solver: A Case Study of an Apparel
Production Plant of Sri Lanka
Z. A. M. S. Juman and W. B. Daundasekara
Department of Mathematics,
Faculty of Science, University of Peradeniya,
Sri Lanka

ABSTRACT
Operations Research (OR) is often concerned with determining the maximum of profit,
performance, yield etc. or minimum of cost, loss, risk, etc. of some real-world objectives.
The Linear Programming Problem (LPP), a branch of Operations Research consists of an
objective function which is linear and constraints are linear equations or inequalities. This
research paper presents a Mathematical Model of a LPP which is to minimize the
production cost, while satisfying operational limitations, of a production plant producing t-
shirts. This production plant is known as Silk Line (pvt) Ltd which is located in Sri Lanka.
The developed model is considered to be a large scale one which provides a fully functional
cost effective system finding optimum number of machine operators and workers in each of
the departments in the production plant as well as finding optimum raw material for the
entire t-shirts production. The optimal solution to the model is found using the commercial
software package called “LINGO SOLVER”. Moreover, a sensitivity analysis is
performed to complete the target (entire t-shirts production) within a given specific period
of time.

Keywords
Linear Programming Problem, Optimum number of machine operators, Lingo Solver,
Sensitivity analysis.

1. INTRODUCTION
Linear programming is an optimization technique which was developed
during the Second World War. An LPP consists of an objective function
which is linear and constraints are linear equations or inequalities. Objective
function and constraints are formed using the decision variables which are
defined according to the given problem. Decision variables are continuous
to be real valued variables which may have lower or upper bounds.

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 1

 International Journal of Computer Science and Business Informatics

[Link]
Objective function is to be maximized or minimized subject to the
constraints which form a convex feasible region. An LPP can be solved
using an iterative algorithm known as the Simplex Algorithm which was
developed by Dantzig in 1950‟s. This is the most the widely used and
accepted algorithm to solve LPPs because of its simplicity. The optimal
solution of the objective function may be bounded or unbounded whereas
optimal solution to the decision variables may or may not be unique.
Linear Programming has a broader area of applications in the fields of
productions, telecommunications, transportations, scheduling etc. The main
focus of linear programming is to optimize the available resources in the
best possible manner while achieving the objective.
This research paper is focusing on optimizing production cost of a t-shirts
production plant which is located in Sri Lanka. Although the plant‟s monthly
target is 30,000 t-shirts, the records show that the target level has not been
reached in a regular basis. Also, it can be observed that the plant‟s
operational costs are mainly due to workers monthly payments which are
more than what is needed. Due to low productivity and unnecessary
expenditures, at present the plant is facing major financial difficulties and
several other problems in designing the production process. Therefore, the
management is interested in implementing a cost effective system to reduce
the production cost while improving the production efficiency of t-shirts. In
fact, in this research paper, a large scale linear programming model is
developed to achieve these objectives. Then LINGO SOLVER is used to
solve the developed large scale model.

The remainder of this paper is organized as follows: Related literature

review is given in section 2. Section 3 deals with the Methods and Materials
of the LPP. In section 4 the solution to the large scale LPP is summarized.
Finally, the conclusion given in section 5 highlights the limitations and
future research scope on the topic.

2. LITERATURE REVIEW

Linear Programming problem and its solution have been studied by many
authors. Few of them are James and Tom [5], Brain [1], Dantzig [2] and
Taha [13]. A number of computer software is available in obtaining the
solutions to LPP. Some are based on the simplex method and its variants,
e.g. CPLEX, LINDO, TORA, MATLAB, EXCEL SOLVER, AMPL and
LINGO, and some are based on the interior point algorithms, e.g. MOSEX.
See Fourer [3]. Though the software based on simplex method and its
variants have been used widely in solving linear programming problems,
they solve LPP in exponential time. An algorithm that solves LPP in

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 2

 International Journal of Computer Science and Business Informatics

[Link]
polynomial time is considered to be efficient. A first attempt of solving LPP
in polynomial time was the development of interior point algorithm by
Karmarkar [8]. How to solve a large scale LPP by interior point method
under MATLAB environment was proposed by Zhang [15]. The existing
interior point algorithms have some drawbacks such as extensive calculation
requirements, large number of iterations and large computer space
requirements. See Terlaky and Boggs [14].In addition, Hitchcock [4] was
first to develop the transportation model. After that, the transportation
problem, a special class of linear programming problem has been studied by
many researchers. Sharma and Sharma [12] presented the transportation
problem in a slightly different form in getting a dual problem which has a
special structure. Then they proposed a new solution procedure to solve the
dual of the incapacitated transportation problem. Sharma and Prasad [11]
presented a heuristic that provides a very good initial solution to the
transportation problem in polynomial time. Schrenk et al. [10] analyzed
degeneracy characterizations for two classical problems: the transportation
paradox in linear transportation problems and the pure constant fixed charge
(there is no variable cost and the fixed charge is the same on all routes)
transportation problems. A new result on complexity of the pure constant
fixed charge transportation problem has been proved. Liu [9] investigated
the transportation problem when the demands and supplies were varying
within their respective ranges. Following these variations the minimal total
cost were also varied within an interval. So, he built a pair of mathematical
programs where at least one of the supply or the demand was varying, to
compute the lower and the upper bounds of the total transportation cost.
Then the Lingo solver was used to solve the both mathematical programs to
attain the lower and the upper bounds of the minimal total transportation
costs. Juman and Hoque [7] demonstrated the deficiency of Liu‟s [9] method
in getting an upper bound of the minimal total costs of transportation. Then
they extended this Liu‟s model to include the inventory costs of the product
during transportation and at destinations, as they are interrelated factors.
In addition, they developed two new efficient heuristic solution techniques -
Algorithms 1 & 2 to find the upper and the lower minimal total cost
bounds respectively. By comparative studies of the solution techniques on
the solutions of small size numerical problems, it is observed that our
proposed heuristic technique (Algorithm1) performs the same or
significantly better in finding the upper bound of the minimal total cost as
compared with Liu‟s [9] approach. Algorithm 2 provided the same lower
bound of the minimal total costs to each of the numerical problems studied
as the one found by Liu‟s [9] approach. Moreover, numerical studies
demonstrated that the inclusion of inventory costs during transportation and
at destinations with the transportation costs changes the lower and the upper
minimal total cost bounds reasonably. Juman, Hoque and Bhuhari [6]

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 3

 International Journal of Computer Science and Business Informatics

[Link]
designed a C++ computational program of Vogel‟s Approximation Method
(to solve an unbalanced transportation problem by considering both the
balanced and unbalanced features respectively with and without adding a
dummy column) in obtaining an initial feasible solution (IFS) to an
unbalanced transportation problem (UTP). In order to get an initial feasible
solution to a large scale TP, this computational program of VAM is
preferred. Moreover, we examine the effect of dealing with the balanced and
the unbalanced features in applying the well-known VAM method for
solving an unbalanced transportation problem. First we illustrated the
solution procedures with numerical examples (chosen from the literature).
Then we demonstrated this effect by a comparative study on solutions of
some numerical problems obtained by VAM by considering the balanced
and the unbalanced features.

3. METHODS AND MATERIAL

3.1 Current approach
The current methodology, adopted by the plant, to increase the production
has failed due to several drawbacks. At present, the plant finds a feasible
solution by manually comparing alternative resource allocations. Although
the management is fully committed to improve the production process in
order to reach the optimum production, it has failed to achieve the objective.
It was observed that the current production process in the plant has several
weaknesses. This is mainly due to following reasons:

 Mismanagement of human resources

 Mismanagement of orders
 Improper resource utilization

3.2 Data Collection

Data were collected using Questionnaires and Interviews. Then the
mathematical model for the month of August 2012 was formulated using the
collected data.

3.3 Formulation of the Mathematical Model

3.3.1Decision Variables of the Model
Decision variables of the proposed model are as follows:
x1,x2,...,x19, x20 and x52, x53 = # trained and untrained cutter, band knife,
normal, 5-tread, 4-tread, flat lock, button hole, button attach, blind hem,
double needle, and fusing machine operators; x21,x22 = # mechanics needed

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 4

 International Journal of Computer Science and Business Informatics

[Link]
in cutting and sewing departments; x23, x24, x54, x44 = # instructors needed in
cutting, sewing, QC (quality control) and packing departments;
x31, x32, x55, x42 = # supervisors needed in cutting, sewing, QC and packing
departments; x29, x30 = # table checkers needed in cutting and sewing
departments; x25, x26, x39 = # helpers needed in cutting, sewing and packing
departments; x33, x34, x56, x43 = # clerks needed in cutting, sewing, QC and
packing departments; x40 = # menders needed in packing department; x41 =
# quality checkers needed in QC department; x38, x35, x37 , x45 = # in chargers
needed in packing, cutting, QC and sewing departments; x48 = # personal
managers; x47 = # account clerks; x27 , x28 = # quality instructors needed in
cutting and sewing departments; x36 = # production managers; x46 = # iron
tables; x49, x50, x51 = # small, medium, large t-shirts produced in a month.
(Here, # denotes „number of ‟)

3.3.2 Objective Function of the Model

The objective of the linear programming model presented below is to
determine the minimum production cost of the production plant.

Mathematical formulation of objective function:

Cost Function
56 51
Minimize Production Cost Z  
i 1
ci xi  4500 x46  c
j  49
j x j , where ci
i  46, 49, 50, 51

is the salary of the ith employee and cj is the raw material cost of the jth type
t-shirt subject to the following constraints :

3.3.3 Constraints of the Model

All the constraints in the model can be categorized into five different types
as shown in Table 1 below.

Table 1 Types of model constraints

Constraints Inequalities

30 x j 120x j1  c j ; j 5,7,9,...,19;

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 5

 International Journal of Computer Science and Business Informatics

[Link]
c j  # t-shirts completed in a day by jth type of machines in the sewing department

Machine xk  xk 1  ck ; k 1,3,5,...,17,19,52 ;
ck  # machine operators needed for the k type
th
of machines in cutting and sewing

departments

Cutting, Sewing, QC and Packing Departments must have a minimum # workers to

perform the duties.

x55  1; # supervisors needed in QC department.

Workers
x33  x56  x43 ; same # clerks needed in cutting, QC and packing departments.

x28  7 ; # quality instructors needed in sewing department.

Raw material 29x49  30x50  32x51  899964 inches

( raw material needed for the entire t-shirt production)

 29  80   30  80   32  80 
  49x    50x   51 x  Rs.1999920
 36   36   36 
Raw material cost
(raw material cost for the entire production)
 11 . 66  x   11 .66  x   11 .66  x  12480
  49   50   51 minutes
 20   20   20 
Time
( time available for the entire t-shirt production )

3.4 Linear Programming Model (LPM)

Finally, Linear Programming Model (LPM) for the existing problem could
be formulated as below. Optimal solution of the model is obtained using the
software known as LINGO as illustrated below:

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 6

 International Journal of Computer Science and Business Informatics

[Link]

Figure 1 LINGO Mainframe Window

Linear Programming Model:

MinZ  4200 X 1  6200 X 2  4200 X 3  6200 X 4  4200 X 52  6200 X 53  4200 X 5  6200 X 6 

4200 X 7  6200 X 8  4200 X 9  6200 X 10  4200 X 11  6200 X 12  4200 X13  6200 X14 
4200 X 15  6200 X 16  4200 X17  6200 X18  4200 X19  6200 X 2010000 X 21  10000 X 22 
15000 X 23  15000 X 24  10000 X 54  10000 X 44  8000 X 27  8000 X 28  4500 X 29  4500 X 30
(1)
 4500 X 40  10000 X 31  10000 X 32  10000 X 55  10000 X 42  5000 X 33  5000 X 34 
5000 X 56  5000 X 43  12000 X 47  5000 X 41  4000 X 25  4000 X 26  4000 X 39  14000 X 48 
30000 X 36  20000 X 35  20000 X 37  20000 X 38  20000 X 45  4500 X 46  2320  36X 49 
2400  36X 50  2560  36X 51

subject to

30X 5 120X 6  1200 ; 30X 7 120X 8  600 ; 30X 9 120X 10  600 ;

30X 11 120X 12  1000 ; 30X 13 120X 14  1500 ; 30X 15 120X 16  1500 ; (2)
30X 17 120X 18  1500 ; 30X 19 120X 20  1500

X 1  X 2  3 ; X 3  X 4  1 ; X 52  X 53  2 ; X 5  X 6  115 ; X 7  X 8  15 ;
X 9  X 10  8 ; X 11  X 12  4 ; X 13  X 14  2 ; X 15  X 16  3; X 17  X 18  2 ; (3)
X 19  X 20  12

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 7

International Journal of Computer Science and Business Informatics

[Link]

3X 2  1200; 3X 4  1200 ; 3X 53  1200 (4)

 X 5  X 6  X 7  X 8  X 9  X 10  X 11  X 12  X 13  X 14  X 15  X 16    (5)
45X 22   X  X  X  X 
 17 18 19 20 

45X 21  X1  X 2  X 3  X 4  X 52  X 53  0 ; 20X 23  X1  X 2  X 3  X 4  X 52  X 53  0 (6)

X 24  2X 23 ; X 54  2X 23 ; X 44  X 23 (7)

20X 31   X 1  X 2  X 3  X 4  X 52  X 53  (8)

 X 5  X 6  X 7  X 8  X 9  X10  X11  X 12  X13  X14  X15  X16  

60X 32   X  X  X  X 
 (9)
 17 18 19 20 

X 55  1 ; 50X 42  X 39  X 40  X 46  ; X 29  8 ; X 30  X 29 (10)

 X 5  X 6  X 7  X 8  X 9  X 10  X 11  X 12  X 13  X 14  X 15  X 16  
9X 26   X  X  X  X  (11)
 17 18 19 20 

20X 25 13X 26  0 ; 20X 39  8X 26  0 (12)

 X 5  X 6  X 7  X 8  X 9  X 10  X 11  X 12  X 13  X 14  X 15  X 16  
40X 34   X  X  X  X  (13)
 17 18 19 20 

4X 33  X 34  0 ; X 33  X 56  X 43 (14)

X 40  12 ; X 41  24
(15)

X 38  1 ; X 35  1 ; X 37  1 (16)

161
X  3; X X ; X  ; X  7 ; 2X X  0; X  1; X 1 (17)
48 47 48 22 28 27 28 36 45
45

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 8

International Journal of Computer Science and Business Informatics

[Link]
X 46  8; 29X 49  30X 50  32X 51  899964 (18)

2320X 49  2400X 50  2560X 51  71997120 (19)

11.66  X  11.66  X  11.66  X  12480

  49   50   51 (20)
 20   20   20 

X 49  14350 ; X 50  10450 ; X 51  5200 (21)

X  0 and integers ; X is the column vector consisting of decision variables. (22)

4. RESULTS
By solving the developed large scale mathematical model (as shown in
section 3.4) using Lingo solver, the solutions obtained are given below in
Table 2 and Figures 2-5. The comparison between the optimal number of
each type of machines and number of machines currently being used in the
plant is shown in the following table:

Table 2The optimal table for the types of machines

Departments Type of machine Available # Optimal # machines
machines
Cutter 3 3
Department of
cutting Band knife 1 1
Fusing 2 2

Normal 115 115

5-tread 13 15
4-tread 8 8
Department of Flat lock 4 8
sawing Button hole 12 13
Button attach 12 13
Blind hem 10 13
Double needle 12 13

where x1+x2 =# cutter machines;x3+x4= # band knife machines; x52+x53 = #

fusing machines;x5+x6 =# normal machines;x7+x8 =# 5-tread
machines;x9+x10= # 4-tread machines;x11+x12 = # flat lock machines;x13+x14

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 9

International Journal of Computer Science and Business Informatics

[Link]
= # button hole machines;x15+x16 = # button attach machines; x17+x18= #
blind hem machines; x19+x20= # double needle machines.

The Table 2 clearly shows the available number of machines and their
respective optimal amount. The cutting department consists of three cutter,
one band knife and two fusing machines in the plant. The optimum
quantities for these machines in the cutting department are the same. The
sawing department consists of hundred and fifteen normal machines,
thirteen 5-tread machines, eight 4-tread machines, four flat lock machines,
twelve button hole machines, twelve button attach machines, ten blind hem
machines, and twelve double needle machines. The optimum quantities for
these machines in the sawing department are 115, 15, 8, 8, 13, 13, 13, 13
respectively.
No. of mechanics in cutting dept.
Number of Workers currently working in the Factory No. of mechanics in sawing dept
No. of instructors in cutting dept.
30
28 No. of instructors in sawing dept.
No. of instructors in QC dept.
26 No. of instructors in packing dept.
No. of quality instructors in cutting dept.
25 No. of quality instructors in sawing dept.
No. of table checker in cutting dept.
No. of table checker in sawing dept.
20 No. of menders in packing dept.
20 No. of supervisors in cutting dept.
No. of supervisors in sawing dept.
No. of supervisors in QC dept.
17 No. of supervisors in packing dept.
15 14 No. of clerks in cutting dept.
12
No. of clerks in sawing dept.
Fig 1 No. of clerks in QC dept.
11
No. of clerks in packing dept.
10
No. of account clerks
No. of quality checkers in QC dept.
7 7 No. of helpers in cutting dept.
6 No. of helpers in sawing dept.
No. of helpers in packing dept.
5
3 3 3 3 No. of personal managers
2 22 2 2 No. of production managers
1 1 1 1 1 1 11 1 11 1 No. of cutting in chargers
No. of QC in chargers
0 No. of packing in chargers
Workers No. of work-study in chargers in sawing dept.
No. of iron tables

Figure 2 Number of workers currently working in the factory

The bar charts in Figure 2 clearly depict the available number of workers
who are currently working in the plant as follows: The cutting department
consists of a mechanic, three instructors, two quality instructors, eleven
table checkers, two supervisors, three clerks, seventeen helpers, and two
cutting-in-chargers. The sawing department consists of seven mechanics,

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 10

 International Journal of Computer Science and Business Informatics

[Link]
one instructor, one quality instructor, twelve table checkers, two
supervisors, six clerks, twenty six helpers, and one work-study-in-charger.
The QC department consists of one instructor, one supervisor, one clerk,
twenty eight quality checkers, and one QC in-charger. The packing
department consists of one instructor, twenty menders, two supervisors, one
clerk, seven helpers, and one packing in-charger. In addition to there are
three account clerks, three personal managers, one production manager, and
fourteen iron tables in the factory.

Optimum Number of Workers needed

for the Monthly Production
No. of mechanics in cutting dept.
No. of mechanics in sawing dept
No. of instructors in cutting dept.
25 24 No. of instructors in sawing dept.
No. of instructors in QC dept.
22 No. of instructors in packing dept.
No. of quality instructors in cutting dept.
No. of quality instructors in sawing dept.
20 No. of table checker in cutting dept.
No. of table checker in sawing dept.
No. of menders in packing dept.
No. of supervisors in cutting dept.
No. of supervisors in sawing dept.
15 14 No. of supervisors in QC dept.
No. of supervisors in packing dept.
12 No. of clerks in cutting dept.
No. of clerks in sawing dept.
No. of clerks in QC dept.
10 9 No. of clerks in packing dept.
8 No. of account clerks
88
No. of quality checkers in QC dept.
No. of helpers in cutting dept.
5 No. of helpers in sawing dept.
5
No. of helpers in packing dept.
4
No. of personal managers
3 3 3
No. of production managers
2
No. of cutting in chargers
1 111 1 1 1
11111 No. of QC in chargers
00 0 00 No. of packing in chargers
0
No. of work-study in chargers in sawing dept.
No. of iron tables

Figure 3 Optimum number of workers needed for the monthly production

The bar charts in Figure 3 clearly depict the optimum number of workers
needed for the monthly production of the plant as follows: The cutting
department consists of one mechanic, one instructor, eight table checkers,
one supervisor, two clerks, fourteen helpers, and one cutting-in-charger. The
sawing department consists of four mechanics, one instructor, eight table
checkers, three supervisors, five clerks, twenty two helpers, and one work-
study-in-charger. The QC department consists of one instructor, twenty four

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 11

 International Journal of Computer Science and Business Informatics

[Link]
quality checkers, and one QC in-charger. The packing department consists
of one instructor, twelve menders, one supervisor, nine helpers, and one
packing in-charger. In addition to there are three account clerks, three
personal managers, one production manager, and eight iron tables in the
[Link] should be noted that the total number of workers working in the
plant is 182 (see Figure 2). The Optimum number of workers needed for the
monthly production of t-shirts is 137 (see Figure 3). Thusour model which is
presented in this paper reduces the total number of workers by 32.9 %.
Expected number of T-shirts to be produced in August 2012

No. of small t-shirts

16000

14000 14350
12000

10000 10450 No. of medium t-

shirts
8000

6000

4000 5200
No. of large t-shirts
2000

Figure 4 Expected numbers of t-shirts (small, medium, and large) to be

produced in August

The bar charts in Figure 4 clearly depict the expected total number of t-
shirts to be produced in August 2012. The expected numbers of small,
medium and large size t-shirts are 14350, 10450, and 5200 respectively.

Result obtained in the Proposed Model

No. of m edium t-shirts

15000 produced in over tim e
Num ber of 3393
10000
Medium T-
5000 7057
shirts
0 No. of m edium t-shirts
1
Medium T-shirts

Figure 5 Number of medium size t-shirts produced in normal and over time

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 12

International Journal of Computer Science and Business Informatics

[Link]

The bar chart in Figure 5 clearly shows the exact numbers of medium size t-
shirts produced in normal and over times. The bottom of the bar chart
represents that the 7057 medium size t-shirts are produced in normal time
whereas the remaining 3393 (shown in top of the bar chart) are produced in
over time. It should be noted that, in order to complete the entire t-shirts
production in a specific time period, 3393 medium sized t-shirts must be
produced in overtime. See the Figure 5 given above.

Current & Optimum Production

Cost in August 2012

4,300,000
Current
production
4,200,000 4,230,283 cost
4,100,000

4,000,000
Optimum
3,900,000 production
3,800,000
cost
3,802,183
3,700,000

3,600,000

3,500,000

Figure 6 Comparative result of current and optimum production cost

Figure 6 clearly shows that the optimal monthly production cost of the plant
is Rs. 3,802,183. But, for the same month the plant estimated average
monthly production cost for this plant to be Rs. 4,230,283.

5. CONCLUSION
According to the optimal solution of the proposed model for the month of
August 2012, the monthly optimal production cost of the plant is Rs.
3,802,183. But in the same month the plant‟s estimated average monthly
cost is Rs.4,230,283. This clearly indicates that the plant is spending in
excess of Rs.428,100 than what is actually needed. Therefore, the plant can
save a maximum of Rs. 428,100 (10%) per month by implementing the
proposed method. Moreover, a sensitivity analysis is performed to complete

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 13

 International Journal of Computer Science and Business Informatics

[Link]
the t-shirts production within the specified period of time (one month). After
performing a sensitivity analysis, it was revealed that the production plant
needs at least 22760 meters of raw material to meet the given target.

In this research only the production cost is considered while assuming that
the monthly demand is a fixed quantity. But, in practice, for the manufacture
(production plant), there are setup costs and inventory costs of raw-material,
work-in-process and finished goods as well. In addition, buyers incur
transportation costs, ordering costs and inventory costs during transportation
and at the destination (buyer/retailer). Thus, future research might be carried
out by taking these costs of setup, inventory, ordering and transportation
into account. Also, demand can be considered as stochastic. Hence, an
integrated single-manufacture-multi-buyer model with stochastic demand
can be obtained. How to develop a streamline method to tackle this large
scale LPP after considering these costs is challenging and potential future
research.

ACKNOWLEDGMENTS
The authors acknowledge that this research is supported by the research
grant of University of Peradeniya, Sri Lanka. Also, they are grateful to the
referees for their valuable, constructive commentsand suggestions.

REFERENCES
[1] Brain, D.B. Basic Linear Programming. London, Spottiswoode Ballantyne Ltd,1984.
[2] Dantzig, G.B. Linear Programming and extensions. Princeton, NJ: Princeton
University press, 1963.
[3] Fourer, R. Survey of linear programming software. OR/MS today, (2001), pp. 58-68.
[4] Hitchcock, F.L. The distribution of a product from several sources to numerous
locations. Journal of mathematical physics, 20, (1941), pp. 224-230.
[5] James, P.I. and Tom, M.C. Linear Programming. Prentice, Hall, Inc, 1994.
[6] Juman, Z.A.M.S., Hoque, M.A. and Buhari, M.I.A sensitivity analysis and an
implementation of the well-known Vogel‟s approximation method for solving
unbalanced transportation problems. Malaysian Journal of Science, 32,1(2013), pp. 66-
72.
[7] Juman, Z.A.M.S., Hoque, M.A. A heuristic solution technique to attain the minimal
total cost bounds of transporting a homogeneous product with varying demands and
supplies. European Journal of Operational Research, 239, (2014)pp. 146-156.
[8] Karmarkar, N. A new polynomial time algorithm for linear programming.
Combinatorial, 4, (1984), pp. 373-395.
[9] Liu, S.T. The total cost bounds of the transportation problem with varying demand and
[Link], 31,(2003), pp.247-251.

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 14

 International Journal of Computer Science and Business Informatics

[Link]
[10] Schrenk, S., Finke, G and Cung, V.D. Two classical transportation problems revisited:
Pure constant fixed charges and the paradox. Mathematical and Computer Modeling,
54,(2011), pp. 2306-2315.
[11] Sharma, R.R.K. and Prasad, S. Obtaining a good primal solution to the uncapacitated
transportation problem. European Journal of Operational Research, 144, (2003) , pp.
560-564.
[12] Sharma, R.R.K and Sharma, K.D. A new dual based procedure for the transportation
problem. European Journal of Operational Research, 122, 3 (2000),pp. 611-624.
[13] Taha H. A. Operation Research: An introduction. Prentice-Hall of India, 8th edition,
2006.
[14] Terlaky, T. and Boggs, P.T. Interior point method. Faculty of information technology,
Delft. Netherlands, 2005.
[15] Zhang, Y. Solving large scale linear programming by interior point method under
MATLAB environment. Technical report, Mathematics Department, University of
Maryland, Baltimore Country, 1996.

This paper may be cited as:

Juman, Z. A. M. S. and Daundasekara, W. B., 2014. An Optimal Solution to
the Linear Programming Problem using Lingo Solver: A Case Study of an
Apparel Production Plant of Sri Lanka. International Journal of Computer
Science and Business Informatics, Vol. 14, No. 2, pp. 1-15.

ISSN: 1694-2108 | Vol. 14, No. 2. AUGUST/SEPTEMBER 2014 15