, CHAPTER 26

LINEAR PROGRAMMING
INTRODUCTION AND GRAPHICAL METHOD
lroduction:
,Inear programming is a mathematical technique that deals with the
unization of a linear function of variables known as objective function subject
set of linear inequalities known as constraints. The .cojective function may
profit, cost, production capacity or any other measure of effectiveness, whic~
10 be obtained in the best possible or optimal manner. The constraints may be
by different resources such as labour, finance, materials, machines,
ket, technology etc. By linearity is meant a mathematical expression in which
xpressions among the variables are linear. The word 'linear' means that the
ments in a situation are so related that they can be plotted as straight lines on a
ph.

near programming may be defined as a method of determining an optimum

mme of inter-dependent activities in view of available resources. This
hnique is readily applicable to problems that can be described as follows:-

11 entrepreneur (manager) has given amounts of resources (man-hours,

hine time, quantity of resourc esetc.) available for use in a productive activity.
output per unit of resources consumed and the return per unit of output
sumed are both known to the management. The objective is to determine that
nbination of inputs which optimises the goals of the firm (e.g., maximises
,fitor minimises cost)."

hnear programming problem has two basic parts:

I the objective function, which describes the primary purpose of the
formulation- to maximise some return (profit) or to mini mise some cost
(production cost, investment cost): and
the constraint set, which is.a system of equalities and lor inequalities
describing the restrictions (conditions or constraints) under which
optimisation is to be accomplished. (For example, these restrictions
may be on machine time, man-hours, materials, contract specification,
etc. )
~"
" general. linear programming can be used for optimisation problems if the
lowing conditions are satisfied:

459
(i) There must be a well-defined objective function (profit, COSI "I which bakeries should fill orders for cookies at different locations, and many .
quantities produced) which is to be either maximised or minimised IIt1,j hers.
which can be expressed as a linear function of decision variables.
In the heavy industry, linear programming has been used in the iron and steel
(ii) There must be restrictions on the amount or extent of attainment or II
Industry to decide the types of products to be made in the rolling mills so as to
objective and these restrictions must be capable of being expressed •
imise the profit.
linear equalities or inequalities in terms of variables.
(iii) There must be alternative courses of action. For example, a SlWII Metal working industries use linear programming for shop loading and for
product may be processed by two different machines and problem m I errnining the choice between producing and buying a particular part. Paper
be to determine how much of the product to allocate to which macluu d textile industries have used linear programming to define the optimal cutting
(iv) Another necessary requirement is that decision variables should I ethod in order to minirnise trim losses. Linear Programming has also been
interrelated and non-negative. The non-negativity condition shows Ih," in determining the best route for aircraft and ships.
linear programming deals with real-life situations for which nellJ,lU
Iication of linear programming has also been made in the service industries.
quantities are generally iIlogicaJ..
Accounting firms use it for asset valuation and for assigning auditors to tasks in
(v) The resources must be in limited supply. For example. if a firm SI,UI
optimal way. Financial institutions and firms have used linear programming
producing greater number of a particular product, it must make srnnlh ,
evaluating investment plans, for the selection of Bond or Mutual Funds
number of other products as the total production capacity is limited.
olios, for Capital Budgeting and for long-range financial planning, etc.
unilarly, in advertising media, linear programming has been employed for
Applications of Linear Programming
igning advertising expenditure to'different media plans.
Despite the restriction that relationships must be approximately linear, Iiiit'll
programming as an D.R. technique has been recognised and widely used III Administration, Education and Politics have also employed linear progranuning
industry as an effective tool for business problems during the last few decade solve their problems, viz., in planning political campaign strategies; for
locating resources in education; for making school assignments in large
The first application of linear programming technique was made in 1947 in III
stricts; for optimal city administration; and for resource allocations in local
.u.S. Air Force. After the Second World War, it became necessary to concentuu
,Iection campaigns. This list of applications of linear programming can go on
on the development of a model that would define the optimal and efficient tl'
ever. However, the most recent applications of linear programming have
ordinadon of a nation's energies in the event of a total war.
rfaced themselves as a result of the energy crisis in which it is used to assess
The use of linear programming in business and industry has been tremendon .ergyoptions.
since 1951. The first and the most widely used industrial applications have 1)"'11
her applications of linear programming include problems of gasoline blending,
in all phases of the petroleum industry, viz., exploration, production. refi1111 I
ructural design, scheduling of a military tanker fleet, minimising the number of
distribution and pollution control. Here an oil refinery has a choice of bUYIII
.ers to meet a fixed schedule, product distribution, cost cutting in business,
crude oil from several different sources with different compositions and .1
t scheduling, stocks and flows, the balancing of assembly lines etc.
differing prices. The refinery can produce different products like aviation 1\11'1
diesel fuel and gasoline with varying quantities. The constraints may be lit In addition, linear programming is used to aid management in maximising
restrictions on quantity of the crude oil available at a particular source, lit olio selections, allocating advertising budgets among competing media to
capacity of the refinery to produce a particular product, etc. Then a mix of Iii imise market penetration, and designing containers for maximum storage
purchased crude oil and the manufactulW products is sought that gives III to minimise transportation costs.
maximum profit.
ISUI11ptionsof Linear Programming
The second most active user of linear programming is the food It IS important to remember the assumptions of linear programming. These are:
industry. For this industry, examples of the use of linear programming are I' (I) Linearity
determine the optimal mix of dog-and-cat feeds; to determine the opll11I1 II Costs, revenues or any physical properties which form the basis of the
allocation of cast food from various plants to different warehouses: to deternun problem vary in direct proportion (i.e. linearly) with the quantities or
numbers of components produced.

460 461
 (ii) Divisibility
factors most effectively by a better selection and distribution of these
Quantities, revenues and costs are infinitely divisible _ i.e. any fraction
or decimal answer is valid. elements. For example, more efficient use of manpower and machines
can be obtained by the use of linear programming.
(iii) Certainty u) It improves the quality of decisions. The individual who makes use of
linear programming methods becomes more objective than subjective.
The technique makes no allowance for uncertainty in the estimat
The individual having a clear picture of the relationships within the
made, although the evaluation of dual values indicates the sensitivity 01
basic equations, inequalities or constraints can have a better idea about
the solution to marginal uncertainty in constraint values.
the problem and its solution.
(iv) Positive solutions III) It can go a long way in improving the knowledge and skill of
tomorrows executives.
Non-negativity constraints are introduced to ensure only positive value.
are considered. (IV) Although linear programming gives possible and practical solutions.
(v) there might be other constraints operating outside the problem which
The usefulness of linear programming models for decision making
must be taken into account, for example, sales, demand. etc. Just
depends on how well they simulate the conditions under which the firm
because we can produce so many units does not mean that they can be
actually operates. The solution is only optimal for the available input
data. . sold. Linear progranuning method can handle such situations also
(vi) because it allows modification of its mathematical solutions.
Interdependence between demand for products is ignored: products may
IV) It highlights the bottlenecks in the production processes. When
be complementary or a substitute for one another.
(vii) bottlenecks occur, some machines cannot meet demand while others
The model relies on costs being split into tixed and variable elements
remain idle for at least part of the time. Highlighting of bottlenecks is
This may be difficult to achieve in practice, furthennore it is diffi<;:ultto
one of the most significant advantages of linear programming.
make allowance for fixed costs in detennining the solution _ any
attempt to absorb them on a unit basis (i.e. treating them as variabl
As an admifustrative tool, linear programming has certain distinct advantages.
·I costs) is likely to result in non-linear relationships.
Among these are the following:
(a) Insight and perspective into problem situations:
This could result in a situation where the solution (based on variable
(b) Consideration of all possible solutions to problems:
costs) suggests a small quantity of a component should be producecl but
(e) Better and more successful decisions:
consideration of the fixed set-up costs makes this unattractive.
(d) Better tools for adjusting to meet changing conditions.
(viii)
Time factors are ignored. All production is assumed to be
instantaneous. ~imitations of Linear Programming
(ix) There are some limitations associated with every mathematical technique and
Costs and benefits which cannot be quantified easily. such as liquidity
hnear programming is no exception. This technique has the following major
goodwill and labour stability, are ignored. This limitation is not specifi
hmitations:
to linear programming: it is a problem that obviously affects other
I. The first major limitation of linear programming is that it treats all
areas. However, in practice. consideration of liquidity is often almost a
relationships as linear, i.e., if direct cost of producing 10 units is sh. 100
important as the results obtained from whatever optimisation techrnqu
is being employed. then on 20 units, it is assumed to be sh. 200. In practice. however. the
direct costs of production may increase as output approaches capacity. It
Advantages of Linear Programming is assumed that the objective function and constraints call be
FOllowing are the main advantages of linear programming: represented as linear inequalities or equations. This will not always be
the case. However, it may be possible in many circumstances to
(i) It helps in attaining the optimum use of productive factors. Lineal
represent a non-linear relationship by linear approxi mations. It should
progranuning indicates how a manager can utilise his productlvi
also be remembered that standard costing, which is widely used in
practice also has the implicit assumption of linearity. In extreme cases

\ 462
463
 It.un the solution. The graphical method of solution may be extended to the
where there are three variables or unknowns. The graphical method is of no
of non-linearity it may be possible to use the more advanced techniqur .cal value when the number of variables exceeds three and is not
of non-linear programming. ended as a method of solution even in the three variables case, because it
2. It is assumed that any activity is infinitely divisible. Evidently in many Involves the construction of a "three-dimensional" graph. The method of
cases such an assumption would be physically unjustifiable. However ution which is used to solve more complex general allocation problems is
LP may give an approximate first solution to many problems whel ned the simplex method.
divisibility cannot be assumed. Where this is not the case, the use of II
more advance development, integer programming, may be appropriat general allocation, linear programming problems can be solved by the help of
3. All the parameters in the linear programming model are assumed to t lowing methods:-
known constants. Whereas, in reality they are frequently neither I) Graphical Method
known nor constants. ) Simplex Method
4. Yet another limitation of linear programming is that it takes mill
account single objective only, i.e., profit maximization, cOM rmulation of Linear Programming' Models
minimization, etc. However many organisations keep in view rnultiplr he three basic steps in formulating a linear programming model are as follows:
objectives such ;\S capacity utilization, profit maximization, liminn 1. Identify the decision variables to be determined and express them in
overtime etc. in arriving at strategic management decisions. In such terms of algebraic symbols.
situations linear programming formulations will yield infeasibl 2. Identify all the limitations or constraints in the given problem and then
solutions and one will instead have to apply goal linear prograrnmin express them as linear inequalities or equations in terms of above
Which takes into account multiple objectives. defined decision variables.
5. It can be adopted only under the condition of certainty, i.e., when thl 3. Identify the objective (criterion) which is to be optimized (maximize or
quantity of resources available, per unit contribution, cost, etc.. ,11 minimize) and express it as a linear function of the above defined
known with certainty. This does not hold good in real situations. decision variables.

LINEAR PROGRAMMING - METHODS OF SOLUTION Ithematical Fonnulation of Linear Programming Problems

The techniques used in the solution of linear programming problems can I there are n decision variables- and m constraints in the problem, the
divided into three main groups according to the methods used for solution ematical formulation of the LP is:
There are:
(a) General allocation method 'mise (maximise or minimise) Z = CI XI + C2X2 + ....+ c"X., subject to the
(b) Transportation method . ints;
(c) Assignment method. au XI + al2 X2+. + alo Xu~b,
321 XI + 3z2 X2+...+ ~-XnS b:!
These techniques are used as under:-

GENERAL ALLOCATION 3m1.XI + a...zXl +...+ 'a'n Xn~ b",

The general allocation method is the basic procedure which is of geneml
X •• Xl,'''' Xu ~ 0
application to problems where a range of possible outputs can arise from Ih
utilisation of limited resources such as land, materials, finance and machm Ive formulation may be put in the following compact form by using the
The technique formulates a mathematical model which portrays the interacuo
mmation notation.
and behaviour of the essential elements of a situation in the form of a set (II
algebraic expressions -. These expressions are either equations or inequalities III mise (rnax. or min.) Z = DI: cj.Xj(objective function)
the operation of the technique, inequalities are converted to equations for txllh
J-l
the plotting of graphs and the formu1ation of matrices. In a problem consrsun
of only two variables or unknowns, a granhical method may.be used hi 465

464