0.R.

Operations
Scopeorresources resources
essentially, andtraced
operations applied
Introduction
1.1
Problems
(12)Design
(11) (I0) (9)Transportation
(8) Problems
() (6) () (9) (3) (2) (1) is invention Operations
Chapter 1
useful at
decision
total
cost.
MarketingShortest minimize
Assignment Resource
Inventory
Maintenance
Sequencing Uses so least
QueuingProduction,Finance Research available. because
in
for or as.to
order to
of Research
solving
Applicattons theory.
the
Management
Technlques)
(Resource
Management route allocation
control obtain new
problems, PlanningManagement Different
or to of days
of and and
probler1s jobs the
techniques
Scheduling he
Resource tinvent
of
The is
Replacement
problems
etc. problems (Some0.Rof desired necess! Newton existence the Operations
Research
(0.R.)
and
problems to teams
to probleims like applicants techniques
Managentent study
control occurred
mention travelling Problems objective had win to and
of of
Froblems optimisation
problems toLagrange.
a to Models)
O.R. to thesinceoptimisation
few. sales do
maximise Techniques. manage
nomenclature
Hence
the
research withthwaer
person the
Rapid
techniques
techniques.
with World
total development
problems on
profit available
military limited War can
It
or Il be is
 1.2 Resource Management Techniques
in Business and Management
1.2Role of operations research
1. Marketing Management Operatlons research
definitely a role to play in
(a) Product selection
technlques
have
(b) Competitive strategies
(c) Advertising strategy etc.
2. Production Management
0.R techniques are very useful in the
The
production managemcnt.
following areas of
(a) Production scheduling
(b) Project scheduling
(c) Allocation of resources
(d) Location of factories and their si|es
(e) Equipment replacement and Maintenance
) Inventory policy etc.
3. Finance Management
The techniques O.R are applied to Budgeting and Investment areas
and especially to
(a) Cash flow analysis
(b) Capital requirement
(c) Credit policies
(d) Credit risks etc.
4. Personal Management
(a) Recruitment policies, and
(b) Assignment of jobs are some of the areas of personnel
management where O.R techniques are useful.
5. Purchasing and Procurement
(a) Rules for purchasing
(b) Determining the quantity
(e) Determining the time of purchaser are some of the areas where
0.R. techniques can be applied
6. Distribution
In determining
(a) location of
warehouses
(b) size of the warehouses
(c) rental outlets
(d) transportation strategies O.R. techniques are useful.
 OperationsRescarch (0.R) 1.3
1.3Role of O.R in Engineering
1. Optimal design of water resources systems
2. Optimal design of structures
3. Production, Planning, Scheduling and control
4. Optimal design of electrical networks
5. Inventory control
6. Planning of maintenance and replacement of equipment
7. Allocation of resources of services to maximise the benefit
8. Design of material handling
9. Optimal design of machines
10. Optimum design of controlsystems
|1. Optimal selection of sites for an industry to mention a few.

1.4 Classification of Models

The first thing one has to do to use O.R. techniques after formulating a
practical problem is to construct a suitable model to represent the
practical problem. Amodel is a reasonably simplified representation of a
real-world situation. It is an abstraction of reality. The models can
broadly be classified as
lconic (Physical) Models
Analogue Models
MathematicalModels
Also as
Static Models
Dynamic Models and, in addition, as
Deterministic Models
Stochastic Models
Models can further be subdivided as
Descriptive Models
Prescriptive Models
Predictive Models
Analytic Models
Simulation Models

lconic Model :
This is a physical., or pictorial representation of various aspects of a
system.
Example: Toy, Miniature model of a building, scaled up model of a
cell in biology etc.
 1,4 Resource Management Techniques
Analogue or Schematic model :
This uses one set of properties to represent another set of
which a system under study has properties
Exarple: Anetwork of water pipes to represent the flow of
in an clectrical network or graphs, organisational charts etc. current
Mathematical model or Symbolic model:
This uses a set of mathematical symbols (letters, numbers
represent the decision variables of a system under consideration, etc) to
These
variables related by mathematical equations or inequalities which
describes the properties of the system.
Example : A linear programming model, A system of
representing an electrical network or differential equations equations
dynamic systems etc. representing
Static Model:
This is a model which does not take
time into account. It assumes that
the values of the variables do not change
of time horizon. with time- during a certain period

Example :A linear programming problem, an

transportation problen1 etc. assignment problem,
Dynamic model is a model which considers time as one of the
important variables.
Exanple:A dynamic programming problenm, Areplacement problem
Deterministic model is a model which does not take uncertainty into
account.

Exanple :A linear programming problem, an

etc. assignment problem
Stochastic model is a model which considers
important aspect of the problem. uncertainty as an

Exanpie :Any stochastic programming problem, stochastic

models etc. inventory
Descriptive model is one which i=st describes a situation or
system.
Exanple : Anopinion poll, any survey.
Predictive model is one which predicts something based on some
Predicting election results before actually data.
the counting is comnpleed.
 Operations Research (O.R) 1.5

Prescriptive model is one which prescribes or suggests a course of

action for a problem.
Exanple : Any programming (linear, nonlinear, dynamic, geometric
erc)problem.
Analytic model is a model in which exact solution is obtained by
mathematical methods in closed form.
Simulation model is a representation of reality through the use of a
model or device which will react in the same manner as reality under a
given set of conditions. Once a simulation model is designed, it takes only
alittle time, in general, to run a simulation on a computer.
It is usually less mathematical and less time consuming and generally
leastexpensive as well, in many situations.
Example : queuing problems, inventory problems.
1.5 Some characteristics of a good Model
(1) Itshould be reasonably simple.
(2) A good model should be capable of taking ínto account new
changes in the` situation affecting its frame significantly with
ease i.e., updating the models should be as simple and easy as
possible,
(3) Assumptions made to simplify the model should be as sinall as
possible.
(4) Number of variables used should be as small in number as
possible.
(5) The model should be open to parametric treatment.
1.6 Principles of Modeling
(1) Do not build up a complicated model while a simple one will
suffice.
(2) Beware of moulding the problems to fit a (favourite !) technique
(3) Deductions must be made carefully.
(4) Models should be validated prior toimplementation.
(5) A model should neither be pressed to do nor criticised for
failing to do that for which it was never intended:
(6) Beware of overselling the model in cases where assumption
made for the construction of the model can be challenged.
(7) The solution of a model cannot be more accurate than the
accuracy of the information that goes into the construction of the
model.
(8) Models are only aids in decision making.
(9) Modelshould bc as accurate as possil le.
1.6 Resource Management Techniques
1.7 General Methods for solving O.R models
(1) Analytic Procedure : Solving models by classical
techniques like differential calculus, finite
obtain analytic solutions..
Mathematical
differences
etc. to
(2) Iterative Procedure: Starts with a trial solution anda eot
ules for improving it by repeating the procedure until further
improvement is not possible.
(3) Monte carlo technique : Taking sample
observations.
computing probability distributions for the variable
random numbers and constructing some functions to using
values of the decision variables. determine
.8 Main Phases of 0.R.
(1) Formulation of the Problems : Identifying the
decision variables involved and the constraintsobjective, the
that arise
involving the decision variables
(2) Construction of a Mathematical Model: Expressing the
measure of effectiveness which may be total profit, total cost,
utility et., to be optimised by a Mathematical function called
objective function. Representing the constraints like budget
constraints, raW materials constraints, resource constraints,
quality constraints etc., by means of mathematical equations or
inequalities.
(3) Solving the Model constructed: Determinmg the solution by
analytic or iterative or Monto-carlo method depending upon the
structure of the mathematical model.
(4) Controlling and updating : A solution which is optimum today
may not be so tomorrow. The valúes of the variables may
change, new variables may emerge. The structural relationship
between the variables may also undergo a change. AIl these are
determined in updating.
A solution from a model remains a solution only so long as the
uncontrollable variables retain their values and the relationship
between the variables does not change. The solution itself goes
"out of control" if the values of one or mnore controlled variables
vary or relationship between the variables undergoes a change.
Therefore (controls must be established to indicate the limits
within which the model and its solution can be considered as
reliable. This is called controlling.
 Operations Research (0.R) 1.7
(5) Testing the model and its solution ie.,
checking as far as possible either from thevalidating
the model :
past
by expertise and experience whether the model available data or
gives a solution
which can be used in practice:
(6) Implementation : Implement using the solution to
desired goal. achieve the
1.9 Limitation
Mathematical models which are the essence of OR do not take into
account qualitative or emotional or some human factors which are quite
real and influence the decision making. All such influencing
factors find
noplace in 0.R. This is the main limitation of O.R. Hence 0.R. is only ar
aid in decision making.
EXERCISE
1. What isO.R? [MU. BE. Oct S6]
2. What is the scope of 0.R ?
3. What are the applications of 0.R. ?

4. List the uses of O.R.

5. Write ashort note on the importance of operations research
in production management [MU. MBA April, 97]
6. Write a short note on the role of operations research in
marketing management.
MU. MBA Nov.96, Nov. 97, April 98]
7. Write a short note on the role of operations research in
production planning. [MU. MBA April, 96/
8. What are the different phases of O.R.? [MKU. BE. Nov 97)
9. What are the characteristics of an 0.R. problem ?
[MKU. BE. Nov 97|
10. What is a model ?
11. What is a Mathen1atical Modei ?

12. ýhat are the different tvpes of Modcls ?

1.8
Resource Management Techniques
13. What are thecharacteristics of a good model ?
14. What are the limitations of amathematical model
15. What are the limitations of an 0.R. Model

IMKU. BE. Nov 97)

16. Explain the general methods of solving O.R. Models.
17. Explain the principles of modelling.
18. State the different types of Models used in O.R.

[MU. BE. Oct 97:

19. What are the various phases in the study of operations
research ?
[BRU. BE. Apr 97)
20. Answer the following questions with examples wherever
necessary.
(a) Necessities of OR in industry,
(b) Fields of application of OR in industry
(c) Deterministic models,
(d) Mention atleast eight mathematical models.
[MKU. BE. Nov 9)
1. What is an iconic model in the study of operations research ?
(MU. BE. Oct. '9ó/
Chapter 2

Linear Programming
Formulation and
Graphical Method
(Formulation.and Graphical Solution)
2.1Introduction
Linear programming problems deal with determining optimal
allocations
of limited resources to meet given objectives. The resources
may be in the form of men, raw materials, market demand, money and
machines etc. Thèobjective is usually maximizing profit, minimizing total
cost, maximizing utility etc. There are certain restrictions on the total
amount of each resource available and on the quantity or quality of each
product made.
Linear programming problem deals with the optimization
(Maximization or Minimization) of a function of decision variables (The
variables whose values determine the solution of aproblem are called
decision variablesof the pYoblem) known as objective function, subject
to a,set of simultaneous linear equations (or inequalities) known as
constraints. The term linear means that all the variables occurring in the
objective function andthe constraints arè of the first degree in the
problems under consideration and the term programming theans the
process of determining a particúlar course of action.
Linear programming techniques are used in many industrial and
economic problems. They are applied in product mix, blending, diet,
transportation and assignment problems. Oil refineries, airlines, railways,
textile industries, chemical industries, steel industries, food processing
industries and defence establishments are also the users of this technique. /

2.2 Requirements for employing LPP Technique :

(BRU. BE. Nov 96]
1. There must be a well defined objective function.
2. There must be alleruative courses of action to cho0se.
 2.2 Resource Managèment Techniques
3. At least some of the resources must be in
give rise to constraints.
4. Both the objective function and
limited supply, whikh
eçuations or inequalities. constraints must be
2.3 Mathematical Formulatlon of L.P.P Yinew
Ifx, (( =1, 2, ... , n) are th¹ n decision
the systenm is subject to m constraints, the variables of the
can be written in the form: problem
and
general Mathematical model

subjectto g, (* .)
(called structural constraints) S,2b, (i=1,2, .. , m)
and
(called the non-negativity
restrictions or constraints)
Procedure for forming a LPP Model:
Step 1 : Identify the
assign symbols to them. unknown decision variables to be determined and
Step 2 : ldentify all the
factors) in the problem andrestrictions or constraints (or
influencingor
inequalities of decision variables. express them as linear equations
Step 3 : Identify the
function of decision objective or aim and represent it also as a linear
Step 4 : Expressvariables.
the complete.
mathematical model.
We consider on!y
formulation of LPP as. a general
put proper those situations where this will
inequalities in the formulation. help the reader to
1. Usage of
manpower,
equil to the availability of time, raw materials etc are always less than or
2. manpower, time, raw materials etc.
to meetProduction
the demand.
is always
greater than or equal to the
requirement so
Example 1]: Afirm
and sells them at manufactures two types of productsA anu
a profit of Rs. 2
product is processed on two on
type Aand Rs. 3 on type B. Each
minute of processing time on machines M, and M,. Type Arequires
requires 1 minute on M, and M,1 and 2 minutes on M2. Type D
avaalabie for not more than 6 hoursminute on M.
40 minutes whileMachine MË P
machine M, IS
 Linear Programming Formulation and Graphical Method 2.3
Available for 10 hours during any working day. Formulate the
problem As a LPP s0 as to maximize the proft.
Solution : Let the firn decide to produce x, units of product Aand r
units of product Bto maximize its profit.
To produce these units of type Aend type Bproducts, # requires
Nt*, processing minutes on M,
2x +*2 processing minutes on M,
Since machine M, is available för not more than 6 hours and 40
minutes and machine B is available for 10 hours doing any working day.
the constraints are
S 400
s 600

and x *2 20.
Since the profit from type A is Rs. 2 and profit from type B is Rs. 3,
the total profit is 2x+ 3x,. Asthe objective is to maximize the profit, the
objective function is maximize Z =2x,+3x.
.:. The complete formulation of the LPP is
Maximize Z = 2x, + 3x)
subject to the constraints
S 400

S 600
2x1
and X X2 2 0.

Example 2| (Production Allocation Problem)

These products are processed on
Afirm produces three products.required to manufacture one unit
three different machines. The time
each of the three products and the daily capacity of the three
of
machines are given in the table below : Machine
Timeper unit(minutes) capacity
Machine
Product 2 Product3 (Minutes/day)
Product 1
2 440
2 3
M 3 470
M, 430
M3
 2.4 Res. irce Management Techniques
It . is required to determine the number of
manufactured for each product daily. The profit
per unitunits to
1,2 and 3is Rs.4, Rs.3 and Rs.6
respectively. It is for
assumed prod
market. that all th
amounts produced are consumed in the
mathematical model for the problem. (MU. B. Tech.
Solution : Let x, x, and x, be the number ForLeathmerul. ateOct 96rtho
units of products
produced respectively. 1,2 and
To produce these amount of products 1,2
and 3, it requires:
2x1 +3x,t 2x, minutes on M,
4x, + 3x minutes on M,
2x +Sx, minutes on M3.
But the capacity of the
machines M1, Mand M, are 440,470 and 430
(minutes/day).
:. The constraints are
2x+ 3x, t 2x3 S 440

4x, +3x3 S 470

2x,+5x2 s 430
and x, , X3 2 0
Since the profit per unit for product 1, 2,and 3 is
respectively, the total profit is 4x, + 3x, + 6x, AsRs.4, Rs. 3 and Rs.0
the objective is to
maximize the profit. the
Z= 4x objective function IS maximize
+3x, + 6x3
:. The complete
formulation of the LPP is
Maximize Z 4x + 3x +6x,
subject to the constraints
2x +3x, t 2x, S 440

4x +3x3 S 470

2x + Sx S 430

and x, X, X3 2 0.
 Linear Programming Formulation andGraphical Method 2.5

Example 3 : (Blending Problem)

Afirm produces an alloy having the following specifications :
() Specific gravity s 0.98
(ii) Chromimum 2 8%
(i) Melting point 2 450°C
Raw materials A, B and C having the properties shown in the
table can be used to make the alloy,
Raw material
Property A B C
Specific gravity 0.92 0.97 1.04
Chromium 7% 13% 16%
Melting point 440°C 490°C 480°C
Cost of the various raw materials per unit ton are
Rs. 90for A, Rs. 280 for B and Rs. 40 for C. Find the proportions in
which A, Band C. be used to obtain an alloy of desired properties
while the costof raw materials isminimum.
Solution:Let x, X, and x be the tons of raw.materials A,B and (C to
be used for making the alloy.
From these raw materials, the firm requires :
0.92 x + 0.97x, + 1.04 x3 specific gravity
7x + 13x, + l6x, chromium
440 x + 490x, 480 melting point.
p
:. By the given specifications, the constraints are
0.92 x + 0.97 x, + 1.04 x S 0.98
7x+13x + l6 x3 8
440 x +490 x, + 480 x; 2 450

and x 2, X3 20.

Since the cost of the various raw materials per unit ton are : Rs. 90 for
A, Rs. 280 for B and Rs,. 40 for C, the total cost is
the
90x,+ 280x, + 40x. As the objective is to minimize the total cost,
objective function is
Minimize Z = 90 x + 280 x+ 40 x

. The complete formulation of the LPP is

Minimize Z = 90 x + 280 x, + 40 x3
2.6 Resource Management Techniqucs
subject to
0.92 x +0.97 x, + 1.04 x S 0.98
7x +13 x2 +16x 28

440 x +490 x, +480 x 2 450

Example 4:(Diet Problem)

Aperson wants to decide the constituents of a diet
his daily requirements of proteins, fats and which wil fur
minmum cost. The choice is to be made from four
foods. The yields per unit of these foods are given in
carbohydrat es at t
different types
table: the fol owin
Food Yield/unit
type Proteins Fats cost/unit
1 3
carbohydrates (Rs)
2 6 45
2 4 2
40
85
6 5 4 65
Minimum 800 200 700
requirement
Formulate the L.P model for the problem.
Solution : Let x,x, X and x, be the units of food of type 1, 2, 3and
4used respectively.
From these units of food of type 1, 2,3 and 4 he
requires
3x, + 4x, + 8x, + 6x4 Proteins/day
2x +2x,+ r, + 5x Fats / day
6x + 4x+ 7x, 4x4
Since the minimum
Carbohydrates/day
requirement of these proteins, fats and
carbohydrates are 800, 200 and 700 respectively, the constraints are
3x + 4x + 8x, + 6x 800
2x + 2x,+ 7x, + 5x, 200
6x + 4x,+ 7x + 4x, 2 700
and x,X, X3, X4 20.
 Linear Programming Fornulation and
Graphical
Since, the costs of these food of type 1.2, 3 and 4 Method
2.7

Rs.85 and Rs. 65 per unit, the total cost is are Rs. 45, Rs.40,
Rs.45x t40x, +85xgt 65x4. As
the objective is to minimize the total cost,the
objective function is
MinimizeZ = 45x +40x, + 85x, +65x4
.: The complete formulation of the
L.P.P is
Minimize = 45x,+ 40x, + 85x + 65 x4
Z
subject to 3x +4x,t 8x3 6x4 800
2x +2x+7x, + Sx4 > 200
6x +4x t7x +4x4 700
and x, X 3, 4 20.
Example 5|: A farmer has 100 acre farm. He can
sell all
tomatoes, lettuce, or radishes he can raise. The price he
Rs.1.00 per kg for tomatoes, Rs. 0.75 a head for lettuce can obtain is
and Rs. 2.00
per kg for radishes. The average yield per -
tomatoes, 3000 heads of lettuce, and 1000 kgs of acre is 2,000 kgs of
radishes. Fertilizer is
available at Rs.0.50 per kg and the amount required
kgs each for tomatoes and lettuce, and 50 kgs for per acre is 100
required for sowing, cultivating and harvesting perradishes. Labour
days for tomatoes and radishes, and 6 man-days for acre is 5 man
400 man-days of labour are available at Rs. lettuce. A total of
20.00 per man-day
Formulate this problem as a LP. model to maximise the farmer's
total profit.
Solution:Let the farmer decide to allot x1, x, and x acre of his farm
to grow tomatoes, lettuce, and radishes respectively to
profit. maximize his total
Since the total area of the fom is restricted to 100 acre
and the total
man-days labour is restricted to 400 man-days, the constraints are
S 100
Sx + 6r, + Sx s 400
and x, X2 20
The Farmer will produce 2000 x, kgs of
lettuce, and 1000 x kgs of radishes.
tomatoes, 3000 x, heads of
.. The total sale of farmer will be
Rs [1 o 2000 x, + 0.75 o 3000 x, +2 o 1000 xl
Fertilizer expenditure will be
= Rs 0.50[100 x + 100 x, + 50 x;)
 hen more Rs.
x
x\-50 and the for 20 young
1875x, x, A spend than costs
120 1875 each andpaise. of
Rs5x,]
be +
x,
Xn-50 profit x,
more and x)l
week to hen
x x,1 + X,+ 2 +
+ -100 X Rs. 30 Rs.80 old 1(x1
+6x, 2000 1875 2080 2080
a house hens
at per worth give hens,
expenditure]
- is old
and
Sx,)-
25x, L.P.P+ bought
eggs only hens 20 5x) 0.lx)
[5x
Techniques
Management
Resource +
x,
+ being to cannotx,
20 x
2250 2080 1850 x
the 1850 3 hasbuy buy for 80
than
20 + x) +
-
s S 400
100 s be layegg he spend s mores (3x t [0.3(3x, (0.5x,6.2
total + of 0. can hens person he to
be (x,
0.3 will
+
= be x
= formulation
x isfunction = each should
that decides tx,be Rs.
house Rs.l Rs. Rs. 0.5x,
-0.1
x1
will - 2000 1850 hens to Sx
will profit(Sale Z Z constraints5x X3
MaximizeMaximize old A assuming 80 each,+ will
week, kind L.P.P. x = Expenditure
on
feeding
= = -
expenditure
Rs Rs [ t X, The feed.
6:Old person Rs. 2x not eggs
Rs 6x,
x, is
net = objectivecomplete + and perto each a
each. maximize
his
profit.
5
onlyRs. can
of profit
Farmer'=s = the Sx
as
eggsweek veek,
The
has
he sale
tosubject Example
Rs.5
of thisSolution: costs since total net
Labour The The lay5 many he
per perFormulate Since hen The
at Also, The
.:. 1 HowRs.6
onesonesRs. young .:.
2.8
2.9 twoB with Profit 95) of to he
and profit.
paise. sixtecn 5-days
and number
bottles bottles,
has types bottles ounces Nov
Method
plantbottles week.
25 his
75/minute 7000 BE. x, 16-ounce
day,
40/minute is 16-ounce
both 16-ounce bottle maximize
3,00,000 produce
[BRU. per
drink&-ounce per and
:
Graphical on available days A B 3-hours
machinemachine
16-ouncebottles
soft used
5 exceed to, to of and
wishes decide 8-ounce
units
and is =0.5x,
-0.1x|
L.P.P a for be is day,
eight-ounce
fur
on available
in designed on
data of
Formulation 8-ounce per on cannot
plannercan bottles plarner minutesminutes
80 20S 6. 80S 20 26.0. number
the eachfollowing 8-hours planner
100/minute
60/minute
and of
S 2 of s > is drink units are
formulation production
Howevér, paise 25,000 productionx,
Progra1nming
2x
+ x and
Sx 0.1 xX2
0.5x, x and
0.5x,
-0.l x1,
X) A
B. The
and x,
40
100 B
and
Z constraints
Maximze The run
the L.P.P. and
are
constraints
The and 15 week.
60 A
+
efficiency. be is of absorb bottles x, machines
complete A A bottles. can production the these
bottle peras a profit.
Linear the 7: machines Machine machine can Let 8-ounce have
this:
bottles produce
to Example16-ounce
of 8-ounce Solutionmaximize
his the week,
per
we
The subject A B marketFormulate
loss Since
9ottling The Weekly of To requires
ome ounce units
'or on the
 eight-ounce objective
16-ounce
exceed
4,80,000s7,20,000
not on the
25,000
should andAs =0.15x
+0.25*,+0.25
x
s
5x, have 0.150.25x,.
4x, s3,00,000 s7,20,000
s4,80,000 s3,00,000 is
drink upto we L.P.P0.15x s4,80,000
s7,20,000
3,00,000
s
Resource+ +
Techniques
Management
2r week, 7000 S
25000 7000 Rs. is
+ function 25000
S 7000s
5x the only 0. is Rs.0.15x, the
bottle 0.
=> > of abSorb per s s of =
2400 2400 production objective formulation + .
&x
16x,
+
bottles x,
and Sx Sx+
+ 4x, 16x x
x,
8-ounce
is
Maximize
Z Maximize
Z 2x5x, +
4x, 8x
+
1642 and
x,
X)
s s can constraints
are: 2x1
The
constraints
s25,000 profit Sx
sixteen-ounce and the
market
the on
40
100 weekly total
profit profit;
complete
the
x the to
the the the The
Sinceand 7.000 Rs.0.25,
Since "maximize
subject
ounces
2.10 and is
 2.L1 B. TheA of leather are for performs
combined). data 100,the Corn, preparation, fatmer
of andprepare, work. A of The for mustfor to
section
components
and product were
available maximize acre Economy. 6000 setsas
day relevantunit 0.0 0.1 painting grow work company
quality. Testing
of An The of TVso
Method A products per to sections, L.P.P
products supply and
B each
for of man-days Rs. Economy
Each B can Rs.20.
Rs.70
and400day for30. nan-days
lower The 600,to the : SectionB average
follows the
Graphical only models
A schedule Rs.
he Rs.100 and a
all
product. The A a Painting
required Assembly assembly that
of
leather if (Bothparts
testing. 0.2 0.2 which profit costs profit assemble
Standard 6 as
is andday. and 8000 sections 40
given
B costs abeans
as this
and spare 10 is aind
product
per B perdaypart production
on a a Formulate
Formulation
EXERCISE
of product
twoand Hours :
are land
requires
yields and two section
yields
to Section
A is standard
corn
types Rs.3 1000 per spare
special
L.P.P. Assembly Assembly
producing 1.0 1.5 available
soy usedColour,
operates twO It
andand make products of of and of andpreparation each 3 2 B.
prepare,, for section
two as speciala weeklyacres acre acrework is pending,
Programming quality
Rs.4time
700
as work section
company : on costs 16colours,
makes could operations--painting, hours 1,000 Each An televisions
production
Model
T.V. Standard for
800 onlythis firm
engaged
in Unit
sale L.P.P Economy
much PriceRs.50
Rs.80 Determine of of Colour running
Rs.4000 cost.
highare company a Formulate to
man-days man-days
Rs.40. for Each
only are
requires of has soyabeans.
Rs.120 is
Linear company profits
as number
a television
Rs.1,00,000
as
24 total
order
of There atleast
is twice for follows:
as
are farmer profit this B. of daily daily
and
A thesufficient costs section
types the
respctive A productB.
available. Model 30. 7 8 Formulate an minimize
A Product a requires expected produce
requires Product
B, 2.A Total
profit.
testing A
or requires
wheat A TheA Section Which
1. type three A B
3.
wheat yields 4. three
is
has and
 for
hour April,
A, hours per
P drinks and P of of
D. figures
of 1 will
at
locatedbottles
plants B. mg. Cost
A whereas
andmodelSpecially units
models B days and
2 require of of is stuff
the the
company
of
L.P.P. soft cost V
corresponding
manufacturing, month
bottles V 1 minimum
Rs.15
four onthe dozen hours one of running
many A and
V1, are V3
that of Find
food
different production
stuffs
packing. D 20 a plants,number 1000 1000 6000 vitamins
A Rs.for
manufactures
Rs.8, a model as
the the16000How
V3
3 mg.
Rs.1.5.
is
time for manufacturing, of food of 100
of week, this Plants during day
ofDloss
and three in 4000. mg. atleast
contributions 1 andof Formulate
two
bottling A, the of vitamins Supposcand
Resource of forof1AThe
Techniques
Mnnagement a C
B amount
a type 2 units
forthcoming plants
3000 1000 2000 of per
that types
these
10 V, B
produces
company andhourpacking.
and and P
that
bottles Rs.
costs of and
of
two of thatbody
assembling dozen
respectively
certain 1
of time.
packing
indicatesandso different
operatingApril
contain
requirements
Va. V, mg. andthe
profit require2,
for Barea1, hours
has the
plant Products 24000
Rs.6000 of of 10 supply
Radio the firm of
C. in stuffs mg. , Rs.1
requires hourmodelwhileDuring
B C mg. V
Eachcapacities
A
for
survey plant two
15 drinks The 10 of is
transistor
have for
and components available demand are
each
fooddaily 100
Consider and mg. A
would
each
vitamin.
modeland of Q. stuff
which radio assembling of and
3,5 onty. hours softat The
:
follows market Ç. respectively these V, V, I
Minimum that
units of run contains
A, packing other a
be bottles firm 7. of of food
of make10
and A A will mg, mg. diet
models
of units dozenCare 6. as that
the andare
of
S.
122 of to
 98] the objective
yalue mustthe
percentage, period
function the of profi
nanely
objective be pictorial
basicfiniteuse.system.be
get programming
as Butthedealline
2.15 function
determine can can
solutions variables.
to
Nov of all
activities the to
any analysis L.P.P can a
read.
MCA.basis
the for values. the variables easy method as
shown
the to activity objective number. a gives the
provides involve
Mcthod during objective to
some
the in proportional integer we which
and from
infeasible
understandSensitivity concepts difficult
of
graphical
two
(MU.on variable different
percentage.
thechangein problem two eliminated linear constraint is
Graphical
formulated
by each L.P.P whichandmay changes. than
increases to in
finite the only graph
each by restricted efficients (optimize) which
solutions of morebasic
Though
and
directly by used
not are programming a
of involving
method to
analysis.
and
automatically tool each the
same
of availability used do constraints one
solution L.P.P. simplesolutions thepowerfulinvolve
the
andconsidered.
this.sincemake
Formulationare contribution resources
graphical explain
the resources not Co only of
problems isresources collectively.known extremise its
graphical graph
by are minimization.
problemsand generalis do with but lines
increase
:
Deterministic
variables and linear involves the problems
methodunbounded situations constraints
the
resource of problems
completely drawing familiarnumber
are not to
a
Programming
programming
The
Proportionality:
1. the quantity :Variables of useful of
Thisconstraints
2.4
Basic
Assumptions the to Programming solving in
of method a clearly is
of also Sum of a
In as
solutions, practical method
assumptions.
following if The
individually problem by really
usage shall total the :
so cost the variables. by
solved not large
i.e., : : or are
Divisibility variables
Optimality in very of
number
Linear
linear its variable.
Additivity
output the
all
Certainty Finiteness
constraints or Graphical representation
of
used redundantillustrated
easily
Graphical are
or to
in
study The
maximization
effectively highlighted the
of
is
aa
who graph,
method
resources Linear
numberMultiple
decision of
Concepts any
The functionthe equal LPP.
the persons
the 2. 3. 4. under 6. 7. 2.5
of then be and 5. The most
the with
 i= and solutionn which the the
the lines
equality
equalities. for
permissible if (or
the 0, is for conver
vertex region Convex Not
convex
or which bounded
the vertex
to x, each region
restrictions
&Ax)
subject of the be the
as variables equations
the choose
choose to within
plane
corresponding
to (feasible
constraints the of
:method vertices said
Rescurce fx)non-negativity
of problem, completely
Techniques
Management the intersection problem,
the is
region points Convex
Graphical
= Inequality of 1. solving all
Z valuesstep at minimization
optimize permissible Z maximizationof lies
in of by of set
the thedrawnpoints
region)values points
for XOY a
procedureand Considerrestrictions.
LP.P, negativityfor
region)
2 lines the
or
the For For regionits Convex
Not
convex
the :the
b, 1,2,3......m
j= Find
Step Findpermissible
Find of
in the () (ii) two
:Example
a =, :1Steplines 4: 5: maximum.A
Working
Given by 3:
convex : any
S, straight bounded
Step Step Step minimum.
Notejoining
the lines.
2.16 or of is
 2.17 94] Theon - theotherconstraints
we as the
method inequality
(0,0),
Nov 0) (-0.5,0).satisfyknown
(-0.5, all
Methodgraphical
BE. equalities. origin satisfies
line also
[MU. the
=-0.5
x’ and Now this
Graphical as (1) (2) (3) (4) (5) satisfies is area
constraints (0,1) on region
the =1 Similarly
interpreting
the this =1
+,
-2x, +=3
3x,
+
2x, ’(0,1) and
by x, also
points within
feasible
and
L.P.P + side x=2
Formulation 3 0 inequality -2x,it origin
1 2 1 2 3 =*}t 1,
= the <1;hence
1-2x point
following
= S S
Max
Z s . = = =1 x, equation The 3 2.1
Iig.
+x, X+x, through the
x X T
=
and
x1
=
Every =0
Programming -2x, x, the
-2x1 ’x=0 on OABCD.
’ put
the andconsider the=0points L.P.P.
passes
=0 +0 S1.
satisfy
Solve
x (1) gives-0 all X, region
the 3 D1
First 2
Linear1: put line Thus+ of
line,
For
the line -2x,feasible
Exampleto
subject Solution:
this s1. space
the substitution, constraints.
on inequality
So, points + x thesolution
2x get
 Solving pointsIr o
is Similarly. (2,0), I2.
points get all
B 0.
space. 2x,) optimum graphical 2 the the S we satisties
+these
two A 4x, the the the
constraints,
3. ,
(0,0), + x1, as All + AIl
solutionB(2,1). Z=3x 5x, -3x, as plane.
plane. satisfy area
the the and12
Solving x,
3. O + S inequality
12
vertext are by type, =1. by 3x 12 4x, x, x, lineother
this
the given : x, L.P.P= 12 4
2 2
0 2 0 in
Techniques
Management
Resource
of x space maximization
=2, S 2 + x, x, this
3x,the the with
vertices the and the
are
Value
ofZ Z -3x+
following4x
Minimize 2x1-X) -3x, the
2x,+
3x in
oninterpreting
x, have - | solution linesatisfy
in point
vertices
20 3
8,x consider
t line and
the x,and We 0 6 2 givesa 2x a origin
line gives
Every
Xy of 2= the
Maximum
find the these Similarly
this constraint ABCDE.
=2 t is Solve
. at Vertex (0,0)A(2,0)B(2,1)
of us it it the
to I. -2x, vertices(0,1)
and
D D(0,1)problem thenon then
is x, Z Let and of
aimof =
intersection
x, of of O 2: 12. 12, 12. region
values the Example subject
to : the side3x,2
= sideconsider
our 2. Solution =
Now = intersectin
have
we
The Sincethe
L.P.P
is 4x origin 3x other teasible
x The + 2x+
+
2.18
have -3x the 2x the
 Linear Programming Formulation and Graphical Method 2.19
constraints. The vertices of the solution space are A (3, 2), B (4, 2),
C (4. .oD.). and E
.).
2r, --2

-3r, +4x=12

2r, + 3r, = 12
=2

-3 -2 3
=0

Fig 2.2

The values of Z at these vertices are given by

Vertex
Value of Z ((: Z=3x + Sx,)

A (3, 2) 19
B (4, 2) 22
42
C(4,6)
102
D) 5
79
4

solution is
Since the problem is of minimization type, the optimum
Minimum Z = 19, x =3, x, =2.

Example 3 : A pineapple firm produces two products canned

material, labour
Dineapple and canned juice. Thespecific amounts of
availability
Ind equipment required to produce each product and the
given below :
»f each of these resources are shown in the table
 Resource ManagementTechni.
Canned
2.20
Canned
Juice
3
Pineapple
2.0
Avaresour12.ilab0clee
Labour(Manhours) 2.3
Equipment(M/chours) 6.9
1.4 4.9
Material(Uni) canned
of canned juice and
Assuming one
margins
unit
respectively. Formulate this
Rs.2 and Rs.1
Pineapple
(MU. BE. Non
h
profit graphicallyalso.
and solve it the number of units of canned juice x be s and
Solution: Let x, be produced.
be
canned pineapple toin
number of units of this problem are the
constraints or restriotions laben
The and material.
equipment
s 12
For labour, 3x, +2x2
S 6.9
For Equip1ment, x +2.3x,
x, S4.9
For material, x, + 1.4
and x,*, 2 0.
the.profit.
Here our objective is to maximize
MaximizeZ = 2x +x,
.:.The objective function is L.P.P.is Maximize Z = 2x, +r
:.The complete formulation of the
subject to thc constraints
2x, s 12
3x +2x,
- + 2.3x, ss 6.9
x +1.4 x, S4.9 and x, x, 2 0.
vertices
The solution space is given below with the shaded area with
(0,0), A(0, 3). B(1.8, 2.2), C (3.2, 1.2) andD (4,0)

3z, + 2a, =12

A,+1.4, = 49

Fig. 2.3
 Linear Programming Formulation and
The values of Z at these Graphical Method 2.21
vertices are given by
Vertex Value ofZ ( Z 2x t x)
O(0, 0)
A(0,3)
B (1.8, 2.2) 5.8
C(3.2, 1.2) 7.6
D (4,0)
Since the problem is of maxímization type, the optimum solution is
Maximum Z = 8, x =4, x, =0.
| Example 4: A Company manufactures 2 types of printed
circuits. The requirements of transistors, resistors and capacitors for
each type ofprinted circuits along with other data are given below :
Circuit Stock available
A B
Transistor 15 10 180
Resistor 10 20 200
Capacitor 15 20 210
Profit Rs.5 Rs.8

How many circuits of each type should the company produce

from the stock to earn maximum pro fit. [MU. BE. Oct 957
Solutio:Let x, be the number of type A circuits and x, be the
number of type B circuits to be produced.
To produce these units of cype A and type B circuits, the company
requires
Transistors =
15x + 10x
Resistof =
10x + 20x
Capacitors I5x + 20:x
Since the availability of these transistors, resistors and capacitors are
180. 200 and 210 respectively, the constraints are
15x +I0x, s l80
10x, + 20x, S 200

15x, +20 x s 210

and x 20. X2
 ResourceManagementTechniques
and from type Bis Rs&
222 type AisRs.S
theprofitfrom
Since &r,
profitis Sx,+ L.PPis
thetotal formulation ofthe Sx +&r,
Thecomplete Maximize 2
ISt,+ 10r, s 180
subjectto s 200
10x, +20r,
15x,+20r, s 210

and ,X2 2 0.
method, the solution space is given belov
graphical
By using0ABCD with vertices 0(0, 0), A(12, 0), B(10, 3). C0
shaded area
and D (0. I0)

24

16
ISs,+10xs I0

D 1Ss, +201,210

Fig. 2.4
The values of Z of these vertices are given by
Vertex i Value of Z (:Z=5x, + 8x,)
O(0.0)
A(12, 0) 60
B(10, 3) 74
C(2,9) 82
D (0,10) 80
solutionI5
Since the problem is of maximization type, the optimum
Maximum Z = 82. X = 2, X, = 9.
 Linear Programming
Formulation and Graphical Method
|Example 5]: Apply graphical methodto solve the 2,23

Z=x-. 2x, Subject to x tx, S 1, 6x +4x, 2 LPP: Maximise

2sx,s4. 24, 0sx S5 and
Solution : By using graphical method, the IMU.MBA.Nov.96/
below with solution space is given
shaded area ABCDE with vertices A2). B (5, 2),
C(3,4), D(3,4) and E(2, 3)

Fig. 2.5
he values of Z at these vertices are given by
Vertex Value of Z (: 2=x -2r,)
4
-

B(5, 2)
C(5,4) -3
D (3, 4) -5
E(2, 3) -4

Since the problem is of maximization type, the optimal solution is

Max1mum Z = 1, x =5, x, =2.
224 Resource Mnnagement lechniques
IEample 6]: ACompany manufacturee
two types of eloth.
thrte different colours of wool. One yard length
rrquire 4 02 of red wool, $ 02Z of green wool snd of3 A type
wool. One yard length of type Bcloth requires $ 02. of
0Z of green wool and 8 0Z of yelow wool. The
manufacturer is 1000 02, of red wool, 1000 (02 ofwool
1200 0Z of yellow wool. Theemanufacturer ean make green wont availahte
on one yRrd of type Acloth and Rs.3 on one yard of profit of
Find the best combination of the quantities of type A: type Ren
which gives him maximum profit by solving the and type Bel
L.P.P. graphically
Solution : Let the manufacturer decide to (MU. BE. Apr
cloth and x, yards of type Bcloth.
produce X yards of ty
To produce these yards of type Aand type B cloth, he
requires
red wool = 4x +Sx, 0z
green wOol =
yellow wool = 3x + 8x 02

Since the ávailability of these red wool, green wool and

yellow woc
are 1000 0Z, 1000 0Z and 1200 OZ
respectively, the constraints are
4x| + 5x2 s 1000
Sx, + 2x, s 1000
3x + 8x s 1200
and x 2 0, X 2 0.
Since the profit from one yard of type A cloth is Rs.5 and the proti
from one yard of type B cloth is Rs. 3, the total profit is Sx + 3x2
The complete formulation of the L.P.P. is
Maximize Z = 5x t 3x
subject to 4x +5x2 1000

S 1000
3x, + 8x s 1200

and x *2 2 0.
 Lincar Programming Formulation and Graphical Method 2.25
By using graphical method, the teasible region is given below with
shaded area OABCD with vertices O(o. 0), A(200. 0), B(17
(2000 1800
c(0017 17 and D (0. 150)

S00
400
Sa, +2x, 1000
300

4x, +5x,= 1000

100 3x, +8, 1200

100 S00 400 s Q 600

Fig 2.6
The values of Zat these vertices are given
.y

Vertex Value of Z (': Z= 5x, + 3x,)

O(0, 0)
A (200, 0) 1000
(3000 1000 1058.8
B 17' 17
(2000 1800\ 905.8
c 7 17 450
D
(0, 150)
is
Since the problem is of maximization type, the optimum solution
3000 1000
Maximum Z 1058.8, x = 11 2 17

Note : In' agiven L.P.P, if any constraint does not affect the feasible
region (or solution space), then the constraint is said to be a redundant
constraint.
2.26 Resource Management Techniques
Example 7: ACompany making cold drinks has
plants located at towns T; and T,. Each plant produces two bot
A, Band Cand their production capacity per day is given di three
Cold Plant at belcw:
drinks
A 6000 2000
B 1000 2500
C 3000 3000

Themarketing department of the company forecasts a demand:

80,000 bottles of A, 22,000 bottles of B and 40,000 bottles of C dur
the month of June. The operating costs per day of plants at T, and 1
are Rs. 6000 and Rs.4000 respectively. Find graphically, the numbe
of days for which each plant must be run in June so as to minimiz
the operating cósts while meeting the market demand.
Solution:Let the plant at T, and T; be run for x and x, day
respectively.
Since the plants at T, and T, run for x and x, days, they will produc:
6000 x, +2000 x bottles of A

1000 x +2500 x bottles of B

3000 x +3000 x bottles of C

are 80,000, 22,00

Since the demand for the cold drinks A, B and C
always greater than or equu
and 40,000 respectively and the production is
to the demand, the constraints are

6000 x + 2000 x, 80000 > 6x,+ 2x, > 80

3x tx, > 40

1000 x t 2500x, 2 22000 ’ X + 2.5r, >22

+ 3000 x2 2 40000 > 3x, + 3x, 2 40

3000 X1
2 0.
and
 Lincar Programming Formulationand
Since the operating costs per day at T, isGraphical Method
2.27
R's. 6000 and at l2 1S
Rs. 4000 and T, l2 run for x and x,
days,the total operating cost is Ks.
6000x +4000x.
Here our objective is to minimize the total
the objective function is minimize Z = 6000 x +operating cost. Therefore
4000 *,
.:. The complete formulation of the L.P.P. is
Minimize Z 6000x,+4000x,
subject to 2 40

* +2.5x, 2 22

3x +3x 240
and x X2 20.
By using graphical method, the feasible region is given below with
shaded area with vertices A (22, 0), B (12, 4), C(0, 40)

40

30

25
20

20
=0

Fig 2.7
constraint 3x,+ 3x, 2 40 does not
From the figure, we see that the
the solution space. So 3x, + 3x, 240 is a redundant constraint.
affect see thát the solution space is
Also from the direction of the arrows, we
unboundedabove. are
B(12,4) and C(0,40)
The values of Z at these vertices A (22,0),
given by
 ResourceManagement Techniques

Value of Z ( Z=-6000x
2.28
+4000x,)
Vertex
1,32,000
A(22, 0)
88,000
B(12, 4)
1,60,000
C(0,40)
s of minimization type, the optimum solution
is
the problemis
Since Rs. 88,000, x = 12 days, x, =4 day,
Minimum Z =
problems
Note : From
the above examples, for observed that
solution, we
involving
the optim
th
and having a finite That is, "if there exiy
variables the feasible region.
existed at a vertex of
solution
of an ;.P.P, i# will be at one of the vertices of th
solution
an optimal
feasibleregion".
cases
2.6 Some more feasible solution which m
give region of
The constraints generally, discussed seven linear programmi
We
be bounded or unbounded. solution for either of them was unione
problems and the optimal problem. In general, alinea
for every
However, it may not be true
programming problem may have :
(i) a unique optimal solution
solutions
(ii) an infinite number of optimal
an unbounded solution
(üi)
(iv) no solution
illustrate the cases.
We now give a few examples to
A and Bon
Example 8|: A firm manufactures two products
profits earned per unit are Rs.3 and Rs.4 respective
Product
which the and M.
Each product is processed on two machines M,
two minutes on
requires one minute of processing time on M, and Machin
M, and one minute on M: machin
while Brequires one minute on M while
M, is available for not more than 7 hours 30 minutes Findth
M, is available for 10 hours during any working day.
manufactured to g
number of units of products Aand Bto be Solve b
LP.P. and
maximum profit. Formulate the above as a
BE.
graphical method. [MU. BE. Apr 92, MSU
 Linear Programming Formulation and Graphical Method 2.29
Solution : Let the firm decide to manufacture x, units of product A
and *2 units of product B.
Toproduce these units of products AandB, it requires
* tx hours of processing times on M
hours of processing times on M,
But the availability of these two machines M, and M, are 450 minutes
and 600 minutes respectively, the constraints are
S 450

2x1 S600
and x, X2 20
Since the profit from product A is Rs.3 per unit and from product B iS
Rs. 4 per unit, the total profit is Rs. 3x, + 4x, and our objective is to
maximize the profit
.:. The complete formulation of the L..P.P is
Maximize Z = 3x, t 4x)
subject to S450 ... ()
S 600 . (i)
...(iii)
and x, *2
space satisfying the constraints
By graphical method, the solutionrestriction (i) is shown shaded in
(), (ii) and meeting the non-negativity
the following figure.

600

2r, *x, =600

400
300

200

l00

1o0 200 30) 40 K00 600

Fig. 2.8
 Resource Management Techniques
2.30
The solution space is the region OABC, The vertices of his
0), B(150, 300) and
space are 0(0.0), A(300,
The values of Z at these vertices are given by C(0,450) sols
Vertex Value of z
O(0,0)
A(300, 0) 900
B(150, 300) 1650
C(0, 450) 1800
Since the problem is of maximization type and the
Zis attained at a single vertex, this problem has a maxiunimqumue valy
solution. optim
:. The optimal solution is
Maximum Z = 1800, x =0, x) = 450.
Example 9:Solve the following L.P.P. graphically.
Maximize Z
100x1 +40x,
=

subject to Sx + 2x S 1000
+2x2 s 900
+22 S 500
and x, X2 20.
Solution : By usinggraphical
shown shaded in the following figure. method, the solution space 0A

500 -
Sk,+2x, =1000
400

00 C Î, +2r, =900
200

100 sih K +2r, =S00

100 20 30 400 S00 SO0

,0

Fig. 2.9
 Linear Programming Formulation andGraphical Method 2.31
The vertices of this convex region are O (0, 0), A (200, 0),
B(125, 187.5) and C(0, 2s0)
The values of Z at these vertices are given by
Vertex Value of Z i (: Z= 100x +40x)
o(0, 0)
A (200, 0) 20,000
B (125, 187.5) 20,000
C (0, 250) 10,000
Here the maximum value of Z occurs attwo vertices Aand B.
same
Any point on the line joining A and B will also give the
maximum value ofZ.
there are
Since, there are infinite number of points between any ponits, the'same
B gives
infinite number of points on the line joining A and
maximum value of Z.
Thus, there are infinte number of optimal solutions for this L.P.P.
optimal solution is said to have
Note : An L.P.P having more than one
That is, the resources can be
alternative or multiple optimal soutions.
the profit.
combined in more than one way to maximize
solve the following L.P.P.
Example 10]: Using graphical method,
Maximize Z 2x + 3x,
subject to
+x 4 and x, 20.

4o0

Fig. 2.10
2.32 Resource Management Techniques
Solution : By using graphical method, the
the following figure. solution space is ha
Here the sclution space is unbounded. The
region (in the finite plane) are A(3,1) and B(0,4) vertices of Ihe fen
Valule of the objective function
2-2r, + 3r, at these
ZIA)= 9and Z(B)- 12
But there are points in this convex region for which 2.
vertie
higher values. In fact, the maximum value of Z0ccurs . wil have
at
this problem has an wnbounded solution., infinity Ha
Example 11:Solve graphically the following L.P.P.:
Maximize Z =
subject to

.. (i)
and x, x) 2 0 (ii)
Solution: Any point satisfying the non-negativity restrictor
(ii) lies in the first
quadrant only. The two solution
(9, and other satisfying ()are shown shaded in the spaces, one satisfy
following figure.

4,=0
3

Fig. 2.11
There being no point, (*, y) common to both the shaded probletregio
That is, we can not find a convex region for this problem. So the
cannot be solved. Hence the problem have xo feasible solution.
 Linear Programming Formulation and Graphical Method 2.33

2.7 Advantage of Linear Programming :

1. It provides an insight and perspective in to the problem
environment. This generally results in clear picture of the true
problem.
2. It makes a scientific and mathematical analysis of the problem
situations.
3. It gives an opportunity to the decision maker to formulate his
strategies consistent with the constraints and the objectives.
4. It déals with changing situations. Once a plan is arrived through the
changing
linear programming it can also be revaluated for
conditions.
maker makes sure that
5. By using linear programming the decision
he is considering the best solution.
[MU. MCA. Nov 98)
2.8 Limitations of Linear Programming :
programming is that it treats all
1. The major limitation of linear isnot true in many real life situations.
relationships as linear. But it
LPP would be meangingfulonly if
2. The decision variables in somesome times we get fractional values
they have. integer values. But meaningful.
optimal solution, where only integer values are
tothe
parameters in the linear programming model are assumed io
3. Allthe But in real life they may not be known
be known constants. probabilistic and they may be liable for
completely or they may be
changes from time to time.
complex if the 'number of variables and
4. The problems are
constraints are quite large. objective problems,
Linear Programming deals with only a single more than one
5. real life situations, there
may be
whereas in
objective.
EXERCISE
essential characteristics and limitations of Linear
1. Explain the [MU. MCA. Nov 98]
Programming Problem.
feasibility region in an LP problem ? Is it necessary
2. What is
be a convex set ? MU. MCA. Nov. 97)
that it should always
redundant constraint ? What does it imply ? Does it
3. What is a solution to an LPP ? [MU. MCA. Nov. 97]
affect the optimal
programming problems.
4. Explain the advantages of linear method of solving a LPP.
S. State the limitations of the graphical
(BRU. BE. Apr. 97, Nov. 97]
2.34 Resource Management Techniques
6. Solve the following by graphical method
Maximize x-3x,
subject to XË t S 300
-2x s 200
21 2 100
y s 200
and X1 20.

7.Bv graphical method solve the [MU. BE. Apr

Maximize Z =
following problem :
subject to 5x, + 4x S 200
+5x4 150
Sx +4x, 100
8x +4x, 80
and x 2 0.
8. Solve the
[MU. BE. Nov
following problem by graphical method:
Maximize Sx+ 8y
subject to the constraints
3x + 2y s 36
x +2y s 20
3x + 4y s 42
and x 0
[MU. BE. Apr
9. Maximize 2x
subject to 3x, + 2x S 12.0
x + 2.3x, S 6.9
x + 1.4x, s 4.9
and x Xy 2 0
Using graphical method. (MU.BE. Apr 9
10. Using graphical
method tosolve
Minimize Z = 3xL.P.P:
+2x,
subject to
10
6
x +4x, 12
(MU. BE Nov
 Lincar Programming Formulation and Graphical Method 2.35

11. Solve the following problem graphically.

Maximize Z
subject to 3x, + 2x S 12.0

and x1, x 20 [MU. BE. Apr 90, Apr 91/

12. Use graphical method to Maximize Z = 6x + 4x,
subject to

3x, +2x, S 9
[MU. BE. Apr 90]
Minimize Z
13.
subject to the constraints 300
X +y s
x-2y s 200
2x +y 2 100
y 2 200
and x p 0
[MU. BE. Apr 93/
by graphicalmethod.
of
units of vitamin A per gramn and 7 units
14. Egg contains 6 units
cost 12 paise per gram. Milk contains 8
vitamin B per gram and units of vitamin B per gram,and costs
and 12
of vitamin A per gram requirement of vitamin A and
The daily minimum optimal
20 paise per gram. units and 120 units respectively.Find the
vitamin Bare 100 [MU. BE. Nov 93/
product mix.
graphically the L.P.P
15. Solve Maximize Z = 3x, + 4x2
constraints
subject to the
2x, + 5x S 120
4x,+ 2x s 80
2 0 [MU. BE. Nov 92/
and x: X)
require twice
2 types of hats. Each hat A
16. A company producessecond hat B. If all are of hat B
only, the
as much labourtime as the li:nits
produce a total of 500 hats a day. The market
company can 150 and 250 hats. The profits on
hat A and hat B to
dailysales of the to get
respectively. Solve graphically
hat A and Bare Rs. 8and Rs.5 |MU. BE. Apr 95}
the optimal solution
2.36 Resource Management Techniques
17.Solve graphically the following L.P.P:
Maximize Z
Sx, +3x,
subject to constraints
3x, + Sx SS 115
S 10
and x 20

18. Solve graphically the following L.P.P. [MU. BE. A

Minimize Z
subject.to
20x +10:x,
x +2x, S 40
2 30
4x, +3x 2 60
and 20
19. Solve
graphically the following L.P.P.
subject to Max Z = 3x + 2y
2x+3y s 9
X -5y -20
and x,y 0
20. Solve
graphically the
Minimizefollowing
Z L.P.P:
subject to - 6x1 -4x
2x, +3x, 30
3x+2x, S24 : 9 i
3
and
21. Solve
graphically the
Maximizefollowing
Z L.P.P.
subject to 3x -214
4
and x 0
22. Solve
subject to
graphically the
Maximizefollowing
Z L.P.P.
-3x, +x, 23
and x, y 0
 Linear Programming Formulation and Graphical
23. A manufacturer of furniture
Method 2.37
makes two
tables. Processing of these products is done on twoproducts, chairs and
A chair requires 2 hours on machine A and 6 hours on machines A and B.
table requires 5 hours on machine A and no time on machines B. A
mnachine B.
There are 16-hours of time per day available on machine A and 30
hours on imachine B. Profit gained by the manufacturer from achair
and atable is Rs. 2 and Rs. 10 respectively. What should be the daily
production of each of the products.
24. A person requires 10, 12 and 12 units of chemicals A, B and C
respectively for his garden. A liquid product contains 5, 2and 1units
of A, B and C respectively per Jar. A dryproduct contains 1,2 and 4
units of A, B and Cper carton. How many of each should he
purchase in order to minimize the cost and meet the requirement.
25. Solve the following problem graphically,
Maximize Z =
40x + 100x,
subject to 12x + 6x, s 3000
4x +10x, S 2000
2x +3x S 900
and x, 2 0 [MU. B: Tech. Leather Oct 96)
products,
26. A garment manufacturing company can make two
Prima and Seconda. Each of the products requires time on a cutting
machine and a finishing Machine. Relevant dataare
Product
Prima seconda

Cutting hours (per unit) 2 1

3 3
Finishing hours (per unit)
Unit cost Rs. 128 120
Selling price Rs. 134 129
Maximum sales (units per week) 200 200
The number of cutting hours.available per week is 390 and the
number of finishing hours available per week is 810. How much
should each product be produced in order to maximize the profit ?
|BRU. BE. Apr 94
27. A firm manufactures refrigerators and air coolers. Production
takes place in two separate departments. Refrigeratorsare produced
in department Iand air coolers are produced in depart1nent II. The
firm's two products are produced and sold on a weekly basis. The
weekly pro-duction cannot exceed 25 refrigerators in department I
and 35 air coolers in department II, because of the limited available
facilities in these two departments. The firm regularly employs a total
 Resource Managcment Techniqucs
departments. A
2.38
in two refrigerator
The
workers
of60 labour, whiiean air
coolers
requires I man-w
wecksfirm receives A profit margin of Rs.300 and Rh
of
cooler
cooler respectively. Determine
r(he equitny
refrigerator and prot
order to
maximize profit. [BRU.
28. The productionand planning department of asoft tri,
following problem. The bottling lant has
faces the
machines Aand B. Ais designed for 80cc bottle and Bfhr
bottles. However each can be used on both types with
efficiency. The following
data is available.
80ce bottle 10
Machine 160 cc bote
12
A 100per min 40 per min
B 60 per min 15 per min
The machine can run 8 hour per day 5days per weck. Proft
cc bottle is 15 paise and 160 cc bottle is 25 paise. Weekly proda
on the drink cannot exceed 3 million cc and the market can
250000 of 80 cc bottles and 70000 of 160 cc bottles per weel
department wishes to maximize the profit subject to productia
marketing restrictions. Solve the problem by graphical method
simplex method. [BRU. BE. N1S
29. Acompany manufactures two products A and B. Each u
Btakes twice as long to produce as one unit of A and if the com
were to produce only Ait would have time to produce 2000 umt
day. The availability of the raw material is sufficient to produc
units per day of both A and B combined. Product B reqa1
special ingredient only 600 units can be made per day. If Afeco
profit of Rs. 2 per unit and B a profit of Rs. 4 per unit find
optimum product mix by graphical method. [BRU. BE.
30. A company produces two different products Aand B.
company makes a profit of Rs. 40 and Rs. 30 per unit on A
respectively. The production process has a capacity of30,000 how
hours. It takes 3 hours to produce one unit of A and onethal
produce one unit of B. The narket survey indicates
maximum number of andthose
units Athat can
is 12,000 units. Formulate the problem be sold is 8,000
and solve it by grap
method toget maximum profit. MU. BE.A
31. Apply graphical method to find non-negative values of
x, which minimise z= 10x, + 25x, subject to x+ >50,.N
Xy s 40.
/M.U. M. B..4. AM
Chapter 3
General Linear Programmi
Problems - Simplex Metho
-

Linear Programming Problem

3,1.1General programming involving more than two variables ma
The linear
expressed as follows:
Cz3....c,
Maximize (or) Minimize Z= c t Cgt
subject to the constraints +a1n S or = or 2b,
+ + an S or =orsb,
+ + a3,n S or =orsb

tamn Sor = or 2b
restrictions
and the non-negativity
y20.
equalities, some others m
Note : Some of the constraints may be inequalities of (2) type
inequalities of (5) type and remaining ones
of them are of same type. which satisfie
,
Definition () : A set of values x, *.....
constraints of the LPP iscalled its solution.
which satisties the
Deftnition (2) : Any solution to a LPP solution.
negativity restrictions of the LPP is called its feasible (maxm
Defniton (3) :Any feasible solution which optimizes called its p
minimizes) the objective function of the LPP is
solution or optimal solution.
Definition (4) : If the constraints of a general LPP be
n

Zaj,sb, (i =1,2, 3, ... ) con

j=|
then the non-negative varizcles which are introducedto
inequalities (1) to the equalities
(i= 1,2, 3, ... )
j=|
 3.2
Resource Management Techniques
are called slack
as the amount ofvariables. The value of these
unused resource. variables can be interpreted
Definition (5) : Ifthe constrairt's of a general LPP be
(i =k, k+ 1, ...) .(1)
then the non-negative variables, s, which are introduced to convert the
inequalities (1) to the equalities
= b (i =k, k+ 1, ...)
j=|
are called surplus variables. The
välue of
interpreted as the amount over and above the these variables can be
required level.
3.1.2 Canonical and Standard formsof LPP :
After the formulation of LPP, the next step is to obtain its
But before any method is used to find its solution, the problem solution.
must be
presented in a suitable from. Two forms are dealt with here, the canonical
form and the standard form.
The canonical form : The general linear progranmming problem can
always be expressed in the following form:
Maximize Z= c; t cy tC33 t ......Cn
subject to the constraints

t amnn S bm
and the non-negativity restrictions
-20.
This form of LPP is called the canonical formof the LPP.
In matrix notation the canonical form of LPP can be expressed as :
Maximize Z = CX (objective function)
subject to AX s b (constraints)
and X 0 (non-negativity restrictions)
whére C = (C C
..C),
 The Sitnplex Method

a12 a22 a2n b 1

A= , X

Characteristics ofthe Canonlcal form :

(0 The objective function is of maximization type.
(i) Allconstraints are of (s) type.
(üi) All variables x, are non-negative.
The Standard Form:
The general linear programming problem in the form
Maximize Z= cx) t cn t.....+cn
subject to the constraints
=b
+ aznn b,

Cm1tm22t tam'n =bm

and x1 *>» .*, 20 is known as standardform.
In matrix notation the standard form of LPP can be expressed as
Maximize Z = CX (objective function)
subject to constraints
AX b and X0
Characteristics of the standard form :
() The objective function is of maximization type.
(ii) All constraints are expressed as equations.
(ii) Right hand side of each constraint is non-negative.
(iv) All variables are non-negative.
Note : equivalent o
(1): The minimization of a function f(x) is
maximization of the negative expression of this function.
i.e., Min f() = -Max {-f(«) }
i.e., Min Z =-Max (- Z)
e.g: Min Z =Cjx t Czx, is equivalent to
Max (-Z) = - Cjx-C2*2
 (2): An
Resource Management Techniques
inequality in one
in the
opposite directiondirection
by
can be
converted into an
e.g:
ax +bx, multiplying both sides by (-1).inequality
(3): An equality
constraint can be expressed as two
e.g: a +bx, at bxy Sc axinequalities.
+ bx, Sc
ax + bx,
(4): An inequality
constraint with its left
form can be expressed as two
c-ax-bx, S-¢
hand side in the absolute
inequalities.
e.g : |ax tbx, |s c>
(S): If avariable is
unconstrained or unrestricted
its sign), it can always be expressed (without specifying
as the difference of two non
negative variables.
e.g. : Ifx, isunrestricted, then
X= -x where x', x"0.
(6):whenever slack / surplus variables are introduced in, the
constraints, they should also appear in the objective function with
zero coefficients.
Example 1: Express the following LPP in the canonical form.
Maximize Z = 2* +3x, t
subject to the constraints
x+ 5x-7x3 -4
and x, X3 20, X, is unrestricted
Solution : As x, is unrestricted, x -}"
where x,,x,"20. .:. The given LPP becomes
Maximize Z = 2x,+3 (*-»)+*3
subject to 4x-3x+3x," t S .6
* + 5x'- Sx" 4
0.
Convert the second constraint s type by multiplying bothsides by -1.
Now the LPP becomes
Maximize Z 2x +3xy' -3x," +x
subject to 4x -3x' +3x," +x s
4
0.
which is inthe canonica! form.
 1VIethod
LPPin standard form
thefollowing
Expressthe
Example2/: Minmize Z = 5x +Z,
S8
constraints
subjecttothe 3r +4x, 23
6r + 7, 2 5 and and.x ) 2 0.
MinZ=-
- Max (-Z)=- Måx Z°,
Solution: Since Maximize Z
LPPbecomes 8
Thegiven
subjectto 3x + 4x, 2 3
6x, +7x; 5 andx, x, 20
variables
slack variable s and surplus
By introducing
ofthe LPP isgiven by
standard form
MaximizeZ* -Sx-Tx, +Os,+Os, +Os
subjectto 3x +4xj-S2 3
5
6x + 7,-S3
0.
and x, X2, S1, S2, S3
Express the following LAP in standard (Matrix) for
Example 3:
MaximizeZ=
6
subject to 8

6r - 4x + ,320
s 10 andx slack
surplus variable s, and vrie
Solution : By introducing the
the standard form of the LPP becomes
Maximize Z = 4x +2r, +6x, +Os t Os2
6
.subject to 2x, +3x, + 2r-s tOs2
3x +4x, + r, +Os+ Os 10
0.
andx,2 S2
Thus the given problem in matrix form is
Maximize Z CX
subject to AX =

X2
where C (4, 2, 6, 0, 0)

2 3 2 -I
A 3 4 0 X= 3
6 -4 1
 3.6 Resource Managcment Techniques
3.1.3The Slmplex Method
While solving a LPP graphically, the region of fenstble-lutions was
found to be convex. The optimal solution if it exists, oceured at 4ome
vertex. If the optimal solution was not unique, fhe optimal points were on
an edge. These observations also hold for the general L.PP. Essentially the
problem is that of finding the particular vertex of the convex tegion vhich
used methcd for
corresponds tothe optimal solution. The motcommonly
or simplex technique
locating the optimal vertex is the simplex.metlhodDantzig
G. in 1947.
or simplex algorithm which was developed by
which consists in moving
Thesimplex method provide_ an algorithm solutions to another vertex in
foasible
from one vertex of the region of function at the succeeding
that the value of the objective
such a way may be) than at the preceding vertex.
vertex is more (or less, as the case
jumping from one vertex to another is then repeated.
This procedure of finite, the method leads to an optima'
vertices is
Since the number of or indicates the
existence of an
of steps
vertex in a finite number
unbounded solution.
of m linear equations with
systani
Definition (1): Given a obtained by setting (n - m) variables
The solution,
nvariables (m < n). remaining m variables is calleda basic
solving for the
equal to zero and
solution. basic
basic variables and they form the
called
The m variables are are put to zero are called as
non
variables which
solution. The (n- m)
basic variables.
to be a non-degenerate busic
Definitioi (2) : A basic solution is said
none ofthe basicvariables is zero.
solution if
solution is said to be a degenerate basic
Definition (3) : A basic variables are zero.
of the basic
solution if oneor more
solution which is also basic is called a
Definifioh (4) : A feasible
basicfeasiblesolution.
to the following
Example 1 : Find all the basic solutions
problem :
Maxinize Z
4
subject to x + 2x,+ 3x
2x, +3x, + 5r, = 7
 The Simpléx Method
solutionsare
basic
whichofthe
findbasicfeasible
Also()
feasible
non-degeneratebasic
() feasible.
optimalbasic with n=3
(i)
: Since there are m 2 equations
(n - m) = (3-2) = I variables,
variable the
equa
Solution
obtainedbysetting two variables. Since
basic solutions are
solving for the
remaining there
are
to zero and equations. We shall have 3C, = 3
table,
differentbasi
variables with 2 given in the following
solutions, whichare Is the Is the
Values of the Value of Is the solution solution
basic variables the obje solution n0n
Basic
Non
basic givenbythe ctive fea degener
feasible
and
S Pari vari constraint sible ?
No ables
function ate ? optimal ?
ables equations
=4

2x +3x, =7 yes yes yes

x2, *,-1
+3x = 4
4 yes yes No

2x, +3x3 = 4
3 2 X3 =0 3x, +5x, =7 3 No yes No

non-degenerate
From the table, we see that, the first two solutions are
non-degenerate and infeasible.
basic feasible solutions and the third is
The first solution is the optimal one.
The optimal solution is
Maximize Z = 5, x =2, x, = I,x =0.
Exan1ple 2: Obtain allthe basic solutions to the following systen
of linear equations :
2x, + 6x, + 2x, t x4 3

6x +4x,t 4x3 +6x4 2

Which of them are basic feasible solutions and which 2

non-degenerate basic solutions ? Is the non-degenerate solution
feasible ?
 3.8
Resource Management Techniques
Solution.: Since there are 4
4C2=6 different basic solutions,variables with 2
which are givenequations,
in the we shall have
Basic
Values of the
following table.
S.No vari Non-basic variables givenbasle Is the
Is the
variables the constraint by solution
Is the
ables solutlon solution
non feastble
equatBons feasible ? degene and non
degene
2r + 6x, 3) rate?

yes No No

|2x + 2x3 =3)

2 2
No yes No

-3|
3
8 No ye_ No

-343
|6x, +2x,=3)
4 |4x, +4x3 =2
1 yes No No

6x, t x4 = 3])
yes No No
1

2r3 tx4=31
6 X3»*4 ) =0413 +6x4=2| No yes No
3=2,X4 =-1
Definition(5): Let Xp be a basic feasible solution to the LPP:
Maximize Z. =
CX

subject to AX
and X 2 0.

Then the vector CR =(CB, CB2,.. CBy), where CB, are components
cost vectur
of C associated with the basic variables, is called the
USSociated with the basic feasible solution XB.
 The Simplex Method
1: Ifa LPP has
Remarks: afeasible solution, then it also has a basic te
solution.
finite auntber of basic feasible
There exists only
2:

3:
LPP.
Let a LPP have a
feasible solution, 1f. we drop one
and introduce another
variables
new
vasiable in the
basis
soseetluli,othe,
oi
solution obtained is also a basic feasible solution. then
Definition (6): Let XB = B-! be a basic feasible solution th
CX, where AX =b and
LPP:
Maximize Z
X>0,
Let CR be the cost vector CorresponaingB,toletAB. For each .
column vector of
vector a, in A, which isnot a m

Then the number Z; =.XCB, a;

i=l

is called the evaluation corresponding to a,and the number (2.,

is called the net evaluation corresponding to a,.
Remark l:If (Z; - C) = 0 for atleast one j for which a,
i= 1,2, ...,m ;then another basic feasible solution is obtained which ei
an unchanged value of the objective function.
Remark 2: (Unbounded solution). Let there exist abasic feasi
solution to a given LPP. If for atleast one i, for which a, S
(i=1,2,..., m) and (Z, C) is negative, then there does not exist
optimum solution to this LPP.
Remark3:Anecessary and sufficient condition for a basic feasit
solution to a LPP to be an optimum (maximum) is that (Z, - C)20 tor
j, for which a, e B.
Remark 4: The two fundamental conditions on which the simpe
method is based are : () Feasibility condition : It ensures that if the
(Starting) solution is basic feasible then during computation only oI
feasible solutions will be obtained. (i) Optimallity condition:
guarantees that only improved solutions willbe obtained.
3.1.4 The Simplex Algorithm optin
Assuming the existence of an initial basic feasible solution, an
solution to any LPP by simplex method is found as follows : maximizedd
Step I : Check whether the objective function is to be d
minimized. If it is to be minimized, then Convert it into a problem
maximization, by
Minimize Z = -Maximize (-Z)
 3.10 Resource Management Techniques
Step 2 : Check whether all b,'s are positive. if any of the b,s is
negative, multiply both sides of that constraint by - so a to make ite
right hand side positive.
Step 3 : By introducing slack / surplus variahles, convent the
inequality constraints into equations and express the given lLPP into its
standard form.
Step 4: Find an initial basic feasible solution and express the above
information conveniently in the following simplex table
C; C 0

Yø
CB1 S1 bi a3

be a21 ag2
a32
S3 bs

unit matrix
body matrix

(2-G) Zo Z-C Z, -C,

C, - row denotes the coefficients of the variables in the
(Where
function. CR Column denotes the coefficients of the basic
objective YR - column denotes the basic
variables in the objective function.
X¡ - column denotes the values of the basic variables. The
variables.
coefficients of the non-basic variables in the constraint equations
coefficients of the basic variables
constitute the body matrix while the the net evaluations
constitute the unit matrix. The row (Z - C)denotes
(or) index for each column).
... n) by using
Step 5:Compute the net evaluations (2 -C)r =1, 2,
the relation Z, -C=CB 4-C
Examine the sign ofZ -
solution X, is
(a) If all (Z, C) 2 0then the current basic feasible
optimal.
(b) If atleast one (Z, - C) <0, then he current basic feasible solutivn
is not optimal, goto the nextstep.
 The Simplex Meth
Siep 6: (Tofind the enterlng variable)
The entering variable is the non-basic variable

some.Cor escIpooneloheunminng
most negative value of( - C). Let it be x, for
variable column is known as the key column
(or)
shown marked with an arrow at the bottom. If morepBvot
than
the same most negative (2-), any of these variables m vati
arbitrarily as the entering variable.
Step 7: (To find the leaving varlable)
Compute the ratio = Min
>of ue. the raio be
solution column andthe entering variable column by
positive denominators) considering o
(a) If all ai, s0, then there is an unbounded
LPP. solution to the
(b) If atleast one a,, > 0, then the leaving
variable is the
variable coresponding to the minimum ratio 0. If 0=Xa,
the basic variabl x, leaves the basis. The leaving
variable
called the key row or pivot row (or) pivot
element at the intersection of the pivot columnequation, a
and pivot r
called the pivot element or key element (or) leading elemem
Step 8 :Drop the leaving variable and introduce the entering va
along with its associated value under CR column. Convert the
element to unity by dividing the pivot equation by the pivot elemer
all other elements in its column to zero by
making use of
(i) New pivot equation = oldpivot equation +pivot
element
(i) New equation (all other rows including
(Z,-C) row)
Old equation -
(Corresponding New pivot
column
coefficient equation
Step 9 : Go to step (5) and repeat the procedure unitl eithe
optimum solution is obtained or there is an indication of an unb
solution.
Note (1) : For maxinmization
problems:
() If all (Z, - C) 2 0, then the current basic feasible solu
optimal.
3.12 Resource Management Techniques
(i) If atleast one (Z,- C) <0, then the current basic feasible solution
is not optimal.
(i) The entering variable is the non-basic variable corresponding to
the most negative value of(2-C).
Note (2) : For minimizalion problems :
() If all (Z;- C) s 0, then the current basic feasible solution is
cptimal.
(i) Ifatleast one (Z, -C)>0, then the current basic feasible solution
is not optimal.
to
(i) The entering variable is the non-basic variable corresponding
the most positive value of (Z-G).
problems, the
Note (3): For both maximization and minimization the minimum
is the basic variable. corresponding to
leaving variable
ratio 0.
the LPP
Example 1}: Use simplex method to sotve
Máximize Z = 4x, + 10x,

50
subject to
2x, + 5x, 100

2x,+3x, s 90 and X1, X 2 0.

84, MU. BE. Nov 91, Nov
94)
MKU.B.Se 94, MU BSc.
[BRU BSc. 86,
introducing the slack variables S1, S, and s3, the
Solution: By
becomes
problem in standard form
Maximize Z 4x,+ 10x, tOs; t Os +Os3
= 50
subject to 2x t x ts +Os, + Os;
= 100
2x, + 5x + Os, t s + Os,
= 90
2x,+ 3x+ Os;+ Os) ts3
and x1, , S], S2, Sg 2 0.
3 equations with 5 variables, the initial basic feasible
Since there are
=2 variables to zero.
solution is obtained by equating (5-3)
s, = 100, s, = 90
The initial basic feasible solution is s 50,
(x =0, x, =0, non-basic)
3.12 Resource Management Techniques
(ii) If atleast one (Z,-C) <0, then the current basic feasible solution
is not optimal.
(ii) The entering variable is the non-basic variable
themost negative value of (2, -C). coresponding to

Note (2) :For minimization problems :

() If all (Z, C) s 0, then the current basic feasible solution is
cptimal.
(i) If atleastone (Z-C) >0, then the current basic feasible solution
is not optimal.
(iüi) The entering variable is the non-basic variable corresponding to
the most positivevalue of (2;-C).
Note (3) : For. both maximization and minimization problems, the
leaving variable is the basic variable. corresponding to the minimum
ratio 0.

Example 1}: Use simplex method to sotve the LPP

Maximize Z =
4x,+ 10x,
subject to 50

2x + Sx S 100

2x, + 3x 90 and X1 Xy 2 0.
[BRU BSc.86, MKU.B.Sc 94, MUBSc. 84, MU. BE. Nov 91, Nov 94]
Solution: By introducing the slack variables s1, S and S3, the
problem in standard form becomes
Maximize Z
4x1 + 10x, + 0s + Os, + Os3
subject to 2x + x t s +Os,t Os3 = 50
2x + 5x + Os t s) + Os3 =
100

2x1 + 3*2 + 0sy t sS3 = 90

and x, N2, S1, S2, S3 2 O.

Since there are 3 equations with 5 yariables, the initial basic feasible
Solution isobtained by equating (5 -3)=2 variables to zero.
The initial basic feasible solution is S, = 50, S = 100, S3 yO
(a =0, x, =0, non-basic)
 The
initialsimplextable is given by he Sitn
G (4 10

Cp Y |X
50 2 1

100 2 (5) 0
0 90 3
4 -10
Here the net evaluations are calculated as Z.
Z,-C -CB 4 -C1 = (000)(22 2jT-4
[where Tde
eo
-4=-4
(000) [153]"- 10=-0)
= (000)[10 0j"- 10 =0
Z,-C4 =Cg a4-C4 (000) [0 10j-0=0
Z-C; =CB a_ Cg = (000) [00 1]"- 0=0
Since there are some (Z, -C) <0, the
is not optimal. current basic teasi
Tofnd the
Since entering
- C,) variable
:
(2,
basic variable enters = -10 is the
most nenegative, the correspu
the
called the key column or basis, The column correspondng
To fnd the pivot
leaving variable : column.
Find the ratio = min
air
min (Bi a
2
min 100 90)
correspom
The min { 50, 20,
rOW mior nikeymumleavirationg 9varia20.ble
he 30} = 20, which
correp
is the basic which
New pivot row or The variable s2 the
called

equation - pivot equation

old leaving variable row is elemeal
and 5 is the pivot
pivot equation + pivot element.
 The Simpex Method
Remarks:
|: If a LPP has a feasible solution, then it also has a
2:
solution.
There exists only finite Autitber of basic feasible
basic fer
3:
LPP.
Let a LPP have a feasible solution. 1f.
solutions
variables and introduce another-vaiable we-drop one of
in the basis the t
new solution obtained is also a basic feasible solution. set, then
Definition (6): Let XB = B-l b be a basic feasible
LPP: Maximize Z = CX, where AX =b
Let CB be the cost vector corresponding to XB. For
solution
and X>0. to%
vector a, in A, which is not a column vector of B, let each colr
m

Then the number Z; =CB; ai

is called the evaluation
is called the net evaluation corresponding to a, and the number (2,-t
Remark l: If (Z; - corresponding
to a,.
C) =0 for atleast one j for which a.,.
i=1,2, ...,m ; then another basic feasible solution is
an unchanged value of the objective function. obtained which give
Remark 2: (Unbounded solution). Let there exist a basic
solution to a given LPP. If for atleast one i, for feasiblk
(i= 1,2,..., m) and (Z, C) is negative, then which
optimum solution to this LP, there does not exist a
Remark 3: A necessary and sufficient condition for a
solution toa LPP to be an optimum (maximum) basic feasibl:
j,for which a, e B. is that (Z, -C) 0 for al
Remark 4 : The two fundamental conditions on which the
method is based are: () Feasibility condition : It simple
(Starting) solution is basic feasible then during ensures that if the inita
feasible solutions will be obtained. () computation only bas
guarantees that only improved solutions will be Optimality
obtained.
condition:
3.1.4 The Simplex Algorithm
Assuming the existence of an initial basic feasible solution, an
solution to any LPP by simplex method is found as optim
Step I : Check whether the objective function follows :
minimized. If it is to be minimized, then convertisittointobe maximized 0
aproblem o
maximization,by
Minimize Z
Maximize (-2)
 3.10 Resource Managemcnt Techniques
Step 2 : Check whether all b,'s are positive. If any of the
b,'s is
negative, multiply both sides of that constraint by -1 so as to make its
right hand side positive.
Step 3 : By introducing slack / surplus variables, convert the
inequality constraints into equations and express the given LPP into its
standard form.
Step 4 : Find an initial basic feasible solution and express the
above
information conveniently in the following simplex table.
Cq........ 0 .... )

bi d12 ........ 1 00
10
C8 S2 ......0
a33 ......... 0 01
S3 by d32

.......

body matrix unit matrix

Z-C) Zo. Z-Ci Z,-C,

(Where C; - row denotes the coefficients of the variables in the
objective function. CR - column denotes the coefficients of the basic
variables in the objective function. YR - column denotes the basic
variables. Xp - column denotes the values of the basic variables. The
coefficients of the non-basic variables in the constraint equations
constitute the body matrix while the coefficients of the basic variables
constitute the unit matrix. The row (Z; C) denotes the net evaluations
(or) index for each column).
Step 5 : Compute the net evaluations (Z, -C) ((=1, 2, ...n) by using
the relationZ-C=CB 4
Examine the sign of Z,-
(a) If all(Z, - C) 20then the current basic feasible solution XR is
optimal.
(0) If atleast one (Z, - C) < 0, then the current basic feasible solution
is not optimal, go to the next step.
 The Simplex Method
enterlng varlable)
findthe
Step6: (Tovariable is the non-basic variable
The entering (4-) Let it be x, tor some j
correspondn
of
most negative valueknown asthe key column (or) pBvot column
variable columnis an arrowatthe bottom. If more than one vari
with may k
shown marked
negative of these variables
(Z-C), any
the same most
entering variable.
arbitrarilyas the
leavingvariable)
Step 7:(Tofindthe
Compute the ratio =
Min 4,>0 (ie, the ratio betNe
variable column by considering o
solution column and the entering
positive denominators)
s0, then there is an unbounded solution to the
(a) If all a,,
LPP.
the leaving variable is the
(b) If atleast one a,, > 0, then
ratio 0. If0=
variable corresponding to the minimum
leaving variable
the basic variabl x, leaves the basis. The
called the key row or pivot row (or) pivot equation, an
element at the intersection of the pivot column and pivot r
called the pivot element or key element (or) leading elemer
vau
Step 8 : Drop the leaving variable and introduce the entering
along with its associated value under Cp column. Convert the
element to unity by dividing the pivot equation by the pivot elemer
all other elements in its column to zero by making use of
(9 New pivot equation = old pivot equation + pivot element
(i) New equation (all other rows including (Z, - C) row)
(Corresponding) New pivot)
=Old equation - column
equation
coefficient
Step 9 : Go to step (5) and repeat. the procedure unitl eithe
optimum solution is obtained or there is an indication of an unbou
solution.
Note (1) : For maxinizationproblems:
() If all (2- C) 20, then the current basic feasible Solutio

optimal.
3.12 Resource Management Techniques
(i) If atleast one (Z;-C)<0,
then the current basic feasible
isnot optimal. solution
(i) The entering variable is the
the most negative value of non-basic variable corresponding to
(Z-C).
Note (2) :For minimizatton problems :
() If all (Z; - C) s0, then the
current basic feasible solution is
cptimal.
(ü) If atleastone (Z-C{) >0, then the
current basic feasible solution
is not optimal.
(üi) The entering variable is the
the most positive value of non-basic variable corresponding to
(Z,-C.
Note (3): For both maximization and
leaving variable is the basic variable minimization problems, the
ratio 0. corresponding to the minimum
Example 1j: Usesimplex method to sotve the EPP
Maximize Z =
4x + 10x
subject to 50

2x + 5x 100

2x,+3x, s 90 and x1, xy 2 0.

[BRU BSc.86, MKU.B.Sc 94, MUBSc. 84, MU., BE. Nov 91, Nov 94]
Solution: By introducing the slack variables S1, Sh and s, the
problem in standard form becomes
Maximize Z 4x + 10x + 0s + Os, + Os3
subject to 2xtx ts +Os, +Os? = 50
2x + 5x, + Os t s+ Os3 = 100
2x +3x,t Os + 0s, t s3 = 90
and x, y, S1, S2, S3 2 0.
Since there are 3 equations with 5 variables, the initial basic feasible
solution is obtained by equating (5-3) =2 variables to zero.
:: The initial basic feasible solution is S = 50,
S= 100, S3 y0
(I =0, x =0, non-basic)
 The initial simplex table is given by
The Simplex Meh
C; (4 10 0 0
0)

0 50 2
0
100 2 (S)
90 2 3 0 20+
0 -4 -10 30
Uere the net
evaluations are calculated as Z-C =Cp 4,-c
Z-C =Cg aj -C; = (000) [22 2]" -4
[where Tdenotes tran
-4=-4
Z,-C, =Cg ay-Cz =(000)[153]"-10 =-10
Z-C; *Cg a3 -C3 = (000) [10 0j - 10=0
Z4-C4 =g aj-C4 = (000) [0 1oj-0 =0
Zs-Cs =Cg as-Cs =(000) [0 01]"-0=0
Since there are some (Z, - C) <0, the current basic
is not optimal.
feasible S
To find tlhe entering variable :
Since (Z,-C,) = -10 is the most negative, the
basic variable x, enters the basis. The column corresponding
called the key columnn or pivot column. corresponding tothis
Tofnd the leaving variable :
Fiind the ratio = min XBi
air i
= min

(50 100
= min
= min { 50, 20, 30} = 20, which corresponds l
.. The leaving variable is the basic variable s, which corresponu
the minimum ratio = 20. The leaving variable row is çalled the p
row or key row or pivot equation and 5 is the pivot element.
New pivot equation =old pivot equation + pivot elenment.
3.14 Resource Management Techniques
(100 2 501 0)+5
20 I0 0

New s, equation =
old s, equation
(corresponding
column
New
pivot
coefficient \equation/
= 50 2 1 (0

() 20 1 0

30 0

New s, equation 90 3 1
60 3
3 0 0

4
= 30 1

New (Z,-C) eqn. 4 -10 0 0

-200 -10.
5 0

= 200 0 2

:. The improyed basic feasible solution is given in the following

simplex table.
First Tteration :
C (4 10 0 0 0)
CB YR S
0 30 5
10 20
0 30 5 0 0
200 0 2 0

Since all Z-C,20the current basic feasible solution is optimal.

.:. The optimal solution is Max Z=200, x =0, x, =20.
Example 2: Find the non-negative values of x and x which
Maximize'Z= 3x +2x,+ 5x3
subject to S 420
3x + 2x S 460
+ 2x, tx S 430 IMU. MBA. Nov 95]
 Solution : Given the LPP
The Simplex Method
Maximize Z = 3xt2x, +5x 3
subject to x t4x, S 420
3x +2x3 s 460
Xt 2r, tx S 430

By introducingnon-negative slack variables S1, S, t

form of the LPP becomes and s, the st

Maximize Z = 3x1 +
2x, t+
subjectto x+ 4x,+ 0xts+0s, +Os3 5xy+ 0s+0s, +og
420
3x +0x, t 2x, + 0s +Sy+
Os3 460
x+2x, t,+Os + Os, t s3 430
0
Since there are 3 equations with 6
solution is obtained by equating (6-3)variables,
=3
the initial basic fezs:
:. The initial basic variables to zero.
(X=X =0, non-basic) feasible solution is s,= 420, s) = 460, s ={!
The initial simplex table is
Initial iteration :
given by
CG (3 2 0 0 0)
Yg S1
0 S3
S 420 1 4
1
S2 460
0 (2) 0 1
460
= 230
2
430 2 1 0 -4301 =430
0 -3 -2 -S
0
Since there are some
isnot optimal. (-C) <0, The current basic feasible solu
To find the
negative, the entering variable : Since (Z, - C;) = 5 is thebas
mo
The column corresponding non-basic variable X3 enters into the pia
column. corresponding to this xa is called the key column of
3.16 Resource Management Techniques
Tofind the leaving variable :
Find the ratio 0 = min min
air
[460 4301
0= min l2 = min {230, 430} = 230
The leaving variable is the basic variable s, which corresponds to
the minimum ratio 0 =230. The leaving variable row is called the key row
or pivot equation and 2 is the pivot element.
New pivot equation = old pivot equation + pivot element
(460 3 0 2 0 1 0)+2
1
230 0 1 0
(its entering New
New s tquation old s equation - column pivot
\coefficient) \equation/
= 420 1 4 0 1
.
() 0 0

420 1 4

(its entering) New

column pivot
New s equation old s, equation X

\coefficient/ \equation/
= 430 2 1 1

230 0 1
(
200 2 0

fits entering New

column X pivot
New (Z,- C) eqn. =old (Z-C) equation - \equation/
coefficient/
0 -3 -2 -5 0

-150 0 -5
()

= 1150 -2

The improved basic feasible solution is given in the following simplex

table.
 The Simplex Meth
istiteraltlon:
(, (3
G 2 5 0
0)

0 420 1 4 0 11 0 42)4
230 0 1 0 0
200-2 (2) 0 0 200
2
Z-C,1s0 2 0 0 0
Since there is an (Z, - Ch)=2, the current basic feasible
not optimal.
solutug
. Here the non-basic variable x, enters into the basis and h
the
variable s, eaves the basis.
Second iteration :
G (3 2 5 0 0)
YB S S
0 S 20 2 1 -2
230 1 0
2 100

Z-G 1350 4 2 1

Since all (Z, -C)20, the current basic feasible solution is optima
.. The optimal solution is Max Z= 1350, x,
=0, X, = 100, X3
Example 3: Solve the following:
Maximize: 15x + 6r+ 9xy+2x4
subject to 2x1 t x,t 5iy+ 6x4 S 20
3x, +xyt 3x,+ 25x, s 24
7x, tx s 70
IMU. MCA,Muy
Soluion: By
the standard form ofintroducingnon-negative slack variables s1 S2
the LPPbecomes.
3.18 Resource Management Techniques
Maximize Z
15x+ 6x; +9x, +2x+ 0s1 + Os, +Os3
subject to 2x t x, t Sx + 6x4t s +0s +Os3 = 20
3x tx, + 3x, + 25x + Os t s, 24
0s3
+ 0x, + Ox, t x4+ Os +Os, +s3 = 70

The initial basic feasible solution is s, =20, S, = 24, S3 = 70

(I =)=X X4 =0, non-basic)
The initial simplex table is given by
Intial iteration :
G (15 6 2 0 0 0)
YB S S3

20 2 5. 6 1 0 20 = 10
2

0 24 1 3 =8*
(3) 25 .1

70
70 7 0 0 0 0 1 7 =10
-15 6 -9 -2
Since there are some (Z; -C)'<0, the current basic feasible solution
isnot optimal.
The non-basic variable x enters into the basis and the basic variable
S, leaves the basis.
First iteration :
C (15 6 9 2 0 0 0)
YB S S3
0 4 ('4) 3 32 0 12*
S1
15 8 1 1 2513 24

-172
S3 14 -7 3
120 -1 6 123 0

Since (Z, - C) = -1 < 0, the current basic feasible solution is not

 The Simplex Method
The non-basic variable x, enters into the
leaves the basis.
basis and the
Second iteration :
basic
6

S
S
6 ) 12 9 -32 3
-2 0
15 4 1 0 -2 -1

42 0 0 14 -132 7

-G 132 0 15 91 3 3 0
Since all (2-C) 20, the current basic feasible solution is oni
. The optimal solution is given by
Max Z =132, x =4, x =12, X3 =0 and x=0.
Example 4: Solve the following LPP by simplex metE
Minimize Z= 8x -2.
subject to -4x, +2x,
Sx-4x 3

and x,x, 2 0. [MKU. BE.

Solution : Since the given objective function is of minimization
we shall convert it in to a maximization type as follows :
Maximize (-) =Maximize Z* =-8x +2,
subject to 4x,+2x, 1

5x - 4x, 3
0.
standard
By introducing non-negative slack variables s1, S2» the
of the LPP becomes
Maximize Z* -8x + 2x, + 0s, + Os
subject to the constraints
-4x +2x, ts+ Os, =
Sx - 4x, +Os ts = 3

and x X, S1 Sy 2
.. The initial basic feasible solution is given by SË
( x, =0, non-basic).
3.20 Resource Management Techniques
Initial iteration :
C (-8 2 0 0)

1 4
(2) 1

0 3 5 4 1

Z-C 0 8
Since (Z,*- Cc) = 2<0, the
-2 0
current basic feasible solution is not
optimal.
The non-basic variable x, enters the basis and the basic variable
S leaves the basis.
First iteration :
C (-8 2 0)
YB 1
S S2
2 -2

0 5 -3 0 2.1
4 0 1

Since all (Z*-C) >0, the current basic feasible solution is optimal.
. The optimal solution is given by maximize Z* = 1, x =0,x =
But Minimize Z=-Maximize (-Z) =- Maximize Z* =-]
Min Z=-1, x =0, x, 3
Aliter : The above problem can be solved without converting the
objective function in to maximization type.
Given Minimize Z = 8x -2r,
subject to the constraints - 4x1 + 2x, S 1
5x - 4x S 3

..
By introducing the non-negative slack variable s, the LPP
becomes
Minimize Z 8x -2x, +Os+ Osz
subject to the constraints -4x + 2x,t st Os
5x -4x +Os+ s2 3
0.

The initial basic feasible solution is given by

S =1, s, =3(basic) (=X,=0, non-basic)
 Initial iteration :
The Simplex Meth
G (8 -2
0)
0
4 (2)
0 5 4 0
0 -8

Since (Z, - C) =2>0, the current basic

optimal.
To fnd the entering variable :
feasible soluti
Since (Z, - Cq) =2 is most positive, the
variable x, enters into the basis.
Tofind the leaving variable :
The leaving variable is the basic
cor esponding ne
variable S1
minimum rätio 9 correspondin
First iteration:
(8 -2 0 0)
YR S
-2
-2
2
0
-3 0
-4 0 -1
Since all (Z,-C)s0, the current basic
feasible solution is optir
The optimal solution is Min Z=-1, X=0,
x,=i
Example 5]: Use simplex method to
Min Z
subject to
-2x, +4x, s 12
10
and x, N3 Xg 2 0
[BRU. MSc. 1986 MKU, BSC 1992, MU. BE. Nov 93, N
MU. B.Tech. Leather
Solution: Since the given objective function is of minimization
we shall convert it in
to a
naximization type as follows :
Maximize (-Z) = Maximize Z* = -X, t 3x3)
3.22 Resource Management Techniques
subject to 3x, -X+2xs s 7
-2x2 +4x3 12
-4x, +3x3 +8xs 10
0.
By introducing non-negative slack variables s,, s, and sa, the standard
form of the LPP becomes
Maximize Z * = *, + 3*, -2xç + 0s + Os, + Os3
subject tothe constraints
3xy- X,t 2x, ts t Os, +Os3 7
-2x, + 4x, + 0x, +Os, t s+Os3 12

Ax) + 3x, + Rx + 0s + Os, t s3 10

and ,,X3, Xs, S1, S2 S3 2 0.
The initial basic feasible solution is given by S = 7, s = 12,
S3 =10 (y=X} W, =0,non-basic)
Initial iteration :
C; (-1 3 -2 0)
Cg Y S
3 -1 2 0

12 -2 (4) 0 0 1 0
4
10
10 4 3 8 0 0 1 -3.33

0 1 -3 2 0
Since (Z,* -C) <0, the current basic feasible solution is not optimal.
The non-basic variable x enters into the basis and the basic variable
S leaves the basis.
First üeration :
3 -2 0 0)
Yp X S1 S2 S3
1
0 10
) 2 1
4
1
= 4*

3 3 0 0
4

0 8 1
S 2 4

-1 3
|3-G 99 2
0 2 0
4