Sunteți pe pagina 1din 20

Quantitative Approach in Management

• Management by facts & figures

• Decisions based on use of specific


techniques
Measurement scales
 Nominal scale- Numbers are used for
identification purpose only e.g. car registration
number
 Ordinal scale- A kind of ranking where No. 1
may not be double as compared to No. 2
 Interval scale- Zero not important but
difference or interval is important e.g.
temperature on Fahrenheit thermometer
 Ratio scale- Here zero is important. E.g.
temperature on Centigrade thermometer. 440
C is double hot as compared to 220 C
Quantitative Techniques

 Linear Programming

 Queuing Theory
Linear Programming ( L P )
It deals with situations which require need for
utilisation of limited resources to best advantage -
• A bank wants to allocate its funds to achieve highest possible
returns. It must operate within liquidity limits set by regulatory
authorities – IRFC related problem
• An advertising agency wants to achieve best possible exposure for
its client’s product at lowest possible cost – PR organization related
problem
• A Food Administrator wants to prepare a high-protein diet at
lowest cost. There are 10 possible ingredients which provide protein ,
available at different prices - Railway Hospital Canteen related
problem
In Railways , maximum possible trains with available running
staff at a particular point , could be an LP situation
Linear Programming ( L P )
• In LP ,we either maximize (e.g. Profit)
or
minimize (e.g. cost)
an Objective Function
under given Constraints

• All factors involved have linear relationships, i.e.


doubling of labour- hours will result in double output.

• Non - negativity
LP Example – Maximisation – Formulation of problem.
 A factory produces 4 products - A,B,C and D.
Profits per product are - 5.5, 2.7, 6.0, 4.1
respectively
 It has 200 skilled workers and 150 Unskilled Worke
 Requirements
A B of labour
C Dper unit
( Aproduct
week hasis-
Skilled hours 5 3 1 8 40 working
Unskilled Hours 5 7 4 11 Hours )
• Let W, X, Y and Z are weekly output quantities of A,B,C and D
respectively
• Objective Function would be - Maximize 5.5 W + 2.7X + 6 Y +4.1 Z
• Subject to constraints 5W +3X + Y + 8Z < 8000 ( 200 X
40 )
5 W + 7X+ 4Y + 11 Z < 6000 ( 150 X40)

problem
A bakery has to produce 1000 kg high-protein biscuits .It must contain
4 ingredients A,B, C, and D which cost 8, 2, 3, 1 per kg. respectively.
 A, B, C and D contain Protein, Fats, Carbohydrates, Suger and Fillers
by weight in following percentages -

Protein Fats Carbohyd.Suger Fillers The batch must contain a


s minimum of 400 kg of
A 50% 30% 15% 5% 0 Proteins, 250 kg of Fats,
B 10% 15% 50% 15% 10% 300 kg of Carbohydrates
C 30% 5% 30% 30% 5% and 50 kg of Suger.
D 0 5% 5% 30% 60%
• Let quantities of A, B, C and D be W , X, Y and Z kgs respectively
•Objective Function would be - Minimize 8W + 2X + 3Y +Z
• Subject to Constraints - W+X+Y++Z > 1000
0.5W +0.1X + 0.3Y > 400
0.3 W +0.15 X + 0.05 Y +0.05 Z > 250
0.15 W + 0.5 X +0 .3 Y + .05 Z > 300
0.05W + 0.15 X + 0.3Y + 0.3 Z > 50
Steps to apply LP

• Decide whether maximisation or minimisation situation

• Identify decision variables

• Frame Objective Function

• Prepare equations denoting constraints

• Solve for optimum solution


L P – 2 variable – Graphical Solution – Maximization
• A manufacture produces 2 products A and B,
whose profits are Rs. 3 and Rs. 4 per Unit respectively .
• Production data -
Per Unit
Machining hours Labour hours Material (kg)

A 4 4 1
B 2 6 1
Total available 100 180 40
•Due to trade restrictions A can not be produced more than 20 Units in a week
and due to agreement with customer at least 10 units of B must be produced.
• Let quantity of A and B be X and Y respectively in a week
Objective Function is Maximize 3X + 4Y
Subject to constraints 4X + 2Y < 100 ; 4 X + 6Y < 180 ; X+Y < 40
X < 20 , Y > 10
Graphical Solution to LP Maximisation problem

X
Y >. 10
60 4X + 6 Y < 180

50
X + Y < 40
40
Optimal solution - X =15 , Y = 20
Max. profit = 3x15 + 4x 20 = 125
30
20 X < 20

10
4 X + 2 Y < 100

0 10 20 30 40 50 60 70 Y
LP- 2 variable graphical solution – Minimization
 A manufacturer wishes to market a new fertilizer, which should be mixture of
two ingredients A and B, which have following properties-

Bone Nitrogen Lime PhosphatesCost/ Kg


Meal
Ingredient20% 30% 40% 10% 12
A
Ingredient 40% 10% 45% 5% 8
•BFertilizer will be sold in bags containing a minimum of 100 kg
• It must contain at least 15 % Nitrogen , it must contain at least 8%
Phosphates and it must contain at least 25% Bone Meal
• Let quantities of A and B be X and Y Kgs respectively
•Objective function would be - Minimize 12X + 8Y
• Subject to constraints - X +Y > 100 ; 0.3X + 0.1Y > 15 ;
0.1 X + 0.05 Y > 8 ; 0.2 X + 0.4Y > 25
Graphical solution to LP Minimisation problem

X
0.2 X + 0.4 Y > 25
120

100 X + Y > 100


Optimum solution
80 X =60 and Y = 40
Minimum cost = 12x60 + 8x40 = 1040
60
0.1 X + 0.05 Y > 8
40
0.3 X + 0.1 Y > 15
20

0 20 40 60 80 100 120 140 160


Y
A carpentry company wants to add 2 new products in its range – a particular
size Door ( Product 1 ) and a particular size Window ( Product 2 ) . These
products are required to be processed in 3 Sections . Details of production
time taken by each product in each Section and total balance production
hours available in the 3 Sections are as follows –
Sections Production time Production
per product in time
Hours available
Find
per week in
Optimum
Product Hours Quantities
1 2 Of
Products
1&2
Let X and Y be quantities of Products 1 & 2 to be produced per week
Objective Function - Maximise 3X + 5 Y
Subject to constraints –
X < 4 , Y < 12 , 3 X + 2 Y < 18
Y
10 X= 4
9
8
2,6 Y=6
6

4 4,3
3X + 2Y = 18
2
2 4 6 8 X
LP
• Simplex method – A tabular method . Can deal with more than
2 variables
• Transportation method - Deals with multi-supply , multi-
consumption points situations to minimise transportation
• Integer programming – When solutions must be full numbers,
and not fractions
• Goal programming – When goal consists of several objective
functions to be satisfied
• Nonlinear programming – When relations are non-linear
• Use of software packages
Waiting Lines-Queuing Theory
Cost implications
Total cost

Cost of service

Costs

Cost of waiting

Optimum service

Level of service
Waiting Lines - Queuing Theory
• Concept of loss of business due to customers’ waiting
• Cost analysis of provision of faster servicing to reduce
queue length

• Marginal cost of extra provisioning during rush hours

• In case of railway passenger – his/her total time spent in


purchasing a ticket including time taken to come to
Booking Window
Important factors of Queuing Situations
• Arrival pattern – fixed time , random
• Service pattern – fixed time , random , increases with increase in no. of
customers
• Queue discipline ( how customers are selected for service ) – first in
first out , last in first out , random
• Customer’s behavior – waits always , doesn’t come if long queue
• Maximum number of customers allowed in the system – finite
capacity , infinite
• Nature of Calling Source - finite , infinite
Waiting Lines - Queuing Theory
Some formulae
• Let A be mean number of arrivals per unit time
and S be mean number of people or items served per unit time
•Then
Average number of customers in the line = A
S–A

Average time taken by the customer = 1


( including service time ) S- A

This is based on assumption that S and A follow


Poisson Distribution
Example of Car Servicing Station
•Let A = 2 cars per hour and S = 3 cars per hour
•So total time a car remains in Station ( incl. service time ) = 1 hour
•Actual waiting time = 1- service time ( 1/3 ) = 2/3 hour
•Let’s say cost of waiting ( Loss of goodwill ) is Rs. 100 per car per hour
•So, total loss in a day ( 8 working hours ) = 2( Arrival Rate )
x2/3(Actual waiting time)x8x 100 = Rs. 1066
•Let’s say , by some method S becomes equal to 4 cars per hour
•So, total time a car remains in Station = ½ hour
•Actual waiting time = ½ - ¼ = ¼ hour
•Hence loss in a day = 2x8x1/4 x 100 = Rs. 400
•Therefore it is worthwhile spending upto Rs. (1066 – 400 )
i.e. Rs. 666 per day to improve servicing time from
3 to 4 cars per hour .
End of presentation

S-ar putea să vă placă și