DEPARTMENT OF MANGEMENT STUDIES

INDIAN INSTITUTE OF TECHNOLOGY, MADRAS

MS 3510 - Fundamentals of Operations Research
Problem Sheet 1 (Formulation of linear models)

1. A trucking company with Rs.20 crores to spend on new equipment is contemplating three types of
vehicles. Vehicle A has a 10-ton payload and is expected to cover 55 km per hour. It costs Rs.40
lakhs. Vehicle B has a 20-ton payload and is expected to cover 50 km pr hour. It costs Rs.65 lakhs.
Vehicle C is a modified form of B: it carries sleeping quarters for one driver, and this reduces its
capacity to 18-tons and raises the cost to Rs.75 lakhs.

Vehicle A requires a crew of one man, and, if driven on three shifts per day, could be run for an
average of 18 hours per day. Vehicles B and C require a crew of two men each. Where, B would be
driven 18 hours per day with three shifts, C could be run for an average 21 hours per day. The
company has 150 drivers available each day and would find it very difficult to obtain further crews.
Maintenance facilities are such that the total number of vehicles must not exceed 30. How many
vehicles of each type should be purchased if the company wishes to maximize its capacity in ton-km
per day?


2. A city's Police Department has the following minimal daily requirements for policemen:

Time of day Period Minimal Number of Policemen
(24-Hr.clock) required during period
-----------------------------------------------------------------------------------------------------------
2 - 6 1 20
6-10 2 50
10-14 3 80
14-18 4 100
18-22 5 40
22 - 2 6 30
----------------------------------------------------------------------------------------------------------

Note: Consider period 1 as following immediately after period 6.
Each policeman works eight consecutive hours.
Formulate a linear programming model to find an optimal schedule.


3. An airfreight company, which operates out of a central terminal, has 8 aircrafts of Type 1, 15 aircrafts
of Type 2, and 12 aircrafts of Type 3 available for each day's flights. The tonnage capacities (in
thousands of tons) are 45 for Type 1, 7 for Type 2, and 4 for Type 3.

The Company dispatches its planes to Cities A and B. Tonnage requirements (in thousands of tons)
are 20 at City A and 30 at City B; excess tonnage capacity supplied to a city has no value. A plane
can fly only once during the day.

The cost of sending a plane from the terminal to each city is given by the following table:

Type-1 Type-2 Type-3

City - A 23 15 1.4
City - B 58 20 3.8
------------------------------------
Formulate this problem as a linear programme.

4. An Advertising company has announced that it can optimally allocate its client's advertising rupees
by means of linear programming. The Company's approach is to identify the various target groups
sought by the client such as teenagers, young married couples, geriatric group, etc. Let the index i
represent the i
th
target group. The client must specify a desired level of exposure E
i
for each group i.
Then each advertising medium is estimated to have a certain effectiveness score for each of the
identified target groups. Let the index j represent the j
th
advertising medium, and a
ij
the effectiveness
score for the i
th
group, for every rupee allocated to the j
th
medium. Each decision variable is
designated as x
j
, which represents the total amount of rupees allocated to the j
th
advertising medium
during a sales promotion campaign. The client's objective is to minimize its total advertising
expenditure while still meeting its desired levels of product exposure.

Assume that there are three target groups and five advertising media. Write the linear programming
model implied by the above description.


5. The Capacity of a communication system is considered to be the maximum number of
call-kilometers that can be set up simultaneously in the system. The following system connects three
locations A, B and C (see Figure below)

The distance between any two of these locations is indicated in the figure:






























For example, two calls between A and C use two access lines from A to switching station plus two
from switching station to C, and count as 300 call kilometers. Formulate this problem of determining
the capacity of this system as a linear programming problem that finds the set of calls which yields
the largest total number of call-kilometers.

14 access
lines
8 access
lines
4
access
lines
50 km
Switching
Station
150 km
125 km
A B
C
6. An airline has to decide how many air-hostesses to hire and train over the next four months. The
requirements expressed as the number of hostess flight-hours needed are 8000 in September, 9000
in October, 7000 in November and 10000 in December.

It takes one month of training before a stewardess (hostess) can be put on a regular flight, so a girl
must be hired a month before she is actually needed. A trainee requires 100 hours of supervision by
experienced stewardesses during the month of training so that 100 hours less are available for flight
service by regular stewardesses.

Each experienced stewardess can work up to 150 hours in a month, and the airline will have 60
regular stewardesses at the beginning of September.

If the maximum time available from experienced stewardesses exceeds a month's flying and training
requirement, they work fewer than 150 hours. None is laid off.

By the end of each month, 10% of the experienced stewardess quit their jobs to get married or for
other reasons. The Airlines pays an experienced stewardess effectively Rs.80000/- per month
including other benefits and Rs.40000/- for a trainee.

Formulate the hiring and training problem as a linear programming problem.


7. A painting company has won a contract for finishing a large, newly constructed office building.

The company has three grades of paints P
1
, P
2
, P
3
and two types of thinners T
1
and T
2
all of which
can be blended together. The final blend must be such that only a single coat of paint will be needed.
The company has recognized constraints on factors such as viscosity, brilliance, drying time and
durability. It is known from experience that viscosity of the final blend of the Paint used should be, at
least, 350 centipoise. `Brilliance' depends on a constituent `B' - for suitable brilliance, the final blend
should contain between 2 to 10 grams of B per litre. Drying, which should be neither too fast nor too
slow to avoid peeling off and dirt settlement respectively, depends on the vapour pressure of the final
blend. For a proper drying time, the vapour pressure should be between 0.2 and 0.5 kg/sq.cm.
Finally, the durability depends on the ingredient `D' in the blend. The final blend should contain, at
least, 110 grams of D per litre.

A table listing the characteristics of the individual constituents that can be blended, is given below.
Assuming that all quantities of interest blend linearly by volume and that the total quantity of paint
required will be independent of the blend used, formulate a linear programme to determine the
minimum-cost paint blend.

P
1
P
2
P
3
T
1
T
2

-----------------------------------------------------------------------------------------------------------------------------------
Cost (Rs.Per litre) 14.00 11.00 8.40 6.60 4.00
Viscosity (Centipoise) 800 600 400 2 20
B-Content (Grams per litre) 4 10 2 0 0
D-Content ( " ) 300 200 100 0 0
Vapour pressure (kg/cm
2
) 0.3 0.06 0.09 1.20 0.80
----------------------------------------------------------------------------------------------------------------------------------`


8. A chemical engineering company manufactures two products, A and B. Both these products involve
two chemical operations for each. Each unit of product A requires 2 hours on operation 1 and 3
hours on operation 2. Each unit of product B requires 3 hours on operation 1 and 4 hours on
operation 2. Available time for operation 1 is 16 hours and for operation 2, 24 hours. The production
of B also results in a by-product C at no extra cost. A part of this by- product can be sold at a profit,
but the remainder has to be destroyed.
Product A sells for a profit of P
1
rupees per unit. Product B sells for a profit of P
2
rupees per unit. By-
product C can be sold at a unit profit of P
3
rupees, but if it cannot be sold it must be destroyed; the
destruction cost is P
4
rupees per unit. Forecasts by sales department show that a maximum of K
units of C can be sold; n units of C result for every unit of B produced.

Formulate a linear programme to determine the production quantities of A and B, keeping C in mind,
so that the total profits will be maximized.


9. Messrs. Ded Lee Fertilizers manufacture two products which give rise to effluents containing three
pollutants. The details are given below:

------------------------------------------------------------------------------------------------------------
Kilogram of Pollutant Legally permissible
Pollutant emitted by 1000 litres of limit of pollutant
PRODUCT-1 PRODUCT-2 (in kgs)
-----------------------------------------------------------------------------------------------------------

1.Pneumonium P
11
P
12
m
1

Stincoxide

2.Margayee P
21
P
22
m
2

cyanide

3.Kumattoxyl P
31
P
32
m
3

bisulphide
------------------------------------------------------------ -------------------------------------------
Profit per R
1
R
2

1000 litres

Sales commitment S
1
S
2

(min)
---------------------------------------------------------------------------------------------------------

Mr.E.Kolji, the production manager, has decided to use two antipollution devices to keep the pollution
within prescribed limits. The data regarding the devices are:

----------------------------------------------------------------------------------------------------------
Pollutant Percentage reduction of Pollutant by the use of
(Abbreviated names)
Device-1 Device-2
----------------------------------------------------------------------------------------------------------
1. PS u
1
v
1


2. MC u
2
v
2


3. KB u
3
v
3

----------------------------------------------------------------------------------------------------------
Profit reduction per
1000 litres r
1
r
2

----------------------------------------------------------------------------------------------------------

Note that the percentage reduction is the same for either product A device can be used to treat part
or whole or the production of a product. It is also possible that part of the production takes place
without the use of any pollution control device.
Formulate this as a linear programming problem to determine the optimal production plan.


10. The captain of a ship has the problem of allocation of jobs to sailors. Each sailor works for four days
during the week and works for a maximum of three consecutive days. A sailor works either in the
A.M. shift or P.M. shift throughout the week. Saturdays and Sundays are holidays. The number of
sailors required for the five days starting from next Monday is:









How should the available sailors be allotted so that maximum number can go on leave on all days of
the week? Formulate an LP.


11. Consider the following network where the numbers on the arcs connecting nodes i and j represent
the capacity of flow.

Formulate an LP to find out the maximum flow from 1 to 5




20 10

5

10


15 5 20







12. A Company has two grades of inspectors I and II who are to be assigned for a quality control
inspection. It is required that, at least, 1800 pieces be inspected per 8-hour day. Grade I inspectors
can check pieces @ 25 per hour with an accuracy of 98%. Grade II inspectors check pieces @ 15
pieces per hour with an accuracy of 95%.

The wage rate of a grade I inspector is Rs.40 per hour while that of a grade II inspector is Rs.30 per
hour. Each time an error is made by an inspector the cost to the company is Rs.20. The company
has available, for the inspection job, 8 grade I inspectors and 10 grade II inspectors. The company
wants to determine the optimal assignment of inspectors which will minimize the total cost of the
inspectors. Formulate a linear programme.


AM PM
Monday 80 60
Tuesday 70 90
Wednesday 40 60
Thursday 60 50
Friday 45 55
1
2
3
4
5
5
13. A private caterer has undertaken a contract for a series of dinners to be given by a club for the `elite
society'. There will be `n' dinners, one on each of the `n' successive days. The club has asked the
caterer to provide a special type of cloth napkins which become useless after the dinner series.

During the dinner series, however, soiled napkins can be washed and reused. Two types of laundry
service are available to the caterer Regular service and Express service. Regular service takes p
days; so a napkin sent on k
th
day to the laundry is available again on day k+p; this service costs u
rupees per napkin. Express service takes q days (q < p) and costs v rupees per napkin (v > u). New
napkin costs `s' rupees each. Let a
k
be the number of napkins needed on the k
th
day (1,2...........n).
The caterer wants to minimize the costs associated with purchasing and laundering napkins.
Formulate a linear programming model for determining the following daily decisions:

1. How many napkins to buy.
2. How many napkins to be sent to Regular laundry service.
3. How many napkins to be sent to Express Laundry service.


14. Messrs. Rooyilal Kapasji and Co. trade in raw cotton. They buy cotton from the growers and sell
them to mill owners for cash. The company has a godown with a capacity to hold H tons of cotton.
On the 1st April, the initial stock is I (I<H). The cash balance is K rupees. Buying and selling prices
are given below:-


Cotton price per ton Sale price per ton

April .... b
1
a
1


May .... b
2
a
2


June .... b
3
a
3


Cotton purchased in a month cannot be sold in the same month but only later. The company desires
a closing stock of F tons on 30th June. All transactions are strictly through cash (no credit).
Formulate a linear programming problem to optimize the buying and selling policies over the quarter.



15. Messrs. Aavakkai Pickle co., must contract for the purchase of mangoes from each of N growers.
These mangoes are then transported to M processing plants to be converted into pickles. The
purchase manager Mr.Gongu Rao, has collected historical data on the mangoes grown by different
farmers and has determined the varieties of Mangoes to expect from each grower.

There are V varieties of mangoes. Plant i requires R
ik
bushels of mangoes of type k, where i =
1,2.........M and k = 1,2......V. Let U
j
be the maximum number of bushels available from grower j and
p
jk
be the proportion of bushels from grower j that has mangoes of type k. (For example, if P
23
= .25,
then a quarter of each bushel purchased from Grower 2 falls in to Type 3 class).

Hence P
jk
0 and

V
k
jk
P
1
1

The cost per bushel of purchasing mangoes from Grower j is C
j
and the cost of transporting one
bushel from Grower j to plant i is C
ji
. Formulate an appropriate linear programme to determine the
optimal purchasing and distribution policies.