Video lectures by Prof Madhusudan Sohani.

Handout of questions and notes

Chapter 1 Linear Programming

A: Graphical Method
The method:
Graphical Method is a much simpler method but the limitation is that we can solve only a two
variable problem, since we can have only two axes, X axis and Y axis. In LPP practice they are
conventionally referred to as X1 axis and X2 axis. The horizontal axis is called as X1 and the
vertical axis is X2. While it’s not wrong if you interchange the two, it’s a standard practice to
follow the convention.
The methodology of plotting the graph will be explained at length in the lecture, but can
be summarised briefly here for reference:
 Identify all the constraints and count them.
 Each constraint will be plotted as a straight line by taking two suitable points
which will fall on the straight line. Now you must identify the area of the graph
which will satisfy the constraint. Typically the constraint will be <= or >=. (In rare
cases even =). The points to be plotted are to be taken in such a way that values
of the variables are not negative.
 After plotting all constraints as straight lines (by relaxing the original constraint
as ‘=’ temporarily), find the feasible area of the graph which will satisfy all the
constraints. This area is shaded in different colour and called as ‘solution space’
 The ‘Solution Space’ will comprise of a polygon which will typically be
quadrilateral, pentagon, hexagon etc and will be having various corner points.
 The next step is to find out ‘Z’ value or the objective function at each corner point
and find which points gives the optimal value (maximum or minimum, as the case
may be).
 Present the solution in a tabular format explained in the lecture that follows.
Q-01 A company manufactures two products, say A and B. The respective profits from these
products are Rs 10 and Rs 8 per unit. Both the products are required to be processed on two
machines which may be called as M1 and M2. Product A needs 6 hours and 2 hours of
processing time on M1 and M2 respectively. The hours needed by Product B are similarly 4
each on both the machines. Find the optimal number of units of these products that can be
made to maximise the profit of the company.
Q-02 A company makes two kinds of leather belts. Belt ‘A’ is a high quality belt and belt ‘B’ is of lower
quality. The respective profits are Rs 3 and Rs 4 from these varieties. Each belt of type A requires
twice as much time as belt B, and, if all belts were of type B, the company could make 1000 per
day. The supply of leather is sufficient for only 800 per day and both the belts require same
amount of leather. Belt A requires a fancy buckle of which only 400 are available per day and an
ordinary buckle needed for B type is available to the extent of 700 per day. Formulate this as a
linear programming problem and solve it using graphical method.
Q-03 (This is similar to the famous diet problem, but in a very simple form).
It is detected that for a healthy human being three vitamin supplements are very essential. These
vitamins an be called as 1, 2 and 3. Minimum 12, 12 and 10 units respectively of these are
required for good health. There is a choice of two foods available in abundant quantities wherein
these vitamins can be found. Let us call these foods as A and B. Food A contains 6, 1 and 2 units
of these vitamins 1, 2 and 3 respectively, whereas food B contains 1, 4 and 1 units in that order.
The cost of one packet of food A is Rs 4 and food B is Rs 8. Your problem is to find out how many
of the packets of these foods should be purchased so that the required quantities of these
vitamins are available at minimum cost. It may be presumed that any excess quantity of vitamins
consumed is harmless.
Q-04 A tour operator is planning a charter trip to a hill resort. The cost of tour package includes
the fare for round trip, surface transportation, food, stay, and site seeing. The trip is
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

restricted to 200 persons and as per past experience, there will be no difficulty in getting 200
persons for the trip. The problem of the tour operator is to determine the best number of
Deluxe, Standard and Economy tour packages for this charter tour. These three plans differ
according to the facilities offered to the customers in terms of quality of accommodation,
food plan, and other tour options. The following table summarises the estimated prices and
corresponding expenses for the travel agent.
Prices and costs for tour packages per person

Tour Plan Price (Rs) Hotel cost (Rs) Meal & other exp. (Rs)
Deluxe 10000 3000 4750
Standard 7000 2200 2500
Economy 6500 1900 2200
In deciding the numbers of persons in each category the following constraints must be taken
into account.
a. At least 10% packages must be of the deluxe type.
b. At least 35% but not more than 70% must be in the standard type.
c. At least 30% must be of the economy type.
d. The maximum number of seats for the deluxe package in the aircraft is restricted to
60.
e. The hotel requires that at least 120 persons should be on the deluxe and standard
packages put together.
The travel agent wishes to determine the number of packages to offer in each category so as
to maximize his profit.
Formulate the above as a linear programming problem in three variables.
Restate the problem in terms of two variables taking advantage of the fact that the total
number of packages is 200.
Solve the restated problem graphically as a two variable problem to determine the gross
contribution. If the cost of hiring aircraft is Rs 200000, what is the highest possible profit?
Q-05 A company produces two products P and Q. One unit of P requires 24 labour hours. One unit of Q
requires 32 hours. Minimum consumption of labour hours is 384. One unit of P requires 16 hours
of machine time and Q requires 20 hours. Availability of machines hours is 320. Minimum market
demand for P is 10 units and maximum market demand for Q is 8 units. Profit per unit of P and Q
Rs 15 and Rs 10 respectively. Solve by graphical method to find the optimal product mix.
(Unsolved question in the book by Prof Nitin Kulkarni OR for BMS)
Q-06 Maximize 6X1 + 20X2
Subject to:
2X1 + X2 <= 32
3X1 + 4X2 <= 80
X1 >= 5
X2 >= 6
I have a very versatile mobile app for Android phones (does not work on iPhone) which
can solve LPP by graphical and simplex method and the simplex tableaus will look exactly
like the ones shown in these video lecture series of mine. I recommend that you should
have it in your mobiles. It can also solve Transportation Problem and your manual
solution obtained can be verified. If you are interested in getting the app please write to
me at mdsohani@hotmail.com with the subject line as ‘OR Application- BMS’. The
application is no longer available in Google Play store in its original versatile form, and
can be obtained only from someone who has it.
Chapter 1 Linear Programming
B: Simplex or algebraic method
The method:
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

(Please ignore this note till at least 3 questions of LPP Simplex are viewed by you in my video
lectures) In any linear programming problem there are 3 possible types of it namely
≤, ≥
and
=
. Each type of constraint requires a different type of action by way of introducing slack,
surplus and artificial variables. For BMS however we shall restrict to solving numerical
problems involving only slack variables that is problems involving only ‘less than or equal to’
constraints, or <=.
Modify the various constraints by introducing the correct no. of slack variables. Also introduce
these variables in the objective function with zero contribution.
The contribution of decision variables will be known from the original objective function. (The
contribution of slack is always to be taken as zero.)
After modifying the original constraints with appropriate slack variables there will be always
a presence of an identity matrix after writing the information in a tabular form, as will be
seen from the first question that is Q-07 in the video lectures.
The order of variables in the table will always be
Decision variables first, followed by all slack variables. After performing the calculation of
index row (also known as Net Evaluation Row in some books), as shown in the video, put an
incoming arrow under the highest positive value for a maximization problem.
The incoming arrow indicates the column of a variable coming into the basis. This is called as
key column.
After identifying key column take the division of ‘Solution value’ upon value in the key column
for each row. This is called as the replacement ratio
If the divisions are negative ignore those rows. Out of positive divisions take the lowest value
& this row is the key row. The intersection of key column & row is called as Key number. Put a
box on the key no. Once the key column & key number is known we can find various key ratios
by putting key numbers in denominator.
The transformations for key row & non-key row are performed using the following formulae:
1) For key row: New Number = Old number upon the key number
2) For Non Key row: New number = (Old number) – (corresponding number of the key
row) * ( Key ratio)
Between 2 consecutive tables there will be only one change in the basis.
The key column variable will come in & the key row variable will go out. For faster
calculations please the video of any problem at least two times to capture the concept of
identity matrix effectively.
For a maximization problem the value of objective function must keep on increasing in
consecutive tables. (Exception: Degenerate problem at some stage)

Q-07 Maximise Z = 40 X1 + 35 X2 subject to:

2 X1 + 3 X2 <= 60
4 X1+ 3 X2 <= 96
X1, X2 >=0
Q-08 Maximise Z= 160X1 + 90 X2
Subject to:
8X1 + 3X2 <= 8
4X1 + 9X2 <= 10
X1, X2 >=0
Q-09 A firm uses three machines in the manufacture of three products. Each unit of product A
requires 3 hours on machine I, 2 hours on machine II and 3 hours on machine III. Each unit of
product B requires 4 hours on machine I, one hour on machine II and 3 hours on machine III.
Each unit of product C requires 2 hours on each of the machines. The contribution of the
three products A, B and C is Rs 30, Rs 40 and Rs 35 per unit respectively. The firm desires to
maximise the contribution. The machine hours available on three machines I, II and III are 90,
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

54 and 93 respectively. Formulate the above as a linear programming problem. Using the
Simplex method, obtain the optimal solution.
Which of the three products shall not be produced by the firm?
Q-10 Maximise 4X1 – 2 X2
Subject to:
X1 + 3X2 <= 6
X1 – X2 <= 2
X1, X2 >=0
Q-11 (Problem with multiple solutions)
Maximise Z = 3X1 + 2X2 + 5X3
Subject to:
X1 + 2X2 + 2X3 <= 8
3X1 + 2X2 + 6X3 <= 12
2X1 + 3X2 + 4X3 <=12
All variables non-negative
(Solved question in the book by Prof Nitin Kulkarni OR for BMS)
Q-12 Maximise Z = 3X1 + 5X2
Subject to:
X1 + X3 = 4
2X2 + X4 = 6
X1 + 2X2 + X5 = 12
X1, X2, ------- X5 >=0
(Mumbai University April 2006)
Q-13 The following LPP needs to be solved by Simplex algorithm. This is a product mix problem
wherein there are 3 products represented by the 3 variables respectively. X1, X2 and X3
represent the number of units of the various products manufactured per month.
Maximise 3X1 + 2X2 + 5X3
Subject to:
X1 + 2X2 + X3 <= 430 --- Availability of raw material
3X1 + 2X3 <= 460 --- Availability of electrical power
X1 + 4X2 <= 420 --- Availability of packing material
X1, X2, X3 >=0
Q-14 Maximise Z= 3X1+ 4X2+ X3
Subject to:
X1 + 2X2 + 3X3 <= 90
2X1 + X2 + X3<= 60
3X1 + X2 + 2X3 <= 80
X1, X2, X3 >=0
Q-15 Only formulate the problem as an LPP. Do not solve the same.
A scrap metal dealer has received an order from a customer for at least 2000 kgs of scrap
metal. The customer requires that at least 1000 kgs of a high quality metal called ‘Alpha’
must be present in the total scrap to be supplied, since he wants to use this ‘Alpha’ for
producing downstream products of high quality. Further the customer has stipulated
that if the total scrap supplied contains more that 175 kgs of a metal that he deems unfit
for commercial use, he will not accept the order. The dealer can purchase scrap from two
different suppliers ‘A’ and ‘B’ in unlimited quantities with the following percentage
composition (by weight) of Alpha and unfit scrap.

Supplier A Supplier B
Alpha 25% 75%
Unfit Scrap 5% 10%
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

Supplier A and B are willing to supply the scrap at Rs 1 and Rs 4 per kg respectively,
as per these compositions. Formulate this as a linear programming problem to
minimise the cost of scrap purchase.
Q-16 A manufacturer has three products A, B and C. These are produced on three machines M1,
M2 and M3. The processing time required by these products are as under:
Product Processing time per unit
M1 M2 M3
A 3 2 1
B 2 3 -
C 2 3 -
Available Capacity 240 270 60
per week (hrs)
The Product A, B and C give a profit of Rs 10, 6 and 6 per unit. Formulate and solve the
problem for maximum profit. (Solved question in the book by Prof Nitin Kulkarni OR for BMS)
Q-17 Only formulate the problem as an LPP. Do not solve the same.
An animal feed company must produce 200 kg of a mixture per day consisting of ingredients
‘A’ and ‘B’. ‘A’ costs Rs 3 per kg and ‘B’ Rs 8 per kg. No more than 80 kgs of ‘A’ can be used
and at least 60 kgs of ‘B’ must be used. Formulate this as an LPP to find out how much of
each ingredient should be used if the company wants to minimise the cost.
Q-18 Do not solve the following problem, but explain the concept of surplus and artificial variables
using the example below. Rewrite the problem in standard form after introducing the
required variables
Minimise Z = 2X1 + 4X2
Subject to:
2X1 + X2 >= 18
3X1 + 2X2 >= 30
X1 + 2X2 >= 25
X1, X2 >=0
Q-19 Maximise Z= 22X1 + 30X2 + 25X3
Subject to:
2X1 + 2X2 <= 100
2X1 + X2 + X3 <= 100
X1 + 2X2 + 2X3 <= 100
X1, X2, X3 >= 0
(A special case of LPP, a degenerate problem)
Q-20 Do not solve the following problem, but explain the concept of duality in LPP and write the
dual of the problem stated here.
Maximise 20X1 + 10X2
Subject to:
X1 + 2X2 <= 40
3X1 + X2 <= 30
4X1 + 3X2 <=60
X1, X2, X3 >=0
Q-21 Mumbai University April 2011 – Reworded where necessary
A business problem is formulated and solved using LPP algorithm. X1 and X2 are production
volumes of products A and B respectively. The resources required for producing these
products are R1 and R2. Total profit is indicated by Z.
Maximise Z= 10X1 + 4X2 subject to:

20X1 + 10X2 <= 1200 --- R1

40X1 + 10X2 <= 1600 --- R2
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

X1, X2 >=0
Simplex algorithm yielded the following solution.
Cj 10 4 0 0
C Basis X1 X2 S1 S2 RHS
0 S1 0 5 1 -1/2 400
10 X1 1 1/4 0 1/40 40
a. Please improve the solution till optimality
Study the optimal solution and answer the following questions
b. Is the solution found by you infeasible?
c. Is there a multiple solution? If so find it.
d. Interpret the complete solution. What is the product mix and profit at optimality?
e. What is the percentage of resourse utilisation?
f. If one unit of R2 becomes unavailable, what is the reduction in maximum profit?
Q-22 The following product mix problem has been solved by Simplex algorithm and the solution
follows. The values at the right hand side of the constraints indicate the available capacities
in various departments
Maximise 30X1 + 40X2 + 35X3
subject to:
3X1 + 4X2 + 2X3 <=90 (Machine 1 hours available)
2X1 + X2 + 2X3 <=54 (Machine 2 hours available)
3X1 + 3X2 + 2X3 <=93 (Machine 3 hours available)
X1, X2, X3 >=0

Ci Basis X1 X2 X3 S1 S2 S3 RHS
40 X2 1/3 1 0 1/3 -1/3 0 12
35 X3 5/6 0 1 -1/6 2/3 0 21
0 S3 1/3 0 0 -2/3 -1/3 1 15

Answer the following questions based on the optimal tableau above:

a. Is this solution optmial?
b. State and interpret the complete optimal solution.
c. What will be the reduction in the Objective Function value if 2 units of product 1
(represented by X1) are made to meet a contractual commitment?
d. What are the shadow prices of one hour of time in each of the 3 machines?
e. What do you understand by values of dual variables? Write the dual of the original
LPP here as a minimisation in 3 variables. What will be the value of objective function
of the dual problem at optimality?
f. What is the marginal value of resources in the primal problem?

Chapter 2 Assignment and Transportation Problem

A: Assignment Problem
The method:
STEPS BEFORE APPLYING THE ALGORITHM (Pre-checks)
Examine whether the given problem is for maximization or minimisation.
- Profit, sales, production, efficiency, happiness, satisfaction are to be maximised.
- Cost, interest payout, time taken, distance travelled, layover time, wastages are to
be minimised.
- If the problem is for maximization it should be converted to minimization by
subtracting all the nos. from the highest no.
- After this check whether the problem is balanced, if not, add the requisite number of
rows or columns to balance the same. All values in such rows or columns should be taken
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

as zero.

STEPS FOR ASSIGNMENT ALLGORITHM

1) Subtract row minimum
2) Subtract column minimum
3) Try making assignment firstly at unique zeros & after unique zeros are over make
assignment at non unique zero. (For making the necessary assignments, follow some structured
approach such as starting left to right, and top to bottom.)
After each assignment cancel remaining zeros in that row or in that column where
assignment is make.
Count no. of assignments. If the number is less than the matrix size, go to the next iteration
by following steps (1) to (5) given in the note below. After obtaining the minimum number of
vertical and horizontal lines for covering all zeros, find the lowest value of uncovered
element (no lines). This value is added at the intersections, and subtracted from elements
which have no lines. Elements which have one line will be left intact.
Make another attempt to obtain the necessary number of assignments, if not repeat the steps
above. Present the final answer in a tabular form.

A Short note on the procedure for drawing minimum number of lines through the Hungarian
Matrix for cancelling all the zeros.
After making an attempt to make assignments, you may not get the required number of
assignments and the number of assignments may be less than the size of the matrix.
This suggests that one more iteration is necessary by modifying the present matrix. At this stage,
you need to draw the minimum number of lines for covering all the zeros. It is not a matter of guess
work to draw the minimum lines. There is a systematic procedure described in the book by
Satyanarayan and Lalitha Raman. The summary of the same is given here.
Step 1. Tick the rows having no assignments
Step 2. Tick the columns having zeros in the ticked rows
Step 3. Tick the rows having assignments in the ticked columns
Step 4. Go back to step number 2 and 3 above, till no more ticking is possible.
Step 5. When ticking is over, draw lines through ‘ticked’ columns and ‘un-ticked’ rows
The required number of lines will always be less than the size of the matrix.

Q-01 A company plans to assign 5 salesmen to 5 districts in which it operates.

The estimates of sales revenue in thousand rupees for each salesman in different districts are
provided in the following table. Find the optimal assignment of salesmen to districts to
maximize the sales. Find one more optimal solution with the same sales level.
D1 D2 D3 D4 D5
S1 40 48 49 30 37
S2 46 32 35 46 41
S3 48 36 41 49 48
S4 36 29 38 44 43
S5 48 44 45 44 47
Q-02 A, B, C and D are four locations in a city where one excess car each is available with a tourist
operator. At his other locations namely P, Q, R,
The distance Matrix: (km)
S and T one car each is required. A, B, C, D, P, Q,
R, S and T are all located in the same metro city
where he operates from. The matrix for distances between locations is available. You need to
find out which of the four places out of five can be provided with a car from the four places,
which have an excess car. What will be the minimum distance travelled by the four cars to
meet the requirement?
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

P Q R S T
A 2 3 10 9 10
B 5 8 1 3 11
C 4 2 3 5 13
D 2 1 5 4 9
Q-03 Like Q-1 above, again 4 salesmen are to be assigned to 4 districts. The best performance is of
Salesman S1 in District D4. For some personal reason, he has requested not be allocated to
District D4. If the numbers in the following matrix are sales in thousand Rupees, will this
request of S1 result in reduction in total sales level? If so, by how much?
D1 D2 D3 D4
S1 20 15 17 22
S2 16 12 15 14
S3 19 20 19 20
S4 18 18 16 19
Q-04 ABC Company Limited has taken the third floor of a multi-storeyed building for rent with a
view to one of their Zonal Offices. There are five main rooms on this floor to be assigned to
five managers. The rooms have their own advantages and disadvantages. Some have
windows, some are closer to the wash rooms or to the canteen or to the secretarial pool. The
rooms are of different sizes and shapes. Each of the five managers were asked to rank their
preferences amongst the rooms 301, 302, 303, 304 and 305. Their preferences were recorded in
a table as indicated below:
Manager
M1 M2 M3 M4 M5
302 302 303 302 301
Room 303 304 301 305 302
304 305 304 304 304
301 305 303
302
Allocate the rooms to the managers for highest satisfaction.
Q-05 A company solicits bids on each of the four projects from five contractors. One one project
may be assigned to any contractor. The bids received (in thousand Rupees) are given in the
accompanying table. Contractor D feels he is unable to carry out project 3 and therefore
submits no bid.
Contractor
s
A B C D E
Jobs
1 18 25 22 26 25
2 26 29 26 27 24
3 28 31 30 - 31
4 26 28 27 26 19

Use the Hungarian method to find the optimal assignment for minimum cost
What is the minimum total achievable cost?
Q-06 An engineering company has branches in Mumbai, Calcutta, Delhi and Nagpur. A branch
manager is to be appointed, one at each city, out of four candidates A, B, C and D. Depending
on the branch manager and city, the monthly business in the city varies and given below in
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

lakhs of Rupees as per details below. Suggest which manager should be assigned to which
city so as to maximum total monthly business. (Mumbai University, April 2012)
City

Mumbai Calcutta Delhi Nagpur

Manager
A 2 3 1 1
B 5 8 3 3
C 4 9 5 1
D 8 7 8 4
Q-07 A company has four sales executives which need to be appointed on one to one basis in 3
sales territories . Profit potential for each ‘executive-territory’ combination is given in
thousand Rupees. Find optimal assignment of executives to territories to maximise total
profit. Which sales executive will remain idle?
Rupees in
000's
T1 T2 T3
E1 60 67 90
E2 80 83 95
E3 70 72 82
E4 85 95 110
Q-08 In a factory five jobs are to be assigned to to five workers on one on one basis. The cost in
hundred Rupees for each employee job combination is given. Find the optimal assignment of
employees and jobs and the optimal cost.
In hundred Rupees
J1 J2 J3 J4 J5
W1 28 22 26 20 27
W2 17 16 18 20 22
W3 11 13 14 16 17
W4 13 12 11 8 10
W5 11 10 6 8 9
Q-09 A Transport Company provides daily bus service between Kolkata and Digha. It takes six
hours for a single one-way trip. The drivers and conductors can be provided with resting
facilities at either of the places i.e. Kolkata & Digha. The crew require a minimum of four
hours before undertaking the return trip. From the following timetable of bus services,
prepare an assignment schedule for routes assignment for minimising the waiting time. Also
indicate where the crew should be based for each pairing so obtained.
Departure Arrival Departure Arrival
Route No. Route No.
Kolkata Digha Digha Kolkata
A1 500 1100 B1 400 1000
A2 700 1300 B2 500 1100
A3 900 1500 B3 1000 1600
A4 1600 2200 B4 1300 1900
A5 1700 2300 B5 1500 2100

Q-10 The quantity of different products in units produced by the workers per day is given in the
following matirx alongwith the profit per unit. Formulate a profit matrix and find the optimal
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

assignment of workers to productwhich will maximise the profit. Find the optimal profit.
Pencil Rubber Pen Ink
Workers
Amit 30 40 100 50
Sumit 25 70 140 30
Vinit 40 90 130 60
Punit 35 45 120 40
Profit in Rs.
per unit 4 2 1 3

(Mumbai University May 2018)

Chapter 2 Assignment and Transportation Problem
A: Transportation Problem
The method:
TRANSPORTATION PROBLEM ALGORITHM
A TP is solved in 2 Phases.
The purpose of Phase I is to obtain IFS or BFS which is the starting point for Phase II.
There are 3 different methods for Phase I to obtain IFS.
PHASE 1
1) NWCR
2) MMM/LCM
3) VAM [VOGEL’S APPROXIMATION METHOD]
IFS / BFS can be obtained using any one of the 3 methods above. However, VAM is
usually the most efficient method.

STEPS BEFORE APPLYING THE ALGORITHM

Examine whether the given problem is for maximization or minimisation.
- Profit, sales, production, efficiency, happiness, satisfaction are to be
maximised.
- Cost, interest payout, time taken, distance travelled, layover time, wastages
are to be minimised.
- If the problem is for maximization it should be converted to minimization by
subtracting all the nos. from the highest no.
- After this check whether the problem is balanced. That is whether the demand is
equal to the supply. If it is not the same, increase the demand or supply as the case may
be by adding a column or a row. All cost values in such balancing rows or columns
should be taken as zero.

NWCR
Make the 1st allocation in the North West corner (in cell O1D1) of maximum possible
quantity.
This will result in deletion of either the row or the column. After making adjustment
to the demand supply values make next allocation in North West corner of remaining
matrix. Complete the allocations this way.

MMM
Examine the entire matrix for lowest cost. If two lowest costs are identical, choose that
cell where allocation can be maximum. Delete the row or column as case may be &
adjust demand supply equation. Make next allocation in the next lowest cost of the
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

remaining matrix & go on doing the same.

VAM
This method does not aim at minimizing cost directly. It aims at minimizing cost
through the method of minimizing penalty. Penalties are defined as the difference
between two lowest cost in a row or a column.
Since, the method aims at minimizing penalty, choose the highest penalty & tick the same.
If it is a row penalty make allocation in the lowest cost in the row.
If it is a column penalty make allocation in the lowest cost in the column.
After doing this either a row or column will get deleted.
If a row is deleted, rewrite the column penalties, if a column is deleted, rewrite the
row penalties.
If 2 or more penalties are tied, choose that penalty which results in lower cost is
chosen first. If the lower cost is also tied, choose that penalty which results in more
allocation.
Go on doing this till all the rows & columns are deleted.
After obtaining the feasible solution count the number of allocations. You should get
((m + n) - 1) allocations for a non degenerate solution (where m = no. of rows, n = no.
of columns)
You will never get more than (m + n)- 1 allocations. If you get less than m + n – 1 the
solution is degenerate. Increase the no. of allocations to m + n – 1 by making new
allocation of a hypothetical quantity epsilon in one of the un-allocated cells, which
will enable you to complete the allocation of row and column costs.

ui = row cost
vj = column cost

PHASE II
The solution obtained in Phase I by using any of the 3 methods needs to be checked
for optimality. This is done in Phase II. After ascertaining that you have the required
number of allocations (that is, you have a non degenerate IFS), break up the cost of
each allocated cell into 2 components; A row cost + A column cost.
You may start allocating the row cost & column cost, by starting in a row or column
having maximum allocations. Sometimes it is convenient to start in a row or column
with all zero costs.
Check the unallocated cells for advantage if any.
An un-allocated cell has an advantage if the cell cost is lower than the total of row cost &
column cost. Identify cell with highest advantage. This cell needs to be brought into your
next solution.
Draw a loop comprising of + - starting with a cell of highest advantage. The loop may
be drawn by trying to travel vertically or horizontally, alternately. There is only one
correct way of drawing a loop for any un-allocated cell.
At each corner of the loop you must have an allocation.
The loop is drawn by travelling vertically & horizontally alternatively.
Check the new solution for optimality if not go to the next iteration.
After reaching the optimal solution check for multiple solution if any. Furnish the
final answer in a tabular form.

Q-01 Obtain the Phase – I solution for the following transportation problem for minimising the
cost of transportation if the numbers in each cell represent unit cost of transportation.
Illustrate all three methods to obtain the cost and compare the total cost of transportation
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

under each of the 3 methods.

7 10 5 90
12 9 4 50
7 3 11 80
9 5 7 60
120 100 110

Q-02 Solve the following Transportation Problem. Obtain the IFS using VAM,
and thereafter obtain the Optimal Solution using MoDi (Modified Distribution) method. The
numbers in side the cell are unit cost of transportation. What is the optimal cost of
transportation? (Note: This is a classic problem from the book of M. Satyanarayanan and
Lalitha Raman. It has got a lot of learning value)

19 17 17 20 18
100
20 21 22 19 23
125
17 18 20 17 18
175
60 80 85 105 70

Q-03 There are 3 factories A, B and C, and three markets X, Y and Z. Supply at the factories is 60,
80 and 85 units.Deamnd at the market places is 75, 110, 40 units.The supply and demand of
units and the unit cost of transportation (in Rs), and the schedule followed from factories to
markets are given below. The numbers shown in bold indicate number of units transported
from factories to markets. The questions are:
 Test the solution for optimality and find the optimal transportation schedule
 Find one more alternative optimal solution
 Comment upon the managerial significance of the multiple / alternate optimal
solution.
Markets / Sup
Factories X Y Z ply
A 6 3 5 60
35 25
B 5 2 2 80
40 40
C 12 7 8 85
85
Demand 75 110 40 225

Mumbai University April 2018

Q-04 As a financial Consultant, you have been assigned the job of sourcing short-term loans for
various construction projects undertaken by ‘Diamond Contractors’.
Various banks have shown interest in providing short-term loans for the projects. However
depending upon each bank’s average and marginal cost of funds, their perception about the
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

risk and viability of the projects, they have offered loans at different rates of interest for the
various projects. The following table summarises the rates offered by banks, the total loan
availability with each bank and loan requirement for each project. Obtain Initial Feasible
Solution by Vogel’s Approximation Method and after checking for optimality, find the best
pattern for financing the projects and the total annual interest outgo for Diamond
Contractors. (For an individual project, the interest outgo is amount of the loan multiplied by
the rate of interest in percentage divided by 100). If there are more alternative solutions,
find at least one more optimal solution.
Rate of Interest in %

Project
Type of Bank P Q R S T Loan available
(Rs in thousands)
Foreign Bank 20 18 18 18 17 100
Co-op Bank 16 16 16 15 16 400
Nationalised Bank 15 15 15 13 14 250

Loan Required 200 150 200 125 75

Q-05 A construction firm in USA has supply centres located in Atlanta, Chicago,
and New York City. These centres have available 40, 20, and 30 units of a particular resource,
respectively. The firm's job sites have asked for the following number of units: Cleveland, 25;
Louisville, 10; Memphis, 20; Pittsburgh, 30; and Richmond, 15. The shipping cost per unit in
dollars between each centre and job site is given in the following table:

To To To To To
Cleveland Louisville Memphis Pittsburgh Richmond
From Atlanta 55 30 40 50 40
From Chicago 35 30 100 45 60
From New York 40 60 95 35 30
Solve this as a Transportation Problem, by obtaining an Initial Feasible Solution (IFS) using
the Vogel’s Approximation Method (VAM). Check for optimality of the solution and show
how it could be improved by using MODI method.
The objective is obviously to minimise the total shipping cost.

Q-06 There are 3 warehouses A, B, C and four market places P, Q, R and T. Supply at warehouses is
8, 9, and 13 units. The demand at market places is 8, 9, 6 and 7 units. The following table
shows the unit cost data of transportation and a feasible solutionto the problem.
a. Test the given solution for optimality
b. If it is not optimal, find the optimal solution using MoDi method and calculate
the optimal cost of transportation.
From the optimal table answer the following questions
c. If 3 units are transported from B to P, how will the cost be affected?
d. If the transporter from A to R is prepared to reduce the cost by 20% even if one
unit of business is given to him, should the offer be accepted?
e. If the management wants to embark on an advertisement campaignin one of the
market places, which one should be selected?
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

f. If the transport cost from C to R decreases from Rs 11 to Rs 4, what will happen

to the solution?
g. What should be decrease in the cost of the cell A-R so that it can be used
alternatively so that multiple solutions are possible?
SUPP
P Q R S LY
19 17 18 17 8
A 8
11 12 5 9 9
B 6 3
7 9 11 12 13
C 8 1 4
Dema
nd 8 9 6 7

(Mumbai University April 2018, question of 15 marks)

Q-07 A company manufactures a heavy piece of equipment at all three of its plants. Depending
upon the differences in local conditions, wages, proximity to raw material supply, power
tariffs etc. the cost of manufacturing is different at each plant. There are four main
machinery markets in metro cities where the equipment is supplied. Based on the local
factors as regards number of competing manufacturers, show room expenses etc. the selling
prices at all these markets are also different. The company is desirous of working out an
optimal dispatch programme which will result in highest profit levels. The relevant data
about the cost of manufacturing, capacity at each plant, demand form various markets and
shipment cost are available and are given below.
Plant Capacity in units Cost of manufacturing
(Thousand Rupees per
piece)

A 15 70
B 25 75
C 10 80
Shipment Cost in thousand Rupees per machine from plant to market
Market  M1 M2 M3 M4
Plant


A 25 17 45 41
B 22 17 29 45
C 9 19 31 38
Demand and selling Price at Markets
Market Selling Price in Demand
thousand Rupees per
machine
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

M1 110 5
M2 110 15
M3 120 15
M4 125 15
You are required to formulate the above as a Transportation problem, obtain an
initial feasible solution by Vogel’s method and improve the initial solution for
Maximum profit.

Q-08 Obtain the IFS for the following Transportation Problem for cost minimisation and obtain the
optimal solution thereafter.
4 6 8 8 40

6 8 6 7 60

5 7 6 8 50

20 30 50 50

Q-09 Solve the following transportation Problem for cost minimisation where the table has the
usual meaning:
8 6 4
10
10 12 2
8
12 8 6
5
6 10 8
6
7 12 4
Q-10 Solve for cost minimisation

D1 D2 D3 D4
O1 6 5 1 3 100
O2 4 8 7 2 125
O3 6 3 9 5 75
70 90 80 60

Q-11 a) Consider the following solution to a transportation problem obtained by experience.

Evaluate if the solution is optimal.
b) Is the solution degenerate?
c) Is there an alternative solution?
d) Contractor on route O2D4 is offering a discount of 20% on his existing rate. Should
the offer be accepted?
e) As a result of previous court case being withdrawn, the management has decided to
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

accept offer from contractor of route O1D2 to a 50% discount on present, but as per
terms maximum possible load has to be given to him. What will be the financial
implications of this decision?
D1 D2 D3 D4
O1 10 10 16 20
300
O2 16 6 17 25
175 25
O3 8 21 10 15
25 75 150

Q-12 XYZ Company has three plants and four warehouses. The supply and demand in units and the
corresponding transportation costs are given. The table below shows the details taken from
the solution procedure of the transportation problem:
Warehouses
I II III IV Supply

Plants A ∟5 ∟10 10
∟4 ∟5 10

B
∟6 ∟8 ∟7 ∟2 25
20 5

C
∟4 ∟
2 ∟5 ∟7 20
5 10 5

Demand 25 10 15 5
Answer the following questions. Give brief reasons:
Check for optimality
Can you get an alternative solution?
If you have to choose the next best cell for allocation, which would you choose and why?
Which is the worst cell for making allocation, and why? Justify the answers by brief
calculations

Q-13 Solve the following transportation problem for profit maximisation where the profit numbers
are gven inside each cell at top right corner and allocations are shown in bold font.
Obtain IFS by using all the three methods namely NWCR, LCM and VAM. Compare all the
three solutions. Take the best solution to phase 2 and obtain the highest possible profit.
Present the solution in a tabular form.
D1 D2 D3
O1 58 56 60 20

O2 50 54 46 20

O3 70 74 76 20

15 30 15
Chapter 3: Network Analysis
A: Critical Path Method
The method:
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

The procedure for drawing the network diagram, numbering of events using
Fulkerson’s Rule, the process of forward path, backward path, finding critical path &
finding the following floats:
A. Total float
B. Free float
C. Interference float (also head event slack)
D. Independent float
The procedure for drawing network diagram:
- Firstly identify the initial & end activities of the project. Activities which do
not depend on any other activities are initial activities. They can be identified if they
have a blank against their name under column preceding activity. The last activities
are those on which, nothing depends. Thus, the initial & end activities are now
known.
- Activities have alphabetical names or description. Events have numbers.
- After drawing the initial activities from the start event go to the top of the table &
start drawing activities by reading the table row by row. Care must be taken to ensure
that the activities in the 1st column actually depend on activities in the 2nd column but
care must also be taken to ensure that the activities in the 1st column do no depend
unnecessarily on any other activity not mentioned in 2nd column.
To ensure this dummy activities may have to be used from time to time.
Dummy activities do not consume any resources, do not represent anything really
happening, have duration 0 days & they are required to be used only to show the
dependency relationship between activities of 1st & 2nd column.

Every network drawn correctly must have the following characteristics:

1. There should be only one start event where no activities are ending.
2. There should be only one end event where no activities are starting.
3. There should be no loop formed in a network diagram (A loop can be
identified if there is an event from where you can start travel in the direction of
arrows & are able to come back to the same event. If a loop has been formed in the
network diagram either there is a mistake in drawing the diagram or there is a
mistake in the logic of the network).

Procedure for numbering the events:

The events are numbered in a network diagram using following procedure:
i) Count the no. of events i.e. circles. Out of these events there would be
only 1 start event & only 1 end event.
ii) The event where activities are only starting has to be identified first.
This is event no. 1.
Also identify the last event where no activities are starting. Thus, the
1st & last events been identified. The in between events will have to be numbered in
such a way that all the activities are starting at a lower numbered event & ending at a
higher numbered event.
This will include even dummy activities.
iii) After identifying the 1st event & giving it no. 1, delete all the activities
starting from this event i.e. event no. 1.
This will result in atleast 1 more event emerging as a start event. This
is to be numbered next i.e. 2.
If more than 1 event emerges as a new start event usually the event
appearing at a higher level in the diagram will be numbered first.
After numbering every event delete activities starting from the events
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

numbered. At every stage atleast 1 or more start event will emerge. This way
continue numbering till the end event. The rough diagram can be ideally used for
deleting activities from events already numbered.
iv) Write the duration of activities respectively.

The Process of Forward Path

The purpose of forward path is to find the earliest time for an event.
We use the following notation to denote any activity of a network diagram.

ei d ej
i j
li lj

i = start event of an activity

j = end event of an activity
ei = earliest time when event i can happen
li = latest time by which event must happen
d = duration of activity
ej = earliest time for event j
lj = latest time for event j
Forward path has to be carried out starting at event no. 1 & going sequentially in the
order 1…2….3….4……n.
After starting at event no. 1 with earliest time as 0 we go sequentially & ask a
question (which are the activities ending at this event?).
If more than 1 activity is ending at an event the earliest time for this event would be
the largest of the numbers obtained by adding the duration of the activity to the
earliest time of previous event.
After reaching the last event in this process, the project duration can be known.
This is also the length of longest path from the 1st to the last event.
At the last event after calculating ej the same number is written as lj.
After writing lj for the last event the process of backward pass commences.
In this process we go in the order n, n – 1, n – 2, …..3, 2 and 1.
The question to be asked in the backward pass: which activities start from this
event?
If more than 1 activity starts at an event pick up the smallest of the numbers as the
latest time for the event. At the last & the first event the earliest & the latest times
will be same. In respect of other events this may or may not be the case.

Finding Critical Path

After performing the calculations for earliest & latest time for each event through the
above process for forward & backward pass we define the freedom or flexibility in
respect of event/activity. Slack is a measure of freedom/flexibility for an event.
Float is a measure of freedom/flexibility for an activity.
Formulae
TF = (lj – ei) – d
FF = (ej – ei) – d
IF/HES = TF – FF
Independent float = (ej – li)
Prepare the float calculation as per table shown in the video
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

Q-01 The time estim

ates of different activities of a project are given below:
Draw the network diagram, number the events and find the critical path.

Duration
Activity Predecessor activity
Days
A - 6
B - 4
C A 3
D B 8
E B 14
F C, D 8
Q-02 The time estimates of different activities of a project are given below. Draw the network
diagram, number the events and find the critical path. Find the total and free float for all
non-critical activities.
Activity Preceding activity Duration
Days
A - None - 4
B - None - 5
C - None - 2
D A 7
E C 2
F A 4
G B, D, E 8
Q-03 A small project is estimated to have the following durations for its various activities. You are
required to find the critical path, duration of the project and the total and free float for each
activity.

Activity A B C D E F G H I J K
Act. 1-2 1-3 1-4 2-5 3-5 3-6 3-7 4-6 5-7 6-8 7-8
Nodes
Duration 2 7 8 3 6 10 4 6 2 5 6
(days)

Draw the following various network diagrams, number the events where event numbers are
not given, find the duration of the project, critical path and total float, free float, interference
float, and independent float.
Q-04 Activity A B C D E F G
Preceeding
Activity - - A A,B C D E,F
Duration (days) 2 5 3 3 4 2 2
Q-05 Activity A B C D E F G
Preceeding
Activity - - A B C,D D E,F
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

Duration (days) 5 8 3 4 6 4 3
Q-06 Activity A B C D E F G H I J
Preceeding
Activity - - - A B C C D E,F G
Duration (days) 5 2 3 6 4 3 4 4 3 4
Q-07 Activity A B C D E F G H
Activity 1-2 1-3 2-4 2-5 3-5 4-6 5-6 6-7
Duration (days) 5 7 3 4 3 2 4 1
Q-08 The time estimates of different activities of a project are given below. Draw the network
diagram, number the events and find the critical path. Find the total and free float for all
non-critical activities.
Activity Predecessor Duration
A - 3
B - 8
C A 9
D A 6
E C 10
F C 14
G C,D 11
H F,G 10
I E 5
J I 4
K H 1

Q-09 The time estimates of different activities of a project are given below. Draw the network
diagram, number the events and find the critical path. Find the total and free float for all
non-critical activities.
Task Precedence Duration in weeks
A Appraisal of book by reviewers -- 8
B Initial pricing of book -- 2
C Assessment of marketability A, B 2
D Revisions by author A 6
E Editing of final draft C, D 4
F Typesetting of text E 3
G Plates for artwork E 4
H Designing and printing of jacket C, D 6
I Printing and binding of book F, G 8
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

J Inspection and final assembly I, H 1

Q-10 For the following information about a network, draw the network diagram, calculate the
critical path and find total, free and interference float.

Name of the Description of the activity Preceding Duration

activity Activities in days
A Buy Sales office - 6
B Hire Salesmen A 4
C Train Salesmen B 7
D Select Ad Agency A 2
E Plan Ad campaign D 4
F Conduct Ad campaign E 10
G Design Packaging - 2
H Set up packaging facilities G 10
I Package Initial Stock J, H 6
J Order stock from manufacturers - 13
K Select Distributors A 9
L Sell to Distributors C, K 3
M Ship to Distributors I, L 5

Chapter 3: Network Analysis

B: Crashing in Critical Path Networks
The method:
Introduction:
We have seen the procedure for identifying critical and non-critical activities of a project. The
critical or longest path is the one which is most important. It’s this path which decides the
duration of the project. As seen, the non-critical activities are those which can be delayed, if
need be, and so long as the delay is within the overall float of these activities the project can
still be completed in time. There are some situations when you want to complete the project
ahead of it’s original schedule either for financial or other reasons and many times it is
possible to shorten the project duration and at the same time, reduce the cost of the project.
This statement may seem contradictory in the beginning, but after careful consideration, we
can actually achieve both the objectives. This is possible because the costs in the project are
usually of two types, one – the direct cost of an acitivity such as labour cost, material cost etc
and the second – in the nature of indirect cost. This indirect cost is usually in the form of
supersivory cost which cannot be attibuted to an individual activity, but is for the whole
project. Since supervisory cost is usually proportional to the duration of the project, lesser
the duration, lesser the supervision needed and lesser the cost therefor. We can sometimes
find the golden mean such that the total of direct and indirect cost is minimum. For
shortening the activity usually more efforts are needed, more labour is needed, even more
material or other inputs are needed. Since it does not make sense to shorten the duration of
non-critical activity,(since the longest or critical path will still remain equally long) such
shortening is not done for non-critical activities. Critical activities can be shortened by
deploying more resourses at an additional cost, but if such additional cost is less than the
possible savings in indirect cost, we will still save of the total cost. This is the gist of CPM
Crashing. The additional cost needed to be incurred for shortening an activity is called as the
crash cost and therefore the exercise of this time-cost trade-off is referred to as Crashing.
Carefully study the following problems on crashing, since the crashing question if asked in
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

the examination has the potential of carrying the marks of a full question of 15 marks can be
the lengthy question at serial number 5 in the examination.
Steps:
Calculate the normal duration and total cost of the project. (Total cost = Direct normal cost +
crashing cost + indirect cost). Initially before the crashing exercise, crash cost would be zero.
Also carry out the estimate of total minimum possible duration of the project, assuming as if
all activities have been crashed to the entire extent. This minimum possible duration and the
normal duration are the limits within which the optimal duration would lie. Optimal duration
is that when the total cost is minimum. Prepare a table showing the cost slope of each
activity, maximum possible crash in an activity and a crashing rank alloted to the activities in
ascending order of crash cost per day (that is cost slope). Start crashing from the critical
activity with lowest rank (lowest crash cost). At the end of each crash, calculate the costs
namely direct cost (Normal cost + crash cost incurred till then) + indirect cost. Sometimes the
indirect cost is given on a per day (or week / month as the case may be) basis or occasionally
it is given in a tabular form of days vs indirect cost. Extra care needs to be taken to ensure
that all other non-critical paths are kept track of since those non-critical paths tend to
become critical when the ‘reduced’ duration of critical path starts approaching the duration
of non-critical paths. It is quite possible that at some stage, some other non-critical path may
become critical when the duration of original critical path equals the duration of next longest
non-critical path. To bring down the duration of the project having two critical paths many
times needs crashing more than one activity. Watch out the video lectures in this regard,
after reading the above conceptual background note. Also see how to present the answer in
the examination.

Q-01 Find the optimal duration and cost for the following project
Activity 1-2 1-3 2-4 2-5 3-4 4-5
Activity A B C D E F
Normal
Time 8 4 2 10 5 3
Normal
Cost 100 150 50 100 100 80
Crash Time 6 2 1 5 1 1
Crash Cost 200 350 90 400 200 100
Cost
Slope?
Indirect cost is Rs. 70 per
day
Mumbai University April
2003

Q-02 The following table gives the activities in a construction project and the precedence
relationships between activities. The table also provides the details about cost and duration.
Draw the network diagram, and find the critical path with normal duration. If the indirect
cost is Rs 95 per day of project duration,
find the optimal cost and duration of the project by crashing the network by one day at a
time.
 Video lectures by Prof Madhusudan Sohani. Handout of questions and notes

Activity Preceding activity Time in days Direct Cost in

Rupees
Normal Crash Normal Crash
A - None - 4 3 60 90
B - None - 6 4 150 250
C - None - 2 1 38 60
D A 5 3 150 250
E C 2 2 100 100
F A 7 6 115 145
G B, D, E 4 3 100 170

Q-03 The time and cost estimates of different activities of a project are given below:

Activity Predecessor Time in weeks Cost in Rs.

activity

Normal Crash Normal Crash

A - 6 4 10000 14000
B - 4 3 5000 8000
C A 3 2 4000 5000
D B 8 3 1000 6000
E B 14 6 9000 13000
F C, D 8 4 7000 8000
Overhead costs amount to Rs 1000 per week. Crash the project to the extent of
16 weeks and estimate the cost. Will optimum duration be more than 16 weeks or less than
16 weeks? Why?
Q-04 Find the optimal duration of the following project for which the indirect cost is Rs 100 per
day.
Predeedi
ng Normal Crash Crash Normal Cost
Activity Activity time time cost Rs cost Rs slope
A - 6 5 200 80 120
B - 4 2 400 200 100
C - 5 3 300 140 80
D A 4 2 400 280 60
E C 3 1 500 350 75
F B 4 2 400 220 90
G D 3 1 280 200 40
H E 2 1 130 100 30