Lecture-2

Operations Management
22/01/2022
BITS Pilani
Pilani|Dubai|Goa|Hyderabad
Gaurav Nagpal

1
Competitive dimensions of
Operations

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Theory of Competitive Advantage

The motivation for international outsourcing comes from the

theory of comparative advantage. This theory focuses on
the economic concept of relative advantage. According to
the theory, if an external provider, regardless of its
geographic location, can perform activities more
productively than the purchasing firm, then the external
provider should do the work. This allows the purchasing
firm to focus on what it does best—its core competencies.
Consistent with the theory of comparative advantage,
outsourcing continues to grow. But outsourcing the wrong
activities can be a disaster. And even outsourcing non-
core activities has risks.
3

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Advantages and disadvantages
of outsourcing

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

 5

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Solution

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

International Operations
Strategies

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Efficiency- Responsiveness conflict

The more responsive you become to your customer needs, the

more cost you are going to incur, and you will become less
efficient, output per unit of input will fall

If you have to deliver fast… I need to have more of warehouses

(more de-centralised distribution), more information systems,
more inventories in the supply chain, more lastmile
manpower.
If you customise more,…..
If you offer more flexibility ….
8

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Differentiation vs Low cost
strategy

Rock-water analogy… Inventory is like water

9

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Question

10

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Solution Approach 1

Let Rx be the units of resource

required to produce a unit of X

Let Ry be the units of resource

required to produce a unit of y

For Canada
4.Rx = 8.Ry
Rx/Ry =2

For Mexico
4.Rx = 6.Ry
Rx/Ry = 1.5

Therefore, Mexico has relative

advantage in producing X.

Canada has relative advantage in

producing Y
11

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Solution Approach 2

Canada gives up 2 units of

good Y for every unit of good X
Opportunity cost of 1 unit of
good X is 2 units of good Y
Opportunity cost of 1 unit of
good Y is 0.5 units of good X

Mexico gives up 1.5 units of

good Y for every unit of good X
Opportunity cost of 1 unit of
good X is 1.5 units of good Y
Opportunity cost of 1 unit of
good Y is 0.67 units of good X

12

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Which country should
produce what?

Solution:

In Portugal,
manhours requirement per unit cloth = 90/80 times manhours requirement per unit wine

In England,
manhours requirement per unit cloth = 100/120 times manhours requirement per unit wine

Therefore, England has comparative advantage in producing cloth and Portugal has
comparative advantage in producing wine

13

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Important to focus on core
competency
 If retail is core competency, why manage real estate?
 Business Process Outsourcing

14

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

 Byju's acquires Aakash Educational
Services for $1 bn via strategic merger
Edtech decacorn Byju’s has closed the deal to acquire exam preparation firm Aakash Educational Services
Limited (AESL) for nearly $1 billion. The stock-and-cash deal is the biggest in the education space, said
sources.

Byju’s said the deal is a strategic partnership with AESL, a leader in test-prep services. “I am happy to have
Aakash Educational Services Limited (AESL), a market leader and the most trusted name in the test-prep
services, on board with us,” said Byju Raveendran, Founder, and CEO, Byju’s. “Our complementary strengths
will enable us to build capabilities, create engaging and personalised learning programmes. The future of
learning is hybrid and this union will bring together the best of offline and online learning, as we combine
our expertise to create impactful experiences for students.” Raveendran said the pandemic has brought the
importance of the blended format of learning to the forefront. “As we unite our forces to bring together
decades of expertise and experience, this partnership will further accelerate Aakash’s growth and success.”

“At Aakash, we are looking to transform student experiences by steering innovative and digitally-enabled learning
solutions. Together with Byju’s, we will work towards building an omnichannel learning offering that will accelerate
test-prep experience to the next level,” said Aakash Chaudhry, managing director, AESL.
15

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Important to have a core
competency
 Establish cost leadership or differentiation in atleast one product for which you would
like to be known
 Then, gradually diversify into other products to scale up your business

 For example:
 Om Sweets started with Dhodha
 Jain Shikanji started with shikanji
 Also important to keep surprising your customers with new product innovations so as
to make them feel that the brand is livid

16

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Product innovation with time:
McDonalds

17

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Product innovation with time:
Cadbury’s

18

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

About optimisation packages

Pulp: PuLP is an open-source linear programming (LP) package which largely uses Python syntax and comes

packaged with many industry-standard solvers. It also integrates nicely with a range of open source and commercial

LP solvers. You can install it using pip (and also some additional solvers)

Gurobi: The fastest and most powerful mathematical programming solver available for your LP, QP and MIP (MILP,

MIQP, and MIQCP) problems.

GLPK: The GNU Linear Programming Kit (GLPK) is a software package intended for solving large-scale linear

programming (LP), mixed integer programming (MIP), and other related problems.

Pyomo is a Python-based, open-source optimization modeling language with a diverse set of optimization capabilities.

The beauty of using Pyomo is that you can try different solvers by just changing the name of the solver. The

standard utility for installing Python packages is pip. You can install Pyomo in your system Python installation by

executing the following in a shell: pip install pyomo. You can install Pyomo in your system Python installation by
19
executing the following in a shell: conda install -c conda-forge pyomo
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Purchasing Problem- allocating
your share of telephone calls
You, as a purchase manager, have been approached by three telephone companies to subscribe to their long-distance

service in the United States. MaBell will charge a flat $16 per month plus $.25 a minute. PaBell will charge $25 a month

but will reduce the per-minute cost to $.21. As for BabyBell, the flat monthly charge is $18, and the cost per min is $.22.

You usually make an average of 200 minutes of long-distance calls a month. Assuming that you do not pay the flat

monthly fee unless you make calls and that you can apportion your calls among all three companies as you please, how

should you use the three companies to minimize the monthly telephone bill? Max cap for any operator is 120 minutes.

Solution: Let x1, x2, x3 be call minutes for 3 operators

Let y1, y2, y3 be the binary variables which demote whether the ith operator is chosen or not,
y1=1 if x1>0, else y1=0
y2=1 if x2>0, else y2=0
y3=1 if x3>0, else y3=0

Obj fun min of total cost = 0.25x1 + .21x2 + .22x3 + 16y1 + 25y2 + 18y3
x1 + x2 + x3 = 200
x1 <= 1000000.y1 If y1=0, x1 has to be zero; and if y1=1, x1 can be any number
x2 <= 1000000.y2 If y2=0, x2 has to be zero; and if y2=1, x2 can be any number
x3 <= 1000000.y3 If y3=0, x3 has to be zero; and if y3=1, x3 can be any number
x1, x2, x3 <= 120
x1, x2, x3 >=0 20

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Formulation

Minimize z = .25x1 + .21x2 + .22x3 + 16y1 + 25y2 + 18y3

subject to
x1 + x2 + x3 = 200
x1 <= 120
x2 <= 120
X3 <= 120
x1 <= 1000000y1
x2 <= 1000000y2
X3 <= 1000000y3
x1, x2, x3 >= 0
y1, y2, y3 = (0,1)
21

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Code using glpk (1)
# Fixed Charge Problem - Choosing a Telephone Company

import pyomo.environ as pyo

from pyomo.environ import *
from pyomo.opt import SolverFactory

#Model
model = pyo.ConcreteModel()

#defining the variables to represent duration of service usage of telephone company

model.x = pyo.Var(range(3), domain = pyo.NonNegativeReals)
x=model.x

#defining binary auxiliary variable

model.y = pyo.Var(range(3), within=Binary)
y=model.y

#Constraints
model.avg_call = pyo.Constraint(expr = sum(x[i] for i in range(3))==200)
model.x0limit = pyo.Constraint(expr = x[0]<=120)
model.x1limit = pyo.Constraint(expr = x[1]<=120)
model.x2limit = pyo.Constraint(expr = x[2]<=120)

#constraints to establish relationship between x and y variables using big M

#you may also creater 3 separate constraints
model.bigM = pyo.ConstraintList()
for i in range(3): 22
model.bigM.add(expr = x[i]<=1000000*y[i])

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956

Code using glpk (2)
#Objective function
model.obj = pyo.Objective(expr = 0.25*x[0]+0.21*x[1]+0.22*x[2]+16*y[0]+25*y[1]+18*y[2],
sense=minimize)

solvername='glpk'

solverpath_folder='C:\\glpk\\w64' #does not need to be directly on c drive

solverpath_exe='C:\\glpk\\w64\\glpsol' #does not need to be directly on c drive

opt = SolverFactory('glpk')
opt.solve(model)

model.pprint()

print('Minimum monthly telephone bill =', pyo.value(model.obj))

for i in range(3): 23
print('x[%i] = %i' % (i, pyo.value(x[i])), 'y[%i] = %i' % (i, pyo.value(y[i])))
BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956
Code using Pulp
import pulp as p

# Create shortest route problem

Lp_prob = p.LpProblem('Problem', p.LpMinimize)

# Create problem Variables

x1 = p.LpVariable("x1", lowBound=0) # Create a positive variable x1
x2 = p.LpVariable("x2", lowBound=0) # Create a positive variable x2
x3 = p.LpVariable("x3", lowBound=0) # Create a positive variable x3
y1 = p.LpVariable("y1", cat='Binary') # Create a binary variable y1
y2 = p.LpVariable("y2", cat='Binary') # Create a binary variable y2
y3 = p.LpVariable("y3", cat='Binary') # Create a binary variable y3

# Objective Function
Lp_prob += 0.25*x1+0.21*x2+0.22*x3+16*y1+25*y2+18*y3

# Constraints:
Lp_prob += x1 <= 120
Lp_prob += x2 <= 120
Lp_prob += x3 <= 120
Lp_prob += x1 <= 1000000*y1
Lp_prob += x2 <= 1000000*y2
Lp_prob += x3 <= 1000000*y3
Lp_prob += x1 + x2 + x3 == 200

# Display the problem

print(Lp_prob)

status = Lp_prob.solve() # Solver

print(p.LpStatus[status]) # The solution status

# Printing the final solution 24

print(p.value(x1), p.value(x2),p.value(x3), p.value(y1),p.value(y2), p.value(y3), p.value(Lp_prob.objective))

BITS Pilani, Deemed to be University under Section 3 of UGC Act, 1956