Supply Chain Management

Lecture 25
Semester Outline
• Tuesday April 20 Chap 15
• Thursday April 22 Simulation Game briefing
• Tuesday April 27 Review, buffer
• Thursday April 29 Simulation Game
Outline
• Today
– Chapter 15
• Sections 1, 2
• Homework 7
– Online today
– Due Thursday April 29 before class
• Homework submitted before April 29 will be graded and
returned on April 29
• Thursday
– Simulation game briefing
What is Revenue Management?
• Revenue management is the practice of differential
pricing to increase supply chain profits
– A strategy that adjusts prices based on product availability,
customer demand, and remaining duration of the sales season
will result in higher supply chain profits
What is Revenue Management?
• Revenue management is the practice of differential
pricing to increase supply chain profits
– A strategy that adjusts prices based on product availability,
customer demand, and remaining duration of the sales season
will result in higher supply chain profits
• Revenue management, also called yield management,
and sometimes smart pricing, is a technique to optimize
revenue from a fixed, but perishable inventory
Revenue Management

Revenue Management:
Maps capacity into demand
Newsvendor problem:
Maps demand into capacity
What is Revenue Management?
• Revenue management, also called yield management,
and sometimes smart pricing, is a technique to optimize
revenue from a fixed, but perishable inventory
• Is revenue management possible for…
– Airline tickets
– Cruise travel
– Restaurants
– Hospitals
– LTL trucking companies
– Apartment rental
– Incoming MBA class
– Vending machines
Revenue Management and
Vending Machines
• Coca-Cola announces that it is
considering vending machines that will
boost prices during hot weather.
– “Coca-Cola is a product whose utility varies
from moment to moment. In a final summer
championship, when people meet in a
stadium to enjoy themselves, the utility of a
chilled Coca-Cola is very high. So it is fair it
should be more expensive. The machine will
simply make this process automatic.”
Douglas Ivester, Chairman and CEO
Conditions for Revenue
Management
• The value of the product varies in different market
segments
– Airline seats: leisure versus business travel
• The product is highly perishable or product waste
occurs
– Fashion and seasonal apparel
– High tech products
• Demand has seasonal and other peaks
– Cruise travel
• The product is sold both in bulk and on the spot market
– Owner of warehouse who can decide whether to lease the entire
warehouse through long-term contracts or save a portion of the
warehouse for use in the spot market
Why Revenue Management?
• Success stories
– American Airlines increased annual revenue by over $1 billion
through revenue management
– Marriott hotels increased annual revenue with $100 million
through revenue management
– National Car Rental was saved from liquidation through revenue
management
– Canadian Broadcasting Corporation increased revenue with $1
million per week
Airfare example
q
Choose the fare that
1000
maximizes the area
800 (revenue) of the rectangle

600

400

200

200 400 600 800 1000 p

Airfare example
q
Choose the fare that
1000 Unaccommodated
demand maximizes the area
800 (revenue) of the rectangle

600 Maximum revenue

= 500*500
400 = $250,000

200 Consumer
surplus

200 400 600 800 1000 p

Airfare example
q
Choose the fare that
1000
maximizes the SUM of areas
800 of the rectangles
Economy class
Maximum revenue
600 = 333*(333 + 667) =
$333,000
400 Business class

200

200 400 600 800 1000 p

Airfare example
q
Choose the fare that
1000
maximizes the SUM of areas
Economy class of the rectangles
800
Economy plus class Maximum revenue
600 = 200*(800+600+400+200)
Business class = $400,000
400
First class
200

200 400 600 800 1000 p

Airfare example
q
Perfect price discrimination
1000 Charging a different price to a different buyer
for the same product without any true cost
800 differential to justify the different price

600 Maximum revenue

= $500,000
400

200

200 400 600 800 1000 p

Is Revenue Management Price
Discrimination?
• The same product sold at different times for different
prices is not necessarily price discrimination, because
at different times...
– The production or distribution costs may be different
– Inventory costs were incurred to keep the product in stock until a
later time
– Consumers value products differently at different points in time
– The product value may change over time, such as perishable or
maturing or seasonal products, fashion goods, antiques.
– Interest is earned if product is sold at an earlier time
– Locking sales in early reduces uncertainty
Revenue Management for Multiple
Customer Segments
• If a supplier serves multiple customer segments
with a fixed asset, the supplier can improve
revenues by setting different prices for each
segment
– What price to charge each segment?
– How to allocate limited capacity among the segments?

Prices must be set with barriers such

that the segment willing to pay more is
not able to pay the lower price
Revenue Management
• Hotels, airlines, opera houses hope this tool will
help them maximize sales and profits
– “The real beneficiary of revenue management has
been the consumer”

Clearly, customers for which revenue

management has decreased the cost of air travel,
have benefited from revenue management. Could
customers for which revenue management has
increased the cost of air travel, also have
benefited from revenue management?
What is Revenue Management?

q q
1000 1000

800 800

600 600

400 400

200 200

200 400 600 800 1000 p 200 400 600 800 1000 p
Example 15-1: Pricing to multiple
segments
• A contract manufacturer has identified two customers
segments for its production capacity—one willing to
place an order more than one week in advance and the
other willing to pay a higher price as long as it can
provide less than a week’s notice for production. The
customers that are unwilling to commit in advance are
less price sensitive and have a demand curve d1 = 5,000
– 20p1. Customers willing to commit in advance are
more price sensitive and have a demand curve of d2 =
5,000 – 40p2. Production cost is c = $10 per unit. What
price should the contract manufacturer charge each
segment if its goal is to maximize profits?
Example 15-1: Pricing to multiple
segments
c = 10
6000

5000

4000
d1 = 5,000 – 20p1
Demand

3000

2000

1000

0
0 50 100 150 200 250 300
Price
Example 15-1: Pricing to multiple
segments
c = 10
6000

5000

4000
Demand

3000

2000

1000 Profit d1 = 5,000 – 20p1

0
0 50 100 150 200 250 300
Price
p-c
Pricing Multiple Segments
• Assume that the demand curve for segment i is
given by
– di = Ai – Bipi
• The goal of the supplier is to price so as to
maximize profits 6000

– Max (pi – c)(Ai – Bipi) 5000

4000
Demand
3000

2000

1000 Profit
0
0 50 100 150 200 250 300
Price
Pricing Multiple Segments
• The optimal price for segment i is given by
– pi = Ai/2Bi + c/2
Example 15-1: Pricing to multiple
segments
• For segment 1:
– pi = Ai/2Bi + c/2 pi = 5,000/(2*20) + 10/2
= $130
– Profit (pi – 10)(5,000 – 20pi)
= (130 – 10)(5,000 – 20*130)
= $288,000
• For segment 2:
– pi = Ai/2Bi + c/2 pi = 5,000/(2*40) + 10/2
= $67.50
– Profit (pi – 10)(5,000 – 40pi)
= (67.5 – 10)(5,000 – 40*67.5)
= $127,650

Total profit $415,650

Example 15-1: Pricing to multiple
segments
• If total capacity is limited to 4,000 units, what
should the contract manufacturer charge each
segment?
– For segment 1: p1 = $130
• Demand d1 = (5,000 – 20p1) = 2,400
– For segment 2: p2 = $67.50
• Demand d2 = (5,000 – 40p2) = 2,300
– Total demand = 2,400 + 2,300 = 4,700

Total demand exceeds production capacity of 4,000

Pricing Multiple Segments
• The goal of the supplier is to price so as to
maximize profits
– Max ∑ki=1 (pi – c)(Ai – Bipi) Maximize profits
– Subject to:
∑ki=1(Ai – Bipi)  Q Production capacity
pi  0 Price
c 10
Q 4000
Segment Price Demand Profit
1 141.666666651042 =5000-20*B4 =(B4-B1)*(5000-20*B4)
2 79.1666666744792 =5000-40*B5 =(B5-B1)*(5000-40*B5)
Total =C4+C5 =D4+D5
Example 15-1: Pricing to multiple
segments
• If the contract manufacturer were to charge a
single price over both segments, what should it
be?
6000 6000

5000 5000

4000 4000
Demand

Demand
3000 3000

2000 2000

1000 1000

0 0
0 50 100 150 200 250 300 0 50 100 150 200 250 300
Price Price

d1 = 5,000 – 20p1 d2 = 5,000 – 40p2

d = (5,000 – 20p) + (5,000 – 40p) = 10,000 – 60p
Example 15-1: Pricing to multiple
segments
• For segment 1 and 2:
– p = Ai/2Bi + c/2 p = 10,000/(2*60) + 10/2
= $83.33
– Max (p – c)(A – Bp) Max (p – 10)(10,000 – 60p)
= (83.33 – 10)(10,000 – 60*83.33)
= $366,650

Differential pricing raises profit

from $366,650 to $415,650
Revenue Management for Multiple
Customer Segments
• If a supplier serves multiple customer segments
with a fixed asset, the supplier can improve
revenues by setting different prices for each
segment
– What price to charge each segment?
– How to allocate limited capacity among the segments?

What if demand is uncertain?

The Park Hyatt Philadelphia
• 118 King/Queen rooms.
• Hyatt offers a pL= $128 (low fare) targeting
leisure travelers.
• Regular fare is pH= $181 (high fare)
targeting business travelers.
• Demand for low fare rooms is abundant.
• Let DH be uncertain demand for high fare
rooms.
• Assume demand for the high fare (business)
occurs only within a few days of the actual
stay

How much capacity should Hyatt

save for the higher priced segment?
Allocating Capacity to a Segment
Under Uncertainty
• Basic tradeoff between committing to an order
from a lower-price buyer or waiting for a high-
price buyer to arrive later on
– Spoilage occurs when the capacity reserved for
higher-price buyers is wasted because demand from
the higher-price segment does not materialize
– Spill occurs if higher-price buyers have to be turned
away because the capacity has already been
committed to lower-price buyers
Allocating Capacity to a Segment
Under Uncertainty

Never sell a unit of capacity for

less than the expected revenue

• Expected revenue = sales probability x sales price

$128  $181.00 = 1.0 x 181

$128  $162.90 = 0.9 x 181
$128  $144.80 = 0.8 x 181
$128  $126.70 = 0.7 x 181
Allocating Capacity to a Segment
Under Uncertainty
$126.70 = 0.7 x 181
Expected revenue = sales probability x sales price

RH(CH) = Prob(demand from higher-price segment > CH) x pH

Never sell a unit of capacity for

less than the expected revenue
Allocating Capacity to a Segment
Under Uncertainty
$128  $126.70 = 0.7 x 181
Expected revenue = sales probability x sales price

RH(CH) = Prob(demand from higher-price segment > CH) x pH

Never sell a unit of capacity for

less than the expected revenue
pL = Prob(demand from higher-price segment > CH) x pH

Prob(demand from higher-price segment > CH) = pL/pH

Allocating Capacity to a Segment
Under Uncertainty
Prob(demand from higher-price segment > CH) = pL/pH

Prob(demand from higher-price segment  CH) = 1 – pL/pH

CH = F-1(1 – pL/pH, DH, H)

Prob

CH
1 – pL/pH pL/pH
Example: Allocating Capacity to a
Segment Under Uncertainty
• Assume that demand for rooms at the high rate is
normally distributed with mean 102 and standard
deviation 20.8. Also assume that the high rate is 181
dollars and low rate (discount rate) is 128 dollars
– Determine probability that expected marginal revenue of higher
rate class will exceed marginal revenue of lower rate class
• pL = 128
• pH = 181
• 1 – pL/pH = 1 – 128/181 = 0.2928
– Convert that probability into the number of rooms
• NORMINV(1 – pL/pH, DH, H) = NORMINV(0.2928, 102, 20.8) = 91

Hence, 91 rooms should be

reserved for the high rate class
Example 15-2 Allocating Capacity
to Multiple Segments
• ToFrom Trucking serves two customer segments. One
segment (A) is willing to pay $3.50 per cubic feet but
wants to commit with only 24 hours notice. The other
segment (B) is willing to pay only $2.00, but is willing to
commit to a shipment with up to one week notice. With
two weeks to go, demand for segment A is forecast to be
normally distributed, with a mean of 3,000 cubic feet and
a standard deviation of 1,000. How much of the available
capacity should be reserved for segment A?
Example 15-2 Allocating Capacity
to Multiple Segments
Revenue from segment A pA = $3.50

Revenue from segment B pB = $2.00

Mean demand for DA = 3,000

segment A
Standard deviation of A = 1,000
demand for segment A
Capacity to be reserved CA = F-1(1 – pB/pA, DH, H) =
for segment A F-1(0.4286,3000,1000) =
2,820
Example 15-2 Allocating Capacity
to Multiple Segments
• ToFrom Trucking serves two customer segments. One
segment (A) is willing to pay $3.50 per cubic feet but
wants to commit with only 24 hours notice. The other
segment (B) is willing to pay only $2.00, but is willing to
commit to a shipment with up to one week notice. With
two weeks to go, demand for segment A is forecast to be
normally distributed, with a mean of 3,000 cubic feet and
a standard deviation of 1,000. How much of the available
capacity should be reserved for segment A?

How should ToFrom change it decision if

segment A is willing to pay $5 per cubic foot?
Example 15-2 Allocating Capacity
to Multiple Segments
Revenue from segment A pA = $5.00

Revenue from segment B pB = $2.00

Mean demand for DA = 3,000

segment A
Standard deviation of A = 1,000
demand for segment A
Capacity to be reserved CA = F-1(1 – pB/pA, DH, H) =
for segment A F-1(0.6, 3000, 1000) =
3,253