Transcript Operations Management

Operations management
Session 17: Introduction to Revenue
Management and Decision Trees
Previous Class
Session 17
Operations Management
2
Today’s Class
 Introduction to Revenue Management
 Decision-making under uncertainty

Decision Trees
 Simulation Game Explanation
Session 17
Operations Management
3
RM: A Basic Business Need
Session 17
Increasing Revenue
Reducing Cost
 What are the basic ways to improve profits?
Revenue
Management
$
Profits
Operations Management
4
Revenue Management
capacity
control
forecasting
market
segmentation
optimization
pricing
Session 17
overbooking
Operations Management
5
Revenue Management
Definitions
‘Selling the right seats to the right customers at the right prices
and the right time.’
(American Airlines 1987)
(Squeezing as many dollars as possible out of the customers)
‘Integrated control and management of price and capacity
(availability) in a way that maximizes company profitability.’
Session 17
Operations Management
6
Revenue Management History
 RM was ‘invented’ by major US carriers after airline deregulation
in the late 1970’s to compete with new low cost carriers
 Matching of low prices was not an alternative because of higher
cost structure
 American Airline’s ‘super saver fares’ (1975) have been first
capacity controlled discounted fares
 RM allowed the carriers to protect their high-yield sector while
simultaneously competing with new airlines in the low-yield sector
 From art to science: By now, there are sophisticated RM tools and
no airline can survive without some form of RM
 Other industries followed - hotel, car rental, cruise lines etc.
Session 17
Operations Management
7
Revenue Management
 How the optimization in Revenue Management
might differ from what we have already learned
(like linear programming)?
Session 17
Operations Management
8
Capacity Investment-1
 New-Fashion buys dyed yarns and makes fashionable
dress. The company knows with certainty that red will
be the color of the year and the demand for a red
gown is 2,000 units per month for the next 5 months.
 The company can invest in a new production line with
advanced technology. The capacity of the new line is
2,000 units per month.


Session 17
The cost of this line is $1,000,000.
The production cost per unit is $130.
Operations Management
9
Capacity Investment-1
 Alternative: The company can also convert an obsolete
line with traditional technology. The capacity of the
production line is also 2,000 units per month.


The cost of this conversion is $500,000.
The production cost per unit is $200.
 Each red gowns are sold for $300 each.
Session 17
Operations Management
10
Capacity Investment-1
 Which technology should the company chose?
-1,000,000+5*2,000*170=0.7M
New
Traditional
-500,000+5*2,000*100=0.5M
 Clearly the new technology is preferable.
Session 17
Operations Management
11
Capacity Investment-2
 New-Fashion company is concerned that orange
instead of red being the color of year.
 The CEO of the company prefers to assume that the
demand for the red gowns will be:


2,000 per month (probability 0.6)
0 (probability 0.4, market will demand 2000 orange gowns)
 Given this information…
Session 17
Operations Management
12
Capacity Investment-2
-1,000,000+5*2,000*170=0.7M
red
New
orange
-1,000,000+0=-1M
red
-500,000+5*2,000*100=0.5M
orange
-500,000+0=-0.5M
Traditional
Session 17
Operations Management
13
Capacity Investment-2
 The optimal decision is to invest in the traditional
technology.
 Intuitively, the traditional technology is preferred when
the demand is uncertain because it has a lower
upfront cost, but higher variable cost of production.
Lesson: Lower upfront costs are preferred when there is
more variability.
Session 17
Operations Management
14
Decision Tree
 A tool to come up strategy under uncertain
environments
Decision
Scenario
Session 17
Operations Management
15
Capacity Investment-3
 A smart consultant realized that a technology can
delay the dye process and enable the company dye
finished gowns after they know the color of the year.
 The technology introduces an additional $30 cost of
dyeing for each unit produced.
 What should the company do?
Session 17
Operations Management
16
Capacity Investment-3
w/o dye
delayed
with dye
delayed
Session 17
Operations Management
17
Observations
 We observe that delay dyeing to collect more
information is beneficial.
Session 17
Operations Management
18
More Observations
 We also observe that if the company delays dyeing it is
optimal to invest in the new technology. While if it decides
to not wait, it is optimal for the company to invest in the
traditional technology. Why?
The new technology costs more, but has lower production costs.
Therefore, once we know demand is high, we prefer to make a
higher initial upfront investment but have a lower marginal
production cost.
 Postpone differentiation and flexibility is desirable
 Sometime, waiting and collecting information is worthwhile
Session 17
Operations Management
19
What did we learn?
 How to use a decision tree to evaluate
alternatives.
 Let’s see another example in a different context.
Session 17
Operations Management
20
Decision Trees
 A new drug must pass through three stages of
clinical trials before it can be brought to market.




Phase 1: Safety is evaluated on a small group.
Phase 2: The effectiveness of the drug is evaluated on a
large group.
Phase 3: Randomized controlled trials are performed on even
larger groups. Comparison is against a “gold standard”
treatment.
(Phase 4: Post-launch safety surveillance.)
When should we contract for production capacity?
Session 17
Operations Management
21
Decision Trees
 Suppose we desire to introduce a new hypertension drug to
market.
 We have completed phase 1 and 2 trials successfully.
 We assess a 90% probability of completing phase 3
successfully (and therefore gaining FDA approval).
 We assume demand for the drug will be 5 million people in
the next year.
 A one-year drug supply for a single person should net us a
$50 profit.
Session 17
Operations Management
22
Decision Trees
 We have the option of contracting for
manufacturing capacity now for $150 million.
 We expect the cost of manufacturing capacity to
increase if we wait until we know the results of
our Phase 3 trial.
 What is the minimum expected cost of delaying
manufacturing such that it is beneficial for us to
wait to contract for manufacturing capacity?
Session 17
Operations Management
23
Decision Trees
contract now
approved $50×5-$150
=$100 million
0.9
not approved
-$150 million
0.1
contract later
0.9× (50×5-P) million
0.9×100-0.1×150=75>0.9× (50×5-P),
83.33 > 250-P
or P>166.67 in order that contracting now is more profitable.
Session 17
Operations Management
24
Decision Trees
We valued the flexibility of being able to
wait until there is no more uncertainty.
Session 17
Operations Management
25
Decision Trees
 Now suppose we have only completed Phase 1,
and that we assess the probability of completing
phase 2 to be 50%.
 We still assess the probability of completing Phase
3 to be 90%.
 We again have the option to contract now at $150
million or to contract later (after either completing
phase 2 or 3).
Session 17
Operations Management
26
Decision Trees
now
pass phase 2
0.5 $75
million
-37.5
do not pass
million
0.5
pass phase 3
0.9
do not pass
0.1
$100 million
-$150 million
-$150 million
later
It does not make sense to contract now.
Session 17
Operations Management
27
What did we learn?
 Decision trees
 How to value the option of delaying decisions to
collect information
 Next class, we will study revenue management
tools based on decision trees
 Still upcoming … simulation game explanation.
Session 17
Operations Management
28
Next Session
 Homework 4 due.
 Game report 1 due.
Session 17
Operations Management
29