DMD 1

1
Welcome to DMD Recitation 1!

¡ Due electronically on Monday February 2, 2015
¡ Airline on Ground (AOG) case – work with your team!
¡ Exercises 2.13 and 2.30 – complete individually!
¡ Submit the PDF files on Stellar.
¡ Google Doc for Teams: bit.ly/DMD16-Teams
¡ To reduce background noise, please mute your phone/

computer!
¡ Please feel free to raise your hand or chat through WebEx if

you have any questions or comments!
15.730 - Decision Trees and Discrete Probability

2
Data, Models,
Data Models
Data, and Decisions
Models, and Decisions
¡ Decision making is hard due to uncertainty
¡ Use data and build models

Making to decision
good make moreis hard
informed decisions
Data Uncertainty Models

3
Outline
¡ Decision Analysis
¡ Tree construction
¡ EMV calculations
¡ Sensitivity Analysis
¡ Binomial Distribution
¡ Understand the formula
¡ Compute probabilities

4
Decision Analysis
¡ Decision Tree
¡ Logical and systematic way of organizing and
representing various decisions and uncertainties
¡ Two types of nodes

¡ Decision node: point where we can choose
between alternatives
¡ Event node: point with uncertainty that we cannot
control

5
Examples of Nodes
We have
control!

6
Examples of Nodes
Outcome is
Random!

7
Event Node Branches

¡ Mutually Exclusive:
No two outcomes can
happen simultaneously
¡ Collectively Exhaustive:
Set of all possible outcomes
represents the entire range of
possible outcomes
Sum of probabilities equals ONE

8
Should We Play the Lottery?

+
- $0.9 $1,000,000
- $1
$0
9

Compare EMVs of each +
decision and pick the $1,000,000
highest! - $0.9
- $1
$0
10

Compare EMVs of each +
decision and pick the $1,000,000
highest! - $0.9
$0
- $1
$0
11
Decision Analysis Procedure

¡ List choices (decision nodes)
¡ List uncertain events (event nodes)
¡ Construct a decision tree
¡ Determine the probabilities of each outcome
¡ Determine the numerical values of the endpoints
¡ Solve using backward induction

¡ Event nodes: calculate EMV
¡ Decision nodes: choose decision with highest EMV
¡ Perform sensitivity analysis (What-if scenarios)

12
Kendall Crab and Lobster

Case
¡ Shortly before noon, Jeff Daniels, director of
Overnight Delivery Operations at Kendall Crab
and Lobster (KCL) watched the weather
channel:
¡ Weather forecast predicted 50% chance that the
storm hits Boston around 5pm
¡ With the chance of Logan airport closing,

business travelers were also nervously awaiting
further weather information
¡ In the past, if a storm of this magnitude hits, 1 in 5
come with strong winds that force Logan to close

13
Operations
¡ Customers can order lobsters for next-day
delivery prior to 5pm on day before delivery
¡ Typical daily order of 3,000 lobsters
¡ At 5:30pm, trucks from United Express pick up the
lobsters and truck them to Logan airport
¡ At 6:30pm, packed lobsters are flown to a processing
and distributing facility in DC
¡ By 10:30am of next day, lobsters are delivered

14
Earnings and Refund Policy

¡ Price charged to customers is $30/lobster, which
includes all transportation costs
¡ When KCL ships a lobster via United Express, its

unit contribution to earnings is $10/lobster
¡ If KCL cannot deliver the lobsters to customers, its

policy is to give each customer a $20 discount
coupon per lobster
¡ Market research has shown that ~70% of the
customers only redeem the coupons

15
Changes to Operations Due to

Weather
¡ Rely on the Massachusetts Air Freight (MAF)
which operates 50 miles away from Boston
¡ If contacted before 5:30pm, MAF will pick up the
lobsters from KCL and deliver them to an airport in
Worcester to fly them to DC
¡ Additional transportation cost of using MAF is $13/
lobster in roughly 67% of the time, $19/lobster in the
remaining 33%

16
Changes to Operations Due to

Weather
¡ Cancel orders and issue coupons
¡ If the lobsters are not packaged yet, the incremental
cost of cancelling the orders is ~$1/lobster
¡ If lobsters were already packages, incremental cost
is ~$1.25/lobster
¡ Deliver lobsters by truck to DC via the Eastern

Parcel Delivery (EPD)
¡ Arrangement needs to be made by noon!
¡ Cost is $4/lobster 50% of the time, $3/lobster 25% of
the time and $2/lobster 25% of the time.

17
What is the Best Decision?

¡ List Decisions that can be made over time
Noon 5:00pm 5:30pm

18

Noon 5:00pm 5:30pm
1. Use EPD to
deliver by land
2. Wait until
5:00pm
3. Cancel orders
and issue
coupons

19

Noon 5:00pm 5:30pm
1. Use EPD to 1. USE MAF to

deliver by land deliver to
2. Wait until Worcester
5:00pm 2. Cancel orders
3. Cancel orders and issue
and issue coupons
coupons

20

¡ List Uncertainties and their probability of occurrence
Noon 5:00pm 5:30pm
If EPD, additional
cost is uncertain:
- $4 w.p. 0.5
- $3 w.p. 0.25
- $2 w.p. 0.25

21

Noon 5:00pm 5:30pm
If EPD, additional Storm:

cost is uncertain: - Yes w.p. 0.5
- $4 w.p. 0.5 - No w.p. 0.5
- $3 w.p. 0.25
- $2 w.p. 0.25

22

Noon 5:00pm 5:30pm
If EPD, additional Storm:

cost is uncertain: - Yes w.p. 0.5
- $4 w.p. 0.5 - No w.p. 0.5
- $3 w.p. 0.25
- $2 w.p. 0.25 Logan Airport:
- Closed w.p. 0.2
- Open w.p. 0.8

23

Noon 5:00pm 5:30pm
If EPD, additional Storm: If MAF, cost is uncertain:

cost is uncertain: - Yes w.p. 0.5 - $13 w.p. 0.67
- $4 w.p. 0.5 - No w.p. 0.5 - $19 w.p. 0.33
- $3 w.p. 0.25
- Closed w.p. 0.2
- Open w.p. 0.8

24

Noon 5:00pm 5:30pm
If EPD, additional Storm: If MAF, cost is uncertain:

cost is uncertain: - Yes w.p. 0.5 - $13 w.p. 0.67
- $4 w.p. 0.5 - No w.p. 0.5 - $19 w.p. 0.33
- $3 w.p. 0.25
- Closed w.p. 0.2 If order canceled:
- Open w.p. 0.8 - Redeem w.p. 0.7
- Not w.p. 0.3

25
Decision Tree
$3
.25
Wait
Noon 5:00pm 5:30pm

26
Decision Tree
$3
.25
Wait
No s
0.5
torm
Noon 5:00pm 5:30pm

27
Decision Tree
$3
.25
$19
Wait
No s
0.5
torm
Noon 5:00pm 5:30pm

28
Calculating the Branch Values $10
profit/
lobster
(10-4)*3,000 = 18,000
$3
.25
(10-3)*3,000 = 21,000
(10-2)*3,000 = 24,000
Wait
Noon 5:00pm 5:30pm

29
profit/
lobster
18,000
$3
.25 21,000
24,000
Wait
Noon 5:00pm 5:30pm

30
profit/
lobster
18,000
$3
.25 21,000
24,000
Wait
(-1)*3,000 + (-20)*(0.7)*3,000 = -45,000
Noon 5:00pm 5:30pm

31
profit/
lobster
18,000
$3
.25 21,000
24,000
Wait
-45,000
Noon 5:00pm 5:30pm

32
profit/
lobster
18,000
$3
.25 21,000
24,000
Wait
No s
10*3,000 = 30,000
0.5
torm
-45,000 10*3,000 = 30,000
Noon 5:00pm 5:30pm

33
profit/
lobster
18,000
$3
.25 21,000
24,000
Wait
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
34
profit/
lobster
18,000
$3
.25 21,000
24,000
Wait
(-1.25)*3,000 +
No s
30,000 (-20)*(0.7)*3,000 =
0.5
-45,750
torm
-45,000
30,000
Noon 5:00pm 5:30pm
35
profit/
lobster
18,000
$3
.25 21,000
24,000
Wait
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
36
profit/
lobster
18,000
(10-13)*3,000 = -9,000
$3
.25 21,000
$19
24,000
(10-19)*3,000 = -27,000
Wait
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
37
profit/
lobster
18,000
-9,000
$3
.25 21,000
$19
24,000
Wait -27,000
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
38
Calculating the EMV
18,000
-9,000
$3
.25 21,000
$19
24,000
Wait -27,000
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
39
Calculating the EMV
18,000 0.67*(-9,000)+0.33*
(-27,000) = -15,000 -9,000
$3
.25 21,000
$19
24,000
Wait -27,000
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
40
Calculating the EMV
18,000
-9,000
$3 -15,000
.25 21,000
$19
24,000
Wait -27,000
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
41
Comparing the EMV
18,000
-9,000
$3 -15,000
.25 21,000
$19
24,000
Wait -27,000
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
42
Comparing the EMV
18,000
-9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
43
Calculating the EMV
18,000
-9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
44
Calculating the EMV
18,000
-9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
-45,750
No s
30,000
0.5
0.2*(-15,000)+0.8*
torm
-45,000 (30,000) = 21,000

30,000
Noon 5:00pm 5:30pm
45
Calculating the EMV
18,000
-9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
21,000 -45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
46
Calculating the EMV
18,000
-9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
21,000 -45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
47
Calculating the EMV
18,000
-9,000
$3 -15,000
.25 21,000
0.5*(21,000) + -15,000
$19
0.5*(30,000) = 25,500
24,000
Wait -27,000
21,000 -45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
48
Calculating the EMV
18,000
-9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
25,000
21,000 -45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
49
Calculating the EMV
18,000
-9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
25,000
21,000 -45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
50
Calculating the EMV
0.5*(18,000) + 0.25*(21,000)
+ 0.25*24,000 = 20,250
18,000
-9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
25,000
21,000 -45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
51
Calculating the EMV
18,000
20,250 -9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
25,000
21,000 -45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
52
Comparing the EMV
18,000
20,250 -9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
25,000
21,000 -45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
53
Comparing the EMV
18,000
20,250 -9,000
$3 -15,000
.25 21,000
-15,000
$19
25,000 24,000
Wait -27,000
25,000
21,000 -45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
54
Sensitivity Analysis
18,000
20,250 -9,000
$3 -15,000
.25 21,000
-15,000
$19
? 24,000
Wait -27,000
?
21,000 -45,750
No s
30,000
0
torm
-45,000
30,000
Noon 5:00pm 5:30pm
55
18,000
20,250 -9,000
$3 -15,000
.25 21,000
-15,000
$19
21,000 24,000
Wait -27,000
21,000
21,000 -45,750
No s
30,000
0
torm
-45,000
30,000
Noon 5:00pm 5:30pm
56
Sensitivity Analysis To wait is
robust to
forecast
18,000
20,250 -9,000
$3 -15,000
.25 21,000
-15,000
$19
21,000 24,000
Wait -27,000
21,000
21,000 -45,750
No s
30,000
0
torm
-45,000
30,000
Noon 5:00pm 5:30pm
57
18,000
20,250 -9,000
$3 -15,000
.25 21,000
-15,000
$19
? 24,000
?
Wait -27,000
?
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
58
18,000
20,250 -9,000
$3 -15,000
.25 21,000
p*(-15,000) +
-15,000
$19
(1-p)*(30,000)
? 24,000
Wait -27,000
?
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
59
18,000
20,250 -9,000
$3 -15,000
.25 21,000
p*(-15,000) +
-15,000
$19
(1-p)*(30,000)
? 24,000
Wait -27,000
-45,750
No s
30,000
0.5
0.5*(30,000) +
torm
0.5*[p*(-15,000) +
-45,000 (1-p)*(30,000)]
30,000
Noon 5:00pm 5:30pm
60
Sensitivity Analysis | Breakeven
0.5*(30,000) + 0.5*[p*(-15,000)
+ (1-p)*(30,000)] = 20,250
18,000
20,250 -9,000
$3 -15,000
.25 21,000
p*(-15,000) +
-15,000
$19
(1-p)*(30,000)
? 24,000
Wait -27,000
-45,750
No s
30,000
0.5
0.5*(30,000) +
torm
0.5*[p*(-15,000) +
-45,000 (1-p)*(30,000)]
30,000
Noon 5:00pm 5:30pm
61
p = 0.433
18,000
20,250 -9,000
$3 -15,000
.25 21,000
p*(-15,000) +
-15,000
$19
(1-p)*(30,000)
? 24,000
Wait -27,000
-45,750
No s
30,000
0.5
0.5*(30,000) +
torm
0.5*[p*(-15,000) +
-45,000 (1-p)*(30,000)]
30,000
Noon 5:00pm 5:30pm
62
p = 0.433
18,000
20,250 -9,000
$3 -15,000
.25 21,000
-15,000
$19
24,000
Wait -27,000
20,250
-45,750
No s
30,000
0.5
torm
-45,000
30,000
Noon 5:00pm 5:30pm
63
p = 0.433
18,000
20,250 -9,000
$3 -15,000
.25 21,000
p*(-15,000) +
-15,000
$19
(1-p)*(30,000)
24,000
Wait -27,000
-45,750
No s
30,000
0.5
0.5*(30,000) + If p ≤ 0.433, then wait

torm
0.5*[p*(-15,000) + and go with MAF

-45,000 (1-p)*(30,000)]
30,000
Noon 5:00pm 5:30pm
64
p = 0.433
18,000
20,250 -9,000
$3 -15,000
.25 21,000
p*(-15,000) +
-15,000
$19
(1-p)*(30,000)
24,000
Wait -27,000
-45,750
No s
30,000
0.5
0.5*(30,000) + If p ≥ 0.433, then ship

torm
0.5*[p*(-15,000) + via trucks with EPD

-45,000 (1-p)*(30,000)]
30,000
Noon 5:00pm 5:30pm
65
Result of Decision Analysis

¡ An optimal decision rule (for every possible
scenario): a “complete contingency plan”
¡ Insights:
¡ If risk neutral, then computing EMV is good enough
¡ But what if risk averse?
¡ Maybe we want to be on the safe side and just
ship by truck to DC!
¡ Variability might be important to look at

66
Random Variables
Discrete
Continuous

67
Simple questions
Probability and Random
Variables
. What is the p
probabilityy that the return is at least 12%?
. P(R 12) = P(R=12) + P(R=13) + P(R=14) = 0.25+0.21+0.09 = 0.55
¡ .Remember!
What is the probability that the return is less than 12%?
¡ Probabilities are always between 0 and 1
. P(R<12) = P(R=10) + P(R=11) = 0.22+0.23 = 0.45
¡ Considering all possible outcomes, the sum of their
. Notice that must
probabilities these
add two probabilities add up to
to ONE 1
R (% per year) Probability
10 0.22
11 0.23
12 0.25
13 0.21
14 0 09
0.09

68
Simple questions
Complements
. What is the p
. P(R 12) = P(R=12) + P(R=13) + P(R=14) = 0.25+0.21+0.09 = 0.55
¡ .What
Whatis the probability
is the thatthat
probability return
theis return
at leastis12%
less than 12%?
P(R ≥ 12) = P(R=12) + P(R=13) + P(R=14) = 0.25 + 0.21 + 0.09 = 0.55
. P(R<12) = P(R=10) + P(R=11) = 0.22+0.23 = 0.45
¡ What is the probability that return is less than 12%?
. Notice that these two probabilities add up to 1
P(R < 12) = P(R=10) + P(R=11) = 0.22 + 0.23 = 0.45
10 0.22
11 0.23
12 0.25
13 0.21
14 0 09
0.09

69
Simple questions
Complements
. What is the p
. P(R 12) = P(R=12) + P(R=13) + P(R=14) = 0.25+0.21+0.09 = 0.55
¡ .What
Whatis the probability
is the thatthat
probability return
theis return
at leastis12%
less than 12%?
P(R ≥ 12) = P(R=12) + P(R=13) + P(R=14) = 0.25 + 0.21 + 0.09 = 0.55
. P(R<12) = P(R=10) + P(R=11) = 0.22+0.23 = 0.45
¡ What is the probability that return is less than 12%?
. Notice that these two probabilities add up to 1
P(R < 12) = P(R=10) + P(R=11) = 0.22 + 0.23 = 0.45
10 0.22
11 0.23 0.55 + 0.45 = 1
12 0.25
13 0.21
14 0 09
0.09

70
Example: return
Expected Valueon investment
R= 10*0.22 + 11*0.23 + 12*0.25 + 13*0.21 + 14*0.09 = 11.72
(Excel tip: SUMPRODUCT(column pi, column ri))
Interpretation: center of gravity

0.3
R (% per year) Probability 0.25
0.25 0.23
0.22
10 0.22 0.21
0.2
11 0.23
0.15
12 0.25
0.09
0.1
13 0 21
0.21
0.05
14 0.09
0
9 10 11 12 13 14 15
Distribution can be PbalancedR at the mean

71
Mean doesn(t
doesn t tell the whole
Expected Value Doesn’t Tell story
the Whole
Consider a secondStory!
portfolio with the following
p g return
distribution 0.57 0.6
S (% per year) Probability 0.5

0 43
0.43
10 0.57 0.4
11 0 0.3
12 0
02
0.2
13 0
0.1
14 0.43
0
9 10 11 12 13 14 15
The two portfolios have the same

average return: Distribution still balances at 11.72
S = 10*0.57
10*0 57 + 14*0
14*0.43
43 = 11
11.72
72 = R

72
How are
How they
are they different?
different?
0.3
0.25
0.25 0.23
0 22
0.22
0.21
0.2
0.15
0 09
0.09
0.1
0.05
0
9 10 11 12 13 14 15
0.6 0.57
0.5
0.43
0.4
0.3
0.2
0.1
0
9 10 11 12 13 14 15

73
Example
Variance calculation of variance
2
R
((10-11.72))2 * ((0.22))
0.3
0.25 + (11-11.72)2 * (0.23)
0.25 0.23
0.22
0.21
0.2 + (12-11.72)
(12-11 72)2 * (0.25)
(0 25)
0.15
+ (13-11.72)2 * (0.21)
0.09
0.1
0.05
+ (14-11.72)
(14 11 72)2 * (0
(0.09)
09)
0 = 1.6016
9 10 11 12 13 14 15
(10-11.72)2

74
Example
Variancecalculation of variance
2
S ((10-11.72))2 * ((0.57))
0.57
0.6
+ (14-11.72)2 * (0.43)
0.5
0.43
0.4
= 3.9216
3 9216
2
0.3 > 1.6016 = R
0.2
0.1 Portfolio S has more variability in

its potential return than portfolio
0
9 10 11 12 13 14 15 R
(10-11.72)2

75
Binomial Distribution
¡ Describes the distribution of
¡ the number of successes
¡ out of n independent “trials”
¡ where each trial has the same probability of success p
¡ Binomial(n,p)
¡ Need to properly define what “trial” and “success”

mean for our problem!

76
¡ Suppose we flip 4 coins n=4
¡ Success is obtaining a head with p = 1/3
¡ X = number of heads
1 2 3 4

77
¡ How can we obtain 2 heads?
1 2 3 4 1 2 3 4
H H T T T T H H
H T H T T H T H
H T T H T H H T
4! 4*3*2*1
= = 6 ways we can obtain 2 heads!
2!(4-2)! 2*1*(2*1)
78
¡ We can think of it a little differently!

¡ Total number of permutations
¡ To place the first head, we have 4 possibilities

79

¡ To place the second head, we have 3 possibilities

80

¡ To place the first tail, we have 2 possibilities
H H

81

¡ To place the second tail, we have 1 possibility
H H T

82

4*3*2*1
= 4!
¡ To place the second tail, we have 1 possibility
H T H T

83

¡ Total number of permutations = 4!
¡ But we don’t care about the order of the heads or the tails!
¡ Suppose we choose the following break-down
H T H T
¡ In how many ways can I place the heads in the grey bins?

84

H T H T
¡ In how many ways can I place the heads in the grey bins? 2!
¡ In how many ways can I place the tails in the red bins?

85

H T H T
¡ In how many ways can I place the tails in the red bins? (4-2)!

86

H T H T
¡ In how many ways can I place the tails in the red bins? (4-2)!
¡ Since we do not care about the order: 4!/(2! * 2!)

87
¡ P(X=2) = probability that we obtain 2 heads and the remaining
are tails
¡ Each grey bin has a probability p=1/3 to have a head
1/3 * 1/3 = (1/3)2
¡ Each red bin has a probability 1-p = 2/3 to have a tail
2/3 * 2/3 = (2/3)4-2
H T H T
4! (1/3)2 * (2/3)4-2
2!(4-2)!
88
¡ In general, if X is Binomial(n,p)
¡ X can take on only the values 0,1,…,n-1,n
¡ The probability that we have x successes out of n trials
n! * (p)x * (1-p)n-x
x!(n-x)!
Number of combinations
with x successes out of n Probability that the
trial remaining (n-k) trials are
Probability that x failures
trials are
successes
89
Binomial RVs
¡ Expected value
E[X] = np
¡ Variance
Var(X) = np(1-p)

90
Revisiting Taylor Swift
15.730 - Decision Trees and Discrete Probability 90

91
n = 50.9M, p = 0.005
E(X) = 254,500
Std(X) = 503
CV(X) = Std(X)/E(X) = 0.0019
91
92
n = 50.9M, p = 0.005
E(X) = 254,500
Std(X) = 503
CV(X) = Std(X)/E(X) = 0.0019 Same p: E(X) = 0.67
Std(X) = 0.816
15.730 - Decision Trees and Discrete Probability CV(X) = 1.219
92
93
n = 50.9M, p = 0.005
E(X) = 254,500
Std(X) = 503
CV(X) = Std(X)/E(X) = 0.0019 P = 0.0005: E(X) = 0.0067
Std(X) = 0.258
15.730 - Decision Trees and Discrete Probability CV(X) = 3.862
93
94
Binomial Distribution -
Example
¡ United’s first class cabin has 10 seats in each plane.
¡ Overbooking policy is to sell up to 11 first class tickets since

cancellations and no-shows are always possible
¡ Suppose that for a given flight

¡ 11 first class tickets are sold
¡ Each passenger has 80% chance of showing up for the flight
¡ Whether a passenger shows up is independent of other passengers
¡ Can we model this as a binomial distribution?

95
Example


¡ Can we model this as a binomial distribution? YES!

96
Example


¡ Trial = Passengers, Success = Showing up n =11, p = 0.8

97
Example
¡ X = number of passengers that show up
¡ X is Binomial(11, 0.8)
¡ What is the probability that 10 passengers show up?

98
Example
¡ What is the probability that 10 passengers show up?
P(X=10) = 11! (0.8)10 * (1-0.8)11-10

10!(11-10)!
= 11 * (0.8)10 * 0.2 ~ 0.236

99
Example
¡ What is the probability that the airline gets away with booking?

100
Example
P(X<=10)

101
Example
P(X<=10) = P(X=0) + P(X=1) + … + P(X=10)

102
Example
P(X<=10) = P(X=0) + P(X=1) + … + P(X=10)
¡ Or alternatively:
P(X<=10) = 1 – P(X>10) = 1 – P(X=11)

103
Example
P(X<=10) = P(X=0) + P(X=1) + … + P(X=10)
P(X<=10) = 1 – P(X>10) = 1 – P(X=11)

=1– 11! 0.811 * (1-0.8)11-11
11!(11-11)!
104
Example
P(X<=10) = P(X=0) + P(X=1) + … + P(X=10)
P(X<=10) = 1 – P(X>10) = 1 – P(X=11)

=1– 11! 0.811 * (1-0.8)11-11
11!(11-11)!
~ 0.914
105
Example
¡ What if the airline overbooked by selling

¡ 11 seats: P(X<=10) ~ 0.914
¡ 12 seats: P(X<=10) ~ 0.725
¡ 13 seats: P(X<=10) ~ 0.498

106
Extensions
¡ Given some additional data
¡ Fares prices
¡ Cost of too many passengers showing up (refunds,
damage to customer relations, etc.)
¡ Is it worthwhile to overbook flights?
¡ Check our assumptions

¡ Is 80% an accurate probability?
¡ Are passengers really independent?

107
Alternate Model
¡ We can define the “success” to be a passenger not showing
up
¡ Y = number of passengers not showing up is Binomial(11,0.2)
¡ P(X=10) = P(exactly 10 passengers show up)
= P(exactly 1 passenger does not show up)
= P(Y=1) = 11!
0.21 * (1-0.2)11-1
1!(11-1)!

108
Wrap-up
¡ Due electronically on Monday February 2, 2015
¡ AOG case – work with your team!
¡ Exercises 2.13 and 2.30 – complete individually!
¡ Submit the PDF files on Stellar.
¡ Google Doc for Teams: bit.ly/DMD16-Teams
¡ Office hours for 30 minutes!
¡ Feel free to raise your hand and unmute yourself

to ask questions!

109
Office Hour

DMD 1

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

DMD 1

Uploaded by

Copyright:

Available Formats

1

Welcome to DMD Recitation 1!

¡ Google Doc for Teams: bit.ly/DMD16-Teams

¡ To reduce background noise, please mute your phone/

¡ Please feel free to raise your hand or chat through WebEx if

15.730 - Decision Trees and Discrete Probability

¡ Use data and build models

Data Uncertainty Models

15.730 - Decision Trees and Discrete Probability

15.730 - Decision Trees and Discrete Probability

¡ Two types of nodes

15.730 - Decision Trees and Discrete Probability

15.730 - Decision Trees and Discrete Probability

15.730 - Decision Trees and Discrete Probability

Event Node Branches

15.730 - Decision Trees and Discrete Probability

Should We Play the Lottery?

Should We Play the Lottery?

Should We Play the Lottery?

Decision Analysis Procedure

¡ List uncertain events (event nodes)

¡ Construct a decision tree

¡ Determine the probabilities of each outcome

¡ Determine the numerical values of the endpoints

¡ Solve using backward induction

¡ Perform sensitivity analysis (What-if scenarios)

15.730 - Decision Trees and Discrete Probability

Kendall Crab and Lobster

¡ With the chance of Logan airport closing,

15.730 - Decision Trees and Discrete Probability

15.730 - Decision Trees and Discrete Probability

Earnings and Refund Policy

¡ When KCL ships a lobster via United Express, its

¡ If KCL cannot deliver the lobsters to customers, its

15.730 - Decision Trees and Discrete Probability

Changes to Operations Due to

15.730 - Decision Trees and Discrete Probability

Changes to Operations Due to

¡ Deliver lobsters by truck to DC via the Eastern

15.730 - Decision Trees and Discrete Probability

What is the Best Decision?

15.730 - Decision Trees and Discrete Probability

What is the Best Decision?

15.730 - Decision Trees and Discrete Probability

What is the Best Decision?

1. Use EPD to 1. USE MAF to

15.730 - Decision Trees and Discrete Probability

What is the Best Decision?

15.730 - Decision Trees and Discrete Probability

What is the Best Decision?

If EPD, additional Storm:

15.730 - Decision Trees and Discrete Probability

What is the Best Decision?

If EPD, additional Storm:

15.730 - Decision Trees and Discrete Probability

What is the Best Decision?

If EPD, additional Storm: If MAF, cost is uncertain:

15.730 - Decision Trees and Discrete Probability

What is the Best Decision?

If EPD, additional Storm: If MAF, cost is uncertain:

15.730 - Decision Trees and Discrete Probability

Noon 5:00pm 5:30pm

Noon 5:00pm 5:30pm

(-1)3,000 + (-20)(0.7)*3,000 = -45,000

0.5[p(-15,000) + and go with MAF

0.5[p(-15,000) + via trucks with EPD