TV pilot

Hale's TV Productions

Hale's TV Productions has acquired a script for a pilot episode of a new television
show. A competitor has heard of the script, and offered Hale $100,000 for the
script and rights to the series concept.

If Hale decides to produce the pilot and market the series themselves, they'll face
after-tax production costs of $100,000. If they then fail to sell the series to a network (s 1),
they'll not earn anything from the pilot. However, if a network offers them a one-year (s 2)
or two-year (s3) contract, they'll net a profit on their investment.

The payoff table (profit in $1000s) for Hale's TV Production follows:

States of Nature
s1 s2 s3
Produce d1 -100 50 150
Decision Pilot
Alternatives Sell to d2 100 100 100
Competitor
Probability of States of Nature 0.2 0.3 0.5

For a consulting fee of $2500, an agency will review the plans for the comedy series and indicate the overall
chance of a favorable network reaction to the series. If the special agency review results in a favorable (I 1) or an
unfavorable (I2) evaluation, what should Hale's decision strategy be? Assume that Hale believes the following
conditional probabilities are realistic appraisals of the agency's evaluation accuracy.

P(I1|s1) = .3, P(I2|s1) = .7

P(I1|s2) = .6, P(I2|s2) = .4
P(I1|s3) = .9, P(I2|s3) = .1

Show the decision tree for this problem.

What is the recommended decision strategy and the expected value, assuming the agency information is obtained.

Page 1
 TV pilot

Is the $2500 consulting fee worth the information? What is the maximum that Hale should be willing to
pay for the consulting information?

network decision 8.70%

reject buy 1 yr buy 2 yrs reject
pilot -$100 $50 $150
decision sell out $100 $100 $100
26.09%
network
likelihood 20.00% 30.00% 50.00% pilot 1 year
$102.17
Pr( review | network decision ) 2 years
decide
65.22%
reject buy 1 yr buy 2 yrs likelihood $102.17
review + 30.00% 60.00% 90.00% 69.00%
sell out
review - 70.00% 40.00% 10.00% 31.00% good $100

Pr( network decision | review ) 69%

reject buy 1 yr buy 2 yrs review 45.16%
review + 8.70% 26.09% 65.22%
review - 45.16% 38.71% 16.13%
$101.50 reject
31%
E[ profit ] 38.71%
bad network
pilot 1 year
review + review - ($1.61)
pilot $102.17 -$1.61 2 years
sell out $100.00 $100.00 decide

maximum $102.17 $100.00 16.13%

$100
$101.50
sell out
$100
Hale's has an expected profit of $101,500 if they acquire the
review and act accordingly. Since they could make $100,000
simply by selling out, the review isn't worth more than $1,500
to them.

Page 2
 TV pilot

In fact, depending on their attitude towards financial risk, they

might want to sell out even after the review. After all, to gain the
extra $1,500 in expected payoff, they have to risk some chance of
a $100,000 loss. (That is, they have to prepare the pilot if the
review is favorable. But even after a favorable review, there's a 8.7%
chance that the network will reject their offering.)

Page 3
 TV pilot

8.70% ($100)
reject

26.09%
etwork $50
1 year
102.17
2 years
65.22% $150

$100

45.16% ($100)
reject

38.71%
etwork $50
1 year
$1.61)
2 years
16.13% $150

$100

Page 4
 Snow removal

Martin's Service Station

Martin is thinking about offering driveway-plowing services this winter. To do so, he'd need
either to purchase a blade which could be mounted onto one of his tow-trucks (which would
let him plow when the tow-truck wasn't needed for towing), or to buy a new snowplow vehicle
dedicated purely to snow removal.

How well his decision works out will depend, of course, on how heavy the snowfall is over
the winter season.

The payoff table ($) for Martin's Service Station follows:

Snowfall
Heavy Moderate Light
s1 s2 s3
Purchase d1 7000 2000 -9000
snowplow
Decision Do not d2 0 0 0
Alternatives invest
Purchase d3 3500 1000 -1500
blade
Probability of States of Nature 0.4 0.3 0.3

Suppose Martin decides to wait to check the September temperature pattern before making a final decision.
Estimates of the probabilities associated with an unseasonably cold September (I 1) are P(I1|s1) = .30,
P(I1|s2)= .20, P(I1|s3) = .05. If Martin observes an unseasonably cold September, what is the recommended
decision? If Martin does not observe an unseasonably cold September (I2), what is the recommended
decision?

61.54% $7000
snowfall heavy
heavy moderate light
snowplow $7,000 $2,000 -$9,000 30.77%
decision nothing $0 $0 $0 snow $2000
blade $3,500 $1,000 -$1,500
moderate
$4231
snowplow light
Page 5 7.69% ($9000

decide $0
nothing
 heavy

Snow removal 30.77%

snow $2000
moderate
$4231

likelihood 40.00% 30.00% 30.00% snowplow light

7.69% ($9000
Pr( September | snowfall )
decide $0
heavy moderate light likelihood nothing
$4231
Sept. cold 30.00% 20.00% 5.00% 19.50% 61.54% $3500
Sept. mild 70.00% 80.00% 95.00% 80.50%
blade heavy
Pr( snowfall | September )
30.77%
cold 19.5% snow $1000
heavy moderate light moderate
Sept. cold 61.54% 30.77% 7.69% $2346
Sept. mild 34.78% 29.81% 35.40% light
7.69% $1500
E[ profit ]
September

Sept. cold Sept. mild $1617.50

snowplow $4,230.77 -$155.28 34.78% $7000
nothing $0.00 $0.00
heavy
blade $2,346.15 $984.47
maximum $4,230.77 $984.47
mild 80.5% 29.81%
$1,617.50 snow $2000
moderate
($155)
Martin should buy the snowplow if September is cold, and snowplow light
only the blade if September is mild. This yields him an expected
35.40% ($9000
profit of $1,617.50.
decide $0
nothing
$984
34.78% $3500
blade heavy

29.81%
snow $1000
moderate
$984
light
Page 6
35.40% $1500
 $984
34.78% $3500
blade heavy
Snow removal
29.81%
snow $1000
moderate
$984
light
35.40% $1500

Page 7
 Snow removal

61.54% $7000
vy

30.77%
$2000
moderate

t
7.69% ($9000) Page 8
vy

30.77% Snow removal

$2000
moderate

t
7.69% ($9000)

61.54% $3500
vy

30.77%
$1000
moderate

t
7.69% $1500

34.78% $7000
vy

29.81%
$2000
moderate

t
35.40% ($9000)

34.78% $3500
vy

29.81%
$1000
moderate

t
Page 9
35.40% $1500
 34.78% $3500
vy
Snow removal
29.81%
$1000
moderate

t
35.40% $1500

Page 10
 Sealcoat

Sealcoat, Inc.

Sealcoat, Inc. has a contract with one of its customers to supply a unique liquid chemical product that will be used
by the customer in the manufacture of a lubricant for airplane engines. Because of the chemical process used by
Sealcoat, batch size for the liquid chemical product must be 1000 pounds. The customer has agreed to adjust
manufacturing to the full batch quantities and will order either one, two or three batches every three months. Since an
aging process of one month is necessary for the product, Sealcoat will have to make its production (how much to
make) decision before the customer places an order. Thus, Sealcoat can list the product demand alternatives of
1000, 2000, or 3000 pounds, but the exact demand is unknown.

Sealcoat's manufacturing costs of $150 per pound, and the product sells at the fixed contract price of $200 per
pound. If the customer orders more than Sealcoat has produced, Sealcoat has agreed to absorb the added cost of
filling the order by purchasing a higher quality substitute product from another chemical firm. The substitute produce,
including transportation expenses, will cost Sealcoat $240 per pound. Since the product cannot be stored more than
two months without spoilage, Sealcoat cannot inventory excess production until the customer's next three-month
order. Therefore, if the customer's current order is less than Sealcoat has produced, the excess production will be
reprocessed and valued at $50 per pound.

The inventory decision in this problem is how much Sealcoat should produce given the costs and the possible
demands of 1000, 2000, or 3000 pounds. From historical data and an analysis of the customer's future demands,
Sealcoat has assessed the probability distribution for demand shown in the table below.

Demand Probability
1000 30%
2000 50%
3000 20%

Develop a payoff table for the Sealcoat problem.

How many batches should Sealcoat produce every three months?
How much of a discount should Sealcoat be willing to allow the customer for specifying in advance exactly how
many batches will be purchased?

Sealcoat has identified a pattern of demand for the product based on the customer's previous order quantity. Let:

I1 = customer's last order was 1000 pounds.

I2 = customer's last order was 2000 pounds.
I3 = customer's last order was 3000 pounds.

The conditional probabilities follow:

P(I1|s1) = .10, P(I2|s1) = .40, P(I3|s1) = .50

P(I1|s2) = .22, P(I2|s2) = .68, P(I3|s2) = .10
P(I1|s3) = .80, P(I2|s3) = .20, P(I3|s3) = .00

Develop an optimal decision strategy for Sealcoat.

Net profit actual demand

1000 2000 3000 E[profit] $150 cost/lb produced

Page 11
 Sealcoat

1000 $50,000 $10,000 -$30,000 $14,000 $240 cost/lb purchased

produced 2000 -$50,000 $100,000 $60,000 $47,000 $200 revenue/lb sold
3000 -$150,000 $0 $150,000 -$15,000 $50 revenue/lb reprocessed
maximum $50,000 $100,000 $150,000 $95,000

likelihood( demand ) 30.00% 50.00% 20.00% The decision tree for this
problem would have 27 terminal
With no other information, the best Sealcoat can do is to branches. Since the tree is
produce 200 pounds, yielding an expected profit of $47,000. merely a way to "picture" your
calculations, this is a good
With perfect information, Sealcoat can produce precisely the correct opportunity to try a problem
amount every month, yielding an expected profit of $95,000. without using the tree.

Pr( last | this ) actual demand likelihood( last )

1000 2000 3000
demand 1000 10.00% 22.00% 80.00% 30.00%
last 2000 40.00% 68.00% 20.00% 50.00%
month 3000 50.00% 10.00% 0.00% 20.00%

Pr( this |last ) actual demand

1000 2000 3000
demand 1000 10.00% 36.67% 53.33%
last 2000 24.00% 68.00% 8.00%
month 3000 75.00% 25.00% 0.00%

E[ profit ] demand last month

1000 2000 3000
production 1000 -$7,333 $16,400 $40,000
this 2000 $63,667 $60,800 -$12,500
month 3000 $65,000 -$24,000 -$112,500
maximum $65,000 $60,800 $40,000

E[ profit from acting optimally ] $57,900

By producing 3000 pounds after a month in which 1000 pounds was demanded,
2000 pounds after 2000 was demanded, and 1000 after 3000 was demanded,
Sealcoat will have an expected profit of $57,900.
22.71%

This moves them 22.71% of the way up from $47,000 (the best they can do with
no information) towards $97,000 (the best they can do with perfect information).
The information contained in last month's demand figure is worth $10,900 to them.

Page 12
 Sealcoat

ct that will be used

process used by
greed to adjust
hree months. Since an
ion (how much to
d alternatives of

ice of $200 per

b the added cost of
e substitute produce,
be stored more than
next three-month
production will be

d the possible
s future demands,

nce exactly how

rder quantity. Let:

Page 13
 Sealcoat

b reprocessed

ion tree for this

would have 27 terminal
Since the tree is
way to "picture" your
ns, this is a good
y to try a problem
sing the tree.

Page 14