DECISION TREE

DECISION TREE
Decision tree is a graphical representation of the decision process
indicating decision alternatives, states of nature, probabilities
attached to the states of nature and profits and losses associated
with it. It consists of nodes & branches. The following symbols are
used:

Decision node
State of nature
Branches
A decision tree is generally useful for multistage situation which
involves a series of decisions each depends on the preceding one.
A decision tree look like S1 D DECISION NODE
p1

CB A1, A2, A3
p2
C1 CB
S2 ALTERNATIVES
A1
C1, C2, C3
p1 S1
A2 CB
CHANCE NODE
D C2
CB p2
CB CHANCE BRANCH
S2
A3

p1, p2
C3 CB p1
S1 PROBABILITY

CB p
2
S1, S2
RESPECTED PAYOFF
S2
 ILLUSTRATION 1
Suppose we have the decision making problem represented by the
following table:
Course of Action
Demand Prob S1 (Subcontracting) S2 S3 (Construct
(Begin facilities)
Overtime)
Low 0.10 10 - 20 - 150
Medium 0.50 50 60 20
High 0.40 50 100 200

Show this decision situation in the form of a decision tree & indicate
the most preferred decision & corresponding expected value.
 SOLUTION
A decision tree which represents possible course of action & state of
nature are shown as:
0.1 X 10 = 1
.1
0
p=
L 46
C1 M
p = 0.5
0.5 X 50 = 25
H
S1 EMV = 46
p=0
.4
0.4 X 50 = 20
S2 L p=0
.1 0.1 X -20 = -2
D C2 M
p = 0. 5
0.5 X 60 = 30 68
H
EMV = 68 p=
EMV = 75 0.4
0.4 X 100 = 40
S3

L p = 0.1 0.1 X -150 = -15

C3 p = 0.5
M
0.5 X 20 = 10 75
EMV = 75 H p=
0. 4
0.4 X 200= 80
SINCE NODE 3 HAS THE HIGHEST EMV,
THEREFORE THE DECESION AT NODE D WILL BE
CHOOSEN THE COURSE OF ACTION S3 I.E.
CONSTRUCT NEW FACILITY.
 ILLUSTRATION 2
A businessman has two independent investment portfolios A & B
available to him but he lacks the capital to undertake both of them
simultaneously. He can choose
A first and then stop, If A is successful
then
The take B or of
probability vice-versa
success of A is 0.6 while that for B it is 0.4. Both
investment schemes requires an initial capital outlays of Rs 10,000
and both return nothing if the venture is unsuccessful. Successful
completion of A will return Rs.20000 and successful completion of
B returns Rs. 24000. Draw a decision tree & determine best
strategy.
 SOLUTION
STOP Rs 9600
Rs 0
0 .4
s = 000
D s 24
0.6 1 ACCEPT B R
EMV = 10160 s =
0 00
2 EMV = 3600
0 C3
E Rs
M A C 1
f=
EMV = 3600 f=
0
- R .6
V 0.4 - Rs 10000 s1
000 - Rs 6000
= 0
1 D
0 - Rs 10000
1 f = 0. 6 Rs 0
6 B C2 P
Rs 12000
Rs S =

.6
O
0 ST

Rs s = 0
0

0
EMV = 6800 240 .4

00
00 EMV = 8000

20
D2 ACC
E PT A C4
EMV = 8000
f = 0.4 - Rs 4000
- Rs 10000
EMV OF NODE C3 = 0.4 x 24000 + 0.6 x (-10,000)
= Rs.3600
EMV OF NODE D1 = Rs.3600
EMV OF NODE C1 = 0.6x(20000+3600)+0.4 x (-10,000)
= 10160
EMV OF NODE C4 = 0.6 x 20000 + 0.4 x (-10,000)
Rs.8000
EMV OF NODE D2 = Rs.8000
EMV OF NODE C2 = 0.4x(24000+8000)+0.6 x (-10,000)
= Rs.6800
Thus the optimal decision is to accept
investment A, and if successful, then accept B.
 ILLUSTRATION 3
A large steel manufacturing company has three option with regards
to production: (i) produce commercially (ii) build pilot plant (iii)
stop producing steel. The management has estimated that their
pilot plant if built has 0.8 chances of high yield & 0.2 chances of
low yield. If the pilot plant does show a high yield, management
assigns a probability of 0.75 that the commercial plant will also
have a high yield. If the pilot plant shows a low yield there is only
a 0.1 chance that the commercial plant will show a high yield.
Finally management’s best assessment of the yield on a
commercial – size plant without building a pilot plant first has a
0.6 chance of high yield. A pilot plant will cost Rs. 3,00,000. The
profits earned under high & low yield conditions are Rs.
1,20,00,000 and – Rs. 12,00,000 respectively. Find the optimum
decision for the company.
 SOLUTION Rs.1,20,00,000
GH 6
EMV = 67,20,000
H I 0.
p=
C1 LO p= 0.4
Rs.1,20,00,000
W

.
-Rs.12,00,000

ly

.Y
er e
m uc
al
75

H
.
ci
o m ro d
0
p=
C
P

EMV = 67,20,000 c e 3 =0 p
d u l l y L . .25
Rs 0 Pr erci EMV = 87,00,000 Y.
C

Stop o a
D o mm -Rs.12,00,000
D1 C
Rs 0
Pi Rs.3
Bu Pla ,000

EMV = 87,00,000 Stop Rs.1,20,00,000

lot ,00
ild nt

Y.
.8

.
0

EMV = 1,20,000 Y
H.

H.
p=

= 0.1
C2 C2 p
p
L. = 0.9
ce ally Y .
EMV = 6936000 u i
p=

r o d e rc -Rs.12,00,000
0
L.

P m
EMV = 1,20,000
.2

om
Y.

C Stop
D2 Rs 0
 ILLUSTRATION 4
Mr. X of ABC Ltd. Wants to introduce a new product in the market.
He has a choice of two different research and development plans A
& B. A costs Rs. 10 lakhs and has a 40 percent chance of success
where as B costs Rs. 5 lakhs with a 30 percent chance of success. In
the event of success, Mr. X has to decide whether or not to
advertise the product heavily or lightly. Heavy advertising will cost
Rs. 4 lakhs but gives 0.7 probability of full acceptance and 0.3
probability of partial acceptance by the market. Light advertising
will cost Rs. 1 lakh with the probability 0.5 of full acceptance and
0.5 probability of partial acceptance. Full market acceptance of
the product develop as per plan A would be worth Rs. 40 lakhs and
as per plan B would be worth Rs. 30 lakhs. Partial acceptance in
both cases will be worth Rs 20 lakhs. Which plan Should Mr. X
adopt and what sort of advertising will be done for marketing the
product? Solve the problem with the help of decision tree.
 SOLUTION
A decision tree which represents the possible courses of action & state of
nature is shown in following figure. Rs. 40 L
Rs. 34 L . A c.
Rs. 30 L F
= 0.7
H.Ad. p
D 1 C 2 P.A
Rs. 12 L s = 0.4 Rs . 4 L
p
c.
L. Rs =0
.3
Ad . 1 Rs. 20 L
C1 f=
. L F.Ac.
.5
0. 6 p=0 Rs. 40 L
Rs. 2 L A 0 L C3 P.A
1
R s. Rs. 0 Rs. 30 L p= .
c
D Rs
0.5
Rs. 20 L
B . 5L Rs. 27 L F.Ac0.7
. Rs. 30 L
=p
s = 0.3 H.Ad. C5 P.A
C4 D2 Rs. 4 L p= .
c
Rs. 24 L Rs 0.3
Rs. 20 L
Rs. 7.2 L L. . 1 L
f=

A .
d. F.Ac0.5 Rs. 30 L
0.

p=
7

Rs. 0 C6 P.A
p = c.
Rs. 25 L 0.5 Rs. 20 L