DECISION
TREE
ANALYSIS
MONTE
CARLO
SIMULATION
GAME
THEORY
EXAMPLE
Project
details
Annual
output
(nos.)
=
2,50,000
Expected
sales
revenue
=
Rs.
50,00,000/-
Total
cost
=
42,50,000/-
Fixed
costs
=
Rs.
25,00,000/-
Calculate
BEP.
Also
calculate
the
eect
on
BEP
if
The
price
comes
down
by
10%
per
unit.
The
xed
cost
increases
to
Rs.
30,00,000/-
The
variable
cost
increases
by
15%
Determine
the
project
sensi^vity
&
advise
your
Client.
DECISION
TREE
ANALYSIS
DECISION
TREE
ANALYSIS
Graphical
technique
Analyzing
pros
&
cons
of
alterna^ve
decisions
It
is
a
diagramma^c
representa^on
of
the
logical
rela^onship
between
the
dierent
parts
of
a
complex
situa^on
&
possible
outcomes
of
dierent
decisions.
DECISION
TREE
ANALYSIS
A
decision
tree
is
made
up
of
nodes
&
branches.
Decision
Nodes
decision
point
represented
by
a
square.
Chance
nodes
chance
point
represented
by
a
circle.
Dierent
alterna^ves
for
the
given
situa^on
emerges
from
the
decision
point.
At
each
chance
point,
the
dierent
possible
outcomes
of
one
of
the
alterna^ve
decisions
are
marked.
CASE
ILLUSTRATION
EXPECTED
MONETARY
VALUE
The
eec^veness
of
any
decision
can
be
measured
only
in
monetary
terms.
So,
the
outcomes
of
all
decisions
are
measured
in
terms
of
expected
monetary
value
(EMV).
EMV
provides
a
common
base
for
comparing
the
outcomes
of
dierent
decisions
and
choosing
the
one
that
is
found
more
advantageous.
RULES
FOR
DRAWING
DECISION
TREE
1. Iden^fy
all
alterna^ve
decisions
available
at
the
ini^al
decision
point.
2. Iden^fy
all
possible
subsequent
decision
points.
3. Iden^fy
all
possible
chance
points
&
the
likely
outcomes
at
the
chance
points.
4. Develop
a
decision
tree
diagram
showing
in
sequence
the
decision
points
&
the
chance
points.
Construct
the
decision
tree
diagram
from
leh
to
right.
Denote
decision
points
by
a
square
&
chance
points
by
a
circle.
RULES
FOR
DRAWING
DECISION
TREE
(contd)
5. Aher
construc^ng
the
decision
tree,
work
backwards
(i.e.
from
right
to
leh)
compu^ng
the
EMV
of
each
chance
point
&
each
decision
point
^ll
the
ini^al
decision
point
is
reached.
6. Determine
the
best
alterna^ve
at
the
ini^al
decision
point.
EXAMPLE
MONTE
CARLO
SIMULATION
MONTE
CARLO
SIMULATION
It
is
a
code
name
given
by
Von
Newmann
&
Ulam
to
the
technique
of
solving
problems
using
random
numbers.
Monte
Carlo
technique
can
be
used
to
solve
a
variety
of
problems
involving
stochas^c
situa^ons
(a
stochas^c
situa^on
is
one
where
some
or
all
parameters
of
the
problem
are
described
by
random
variables).
Best
alterna^ve
when
a
mathema^cal
solu^on
is
highly
complex/
impossible.
STEPS
IN
MONTE
CARLO
TECHNIQUE
1. From
the
given
probability
of
occurrence
of
events,
establish
cumula^ve
probability.
2. Assign
tag
nos.
to
the
events
in
such
a
way
that
the
tag
nos.
represent
the
cumula^ve
probability.
3. Obtain
random
nos.
from
a
random
no.
table.
4. Correlate
the
random
nos.
with
the
tag
nos.
assigned
to
the
events
&
iden^fy
the
value
for
the
respec^ve
events.
CASE
ILLUSTRATION
It
is
observed
that
the
demand
for
a
product
varies
in
a
random
fashion.
The
demand
per
day
is
observed
to
have
the
following
probability.
Simulate
the
demand
for
30
days
using
Monte
Carlo
simula^on.
Demand
per
25
33
42
51
day
Probability
0.15
0.25
0.45
0.15
GAME
THEORY
GAME
THEORY
Business
rms
compete
with
each
other.
Hence,
there
are
conic^ng
interests.
The
businesses
try
to
achieve
their
goals
by
formula^ng
strategies.
Technique
developed
by
Von
Newmann
&
named
as
Game
theory.
PROPERTIES
OF
THE
GAME
1. No.
of
compe^tors
are
nite.
2. Each
player
has
nite
no.
of
strategies.
3. All
players
need
not
have
same
no.
of
strategies.
4. Each
player
chooses
to
play
a
single
strategy
out
of
those
available
to
him.
All
players
are
assumed
to
make
their
choice
simultaneously.
5. The
outcome
of
the
play
depends
upon
the
strategies
followed
by
the
players.