Professional Documents
Culture Documents
TOOLS
• A good decision –one that uses analytic decision making –based on logic and
considers all available data and possible alternatives.
• It also follows these six step:
1. Clearly define the problem and the factors that influence it.
2. Develop specific and measurable objectives.
3. Develop a model –that is, a relationship between objectives and variables (which are
measurable quantities).
4. Evaluate each alternative solution based on its merits and drawbacks.
5. Select the best alternative.
6. Implement and evaluate the decision and then set a timetable completion.
OBJECTIVE OF DECISION ANALYSIS
• A method for reducing UNCERTAINTY in the decision making process through DATA
ANALYSIS.
• Provides a RATIONAL PROCESS for making decisions that is usually far better than
random selection.
• Easy to update if conditions change
ANALYTIC DECISION MAKING
• One that requires models, objective and quantifiable variables, often in form of
probabilities and payoffs.
Introduction to two decision making tools
These two tools are used in numerous OM situations ranging from new product
analysis to capacity planning, to location planning to supply chain disaster planning
• Decision Table
• Decision Tree
DECISION TERMINOLOGY
• Getz Product Company is investigating the possibility of producing and marketing backyard storage sheds. Undertaking
this project would require the construction of either a large o a small manufacturing plant. The market for the
product produced –storage sheds–could be either favorable or unfavorable. Getz, of course has the option of not
developing a new product line at all
States of Nature
“ Out of control”
Set up cost /
Possible
Loss
Alternative
Decisions
“what could
we do”
Probabilities
DECISION TABLE
MAXIMAX
This decision criterion locates the alternative with the highest possible gain
The decision for hopeless OPTIMIST. What is the best that can happen?
No probabilities are involved.
The maximum payoff from any state of nature for each decision alternative
THE CRITERIA FOR DECISION MAKING
UNDER UNCERTAINTY
MAXIMIN
This decision criterion locates the alternative that has the least possible loss.
The decision for the hardened PESSIMIST or the very cautious.
What is the best of the worst that can happen? No probabilities are involved
The best minimum payoff from any state of nature for each decision alternative.
THE CRITERIA FOR DECISION MAKING
UNDER UNCERTAINTY
EQUALLY LIKELY
This approach assumes that each state of nature is equally likely to occur.
200,000 + (-180,000)(.5) = 10,000
Given even state probabilities, and using a weighted average, what is the best payoff?
In equally likely criterion we set probabilities to be the same/even.
DECISION MAKING UNDER RISK
• Several possible state of nature may occur, each with an assumed probability.
Given a decision table with conditional values and probability assessments for
all states of nature, EVM for each alternative can be determined.
FORMULA:
In this case Getz should
built a SMALL PLANT
having the highest EMV
SOLUTION:
EMV (A1) = (0.6)(200,000)+(0.4)(-180,000) = 48,000
EMV (A2) = (0.6)(100,000)+(0.4)(-20,000) = 52,000 The
EMV (A3) = (0.6)(0)+(0.4)(0) = 0 Maximum
EMV
DECISION MAKING UNDER CERTAINTY
• Supposed that Getz operations manager has been approached by a marketing research firm that proposes to
help him make the decision about whether to build the plant to produce storage sheds.The marketing
researcher claim that their technical analysis will tell Getz with certainty whether market is favorable for the
proposed product.
• The marketing firm would charge Getz 65,000 for this information, that could prevent Getz from making a
very expensive mistake in the first place.
THE QUESTIONS ARE: Insights:
If manager able to determine which state of nature
• Should Getz hire the firm to make the study? would occur then he would KNOW which
• Is it worth 65,000? DECISION to make (KNOWLEDGE now has a
VALUE)
• What it might be worth?
becomes
Probability payoff + (with
knowledge)
PERFECT INFORMATION
• Solid, quality reliable information can be hard to come by. Firms hire specialists,
consultants, academics, etc. to use their expertise in order to greatly improve
the quality of fundamental data.
- Geologists, Economist Assessment, Marketing Specialist, etc
The goal is to reduce RISK and increase CERTAINTY
Solution:
EVwPI = 120,000
Maximum EMV = 52,000 (A2 small plant)
Insights:
Formula:
EVPI = Expected Value with Perfect Information – Maximum EMV As the conditional value
Solution: of “large plant/favorable
market” increases, the
EVPI = 120,000 - 52,000 = 68,000 EVM also increases and
• The most Getz should be willing to pay for the perfect information is 68,000. This will affect the value of
conclusion of course is based on the assumption that the probability of the first state EVwPI and EVPI, would
of nature is 0.6 and the second is 0.4. also increase
How does EVPI change if the “large plant/favorable market” conditional value is
250,000?
EVPI = 150,000 - 78,000
(Answer: the EMV (A1) = 78,000, AI is now the preferable decision) = 72,000
DECISION TREE
F ive s t e p s in a n a ly z in g
p rob le m s w it h d e c is ion t re e :
1 . D e f in e t h e p rob le m
2 . S t r u c t u re or d r aw t h e
d e c i s i on t re e .
3 . A s s ig n p rob a b ilit ie s t o t h e
s t a t e of n a t u re .
4 . E s t i m a t e p ayof f s for e a c h
NEED TO p os s ib le c om b in a t ion of
d e c is ion a lt e r n a t ive s a n d
KNOW s t a t e s of n a t u re .
5 . S olve t h e p rob le m by
c om p u t in g t h e E M V for
e a c h s t a t e s of n a t u re
n od e .
C O M P L E T E D D E C I S I O N T R E E F O R G E T Z P RO D U C T S
• Th e b ran ch leavin g th e
NEED TO decision n ode to th e
KNOW state -of -n atu re n ode
with th e h igh est E MV
will b e ch osen . in Getz’s
case , a small p lan t sh ou ld
b e b u ilt
A MORE COMPLEX DECISION TREE
• Note all possible outcomes and alternatives are included in their logical sequence and should be
examined carefully.
LETS EXAMINE THE TREE
1 s t d e c is ion p oin t =
c o n d u c t m a r ke t s u r vey
v s n ot t o c on d u c t
m a r ke t s u r vey
2nd decision point = to
b u ilt la r g e p la n t , s m a ll
p la n t o r t o d o n o t h in g
P ro b a b i l i t y o f f avo r a b l e
s u r vey re s u lt = 4 5 %
NEED TO P ro b a b ilit y o f n e g a t ive
s u r vey re s u l t = 5 5 %
KNOW
U P P E R PA RT O F T R E E
• S t a t e - o f - n a t u re - n od e 2 &
3 f avor a b l e m a r ke t
p rob a b ilit y = 7 8 %
• S t a t e - o f - n a t u re - n od e 2 &
3 u n f avor a b l e m a r ke t
p rob a b ilit y = 2 2 %
S t a t e - o f - n a t u re - n od e 4
& 5 f avor a b le m a r ke t
p rob a b ilit y = 2 7 %
S t a t e - o f - n a t u re - n od e 4
& 5 u n f avor a b le m a r ke t
p rob a b ilit y = 7 3 %
NEED TO
KNOW L OW E R PA RT O F T H E
TREE
S t a t e - o f - n a t u re - n od e 6
& 7 f avor a b l e m a r ke t
p rob a b ilit y = 6 0 %
S t a t e - o f - n a t u re - n od e 6
& 7 u n f avor a b l e m a r ke t
p rob a b ilit y = 4 0 %
Therefore:
As the EMV of not conducting survey is 52,000 vs an EMV of 49,200 for
conducting the study –the best choice is to NOT SEEK MARKETING
information. Getz should built the SMALL PLANT.
Getz estimates that if he conducts a market survey, there is really only a 35%
chance the results will indicate a favorable market for the sheds. How does the
tree change?
(Answer: the EMV of conducting a market survey = 38,800 so Getz should = (.35)(106,400) + (.65)(2,400)
still not do it.)
THANK YOU