Transcript Decision Analysis

IndE 311
Stochastic Models and Decision Analysis
UW Industrial Engineering
Instructor: Prof. Zelda Zabinsky
Decision Analysis-1
Operations Research
“The Science of Better”
Decision Analysis-2
Operations Research
Modeling Toolset
311
Queueing
Theory
Simulation
Inventory
Theory
Forecasting
310
Markov
Chains
PERT/
CPM
Decision
Analysis
Stochastic
Programming
Markov
Decision
Processes
Dynamic
Programming
Network
Programming
Linear
Programming
Integer
Programming
Nonlinear
Programming
Game
Theory
312
Decision Analysis-3
IndE 311
• Decision analysis
– Decision making without
experimentation
– Decision making with
experimentation
– Decision trees
– Utility theory
• Markov chains
–
–
–
–
–
–
Modeling
Chapman-Kolmogorov equations
Classification of states
Long-run properties
First passage times
Absorbing states
• Queueing theory
–
–
–
–
–
Basic structure and modeling
Exponential distribution
Birth-and-death processes
Models based on birth-and-death
Models with non-exponential
distributions
• Applications of queueing
theory
– Waiting cost functions
– Decision models
Decision Analysis-4
Decision Analysis
Chapter 15
Decision Analysis-5
Decision Analysis
• Decision making without experimentation
– Decision making criteria
• Decision making with experimentation
– Expected value of experimentation
– Decision trees
• Utility theory
Decision Analysis-6
Decision Making without
Experimentation
Decision Analysis-7
Goferbroke Example
•
•
•
•
•
Goferbroke Company owns a tract of land that may contain oil
Consulting geologist: “1 chance in 4 of oil”
Offer for purchase from another company: $90k
Can also hold the land and drill for oil with cost $100k
If oil, expected revenue $800k, if not, nothing
Payoff
Alternative
Oil
Dry
1 in 4
3 in 4
Drill for oil
Sell the land
Chance
Decision Analysis-8
Notation and Terminology
• Actions: {a1, a2, …}
– The set of actions the decision maker must choose from
– Example:
• States of nature: {1, 2, ...}
– Possible outcomes of the uncertain event.
– Example:
Decision Analysis-9
Notation and Terminology
• Payoff/Loss Function: L(ai, k)
– The payoff/loss incurred by taking action ai when state k occurs.
– Example:
• Prior distribution:
– Distribution representing the relative likelihood of the possible
states of nature.
• Prior probabilities: P( = k)
– Probabilities (provided by prior distribution) for various states of
nature.
– Example:
Decision Analysis-10
Decision Making Criteria
Can “optimize” the decision with respect to several criteria
• Maximin payoff
• Minimax regret
• Maximum likelihood
• Bayes’ decision rule (expected value)
Decision Analysis-11
Maximin Payoff Criterion
• For each action, find minimum payoff over all states of nature
• Then choose the action with the maximum of these minimum
payoffs
State of Nature
Action
Oil
Dry
Drill for oil
700
-100
Sell the land
90
90
Min
Payoff
Decision Analysis-12
Minimax Regret Criterion
• For each action, find maximum regret over all states of nature
• Then choose the action with the minimum of these maximum regrets
(Payoffs)
State of Nature
Action
Oil
Dry
Drill for oil
700
-100
Sell the land
90
90
(Regrets)
State of Nature
Action
Oil
Dry
Max
Regret
Drill for oil
Sell the land
Decision Analysis-13
Maximum Likelihood Criterion
• Identify the most likely state of nature
• Then choose the action with the maximum payoff under that state of
nature
State of Nature
Action
Oil
Dry
Drill for oil
700
-100
Sell the land
90
90
0.25
0.75
Prior probability
Decision Analysis-14
Bayes’ Decision Rule
(Expected Value Criterion)
• For each action, find expectation of payoff over all states of nature
• Then choose the action with the maximum of these expected
payoffs
State of Nature
Action
Oil
Dry
Drill for oil
700
-100
Sell the land
90
90
0.25
0.75
Prior probability
Expected
Payoff
Decision Analysis-15
Sensitivity Analysis with
Bayes’ Decision Rule
• What is the minimum probability of oil such that we choose to drill
the land under Bayes’ decision rule?
State of Nature
Action
Oil
Dry
Drill for oil
700
-100
Sell the land
90
90
p
1-p
Prior probability
Expected
Payoff
Decision Analysis-16
Decision Making with
Experimentation
Decision Analysis-17
Goferbroke Example (cont’d)
State of Nature
Action
Oil
Dry
Drill for oil
700
-100
Sell the land
90
90
0.25
0.75
Prior probability
• Option available to conduct a detailed seismic survey to obtain a
better estimate of oil probability
• Costs $30k
• Possible findings:
– Unfavorable seismic soundings (USS), oil is fairly unlikely
– Favorable seismic soundings (FSS), oil is fairly likely
Decision Analysis-18
Posterior Probabilities
• Do experiments to get better information and improve
estimates for the probabilities of states of nature. These
improved estimates are called posterior probabilities.
• Experimental Outcomes: {x1, x2, …}
Example:
• Cost of experiment: 
Example:
• Posterior Distribution: P( = k | X = xj)
Decision Analysis-19
Goferbroke Example (cont’d)
• Based on past experience:
If there is oil, then
– the probability that seismic survey findings is USS = 0.4 =
P(USS | oil)
– the probability that seismic survey findings is FSS = 0.6 =
P(FSS | oil)
If there is no oil, then
– the probability that seismic survey findings is USS = 0.8 =
P(USS | dry)
– the probability that seismic survey findings is FSS = 0.2 =
P(FSS | dry)
Decision Analysis-20
Bayes’ Theorem
• Calculate posterior probabilities using Bayes’ theorem:
Given P(X = xj |  = k), find P( = k | X = xj)
P(   k | X  x j ) 
P(X  x j |    k ) P(   k )
P(X  x


j
|   i ) P(  i )
i
Decision Analysis-21
Goferbroke Example (cont’d)
• We have
P(USS | oil) = 0.4
P(USS | dry) = 0.8
P(FSS | oil) = 0.6
P(FSS | dry) = 0.2
P(oil) = 0.25
P(dry) = 0.75
• P(oil | USS) =
• P(oil | FSS) =
• P(dry | USS) =
• P(dry | FSS) =
Decision Analysis-22
Goferbroke Example (cont’d)
Optimal policies
• If finding is USS:
State of Nature
Action
Oil
Dry
Drill for oil
700
-100
Sell the land
90
90
Expected
Payoff
Posterior probability
• If finding is FSS:
State of Nature
Action
Oil
Dry
Drill for oil
700
-100
Sell the land
90
90
Expected
Payoff
Posterior probability
Decision Analysis-23
The Value of Experimentation
• Do we need to perform the experiment?
As evidenced by the experimental data, the experimental outcome is
not always “correct”. We sometimes have imperfect information.
• 2 ways to access value of information
– Expected value of perfect information (EVPI)
What is the value of having a crystal ball that can identify true
state of nature?
– Expected value of experimentation (EVE)
Is the experiment worth the cost?
Decision Analysis-24
Expected Value of Perfect Information
• Suppose we know the true state of nature. Then we will
pick the optimal action given this true state of nature.
State of Nature
Action
Oil
Dry
Drill for oil
700
-100
Sell the land
90
90
0.25
0.75
Prior probability
• E[PI] = expected payoff with perfect information =
Decision Analysis-25
Expected Value of Perfect Information
• Expected Value of Perfect Information:
EVPI = E[PI] – E[OI]
where E[OI] is expected value with original information
(i.e. without experimentation)
• EVPI for the Goferbroke problem =
Decision Analysis-26
Expected Value of Experimentation
• We are interested in the value of the experiment. If the
value is greater than the cost, then it is worthwhile to do
the experiment.
• Expected Value of Experimentation:
EVE = E[EI] – E[OI]
where E[EI] is expected value with experimental
information.
E [EI ]   E [value | experiment al result j ] P (result j )
j
Decision Analysis-27
Goferbroke Example (cont’d)
• Expected Value of Experimentation:
EVE = E[EI] – E[OI]
E [EI ]   E [value | experiment al result j ] P (result j )
j
EVE =
Decision Analysis-28
Decision Trees
Decision Analysis-29
Decision Tree
• Tool to display decision problem and relevant
computations
• Nodes on a decision tree called __________.
• Arcs on a decision tree called ___________.
• Decision forks represented by a __________.
• Chance forks represented by a ___________.
• Outcome is determined by both ___________ and ____________.
Outcomes noted at the end of a path.
• Can also include payoff information on a decision tree branch
Decision Analysis-30
Goferbroke Example (cont’d)
Decision Tree
Decision Analysis-31
Analysis Using Decision Trees
1. Start at the right side of tree and move left a column at a time. For
each column, if chance fork, go to (2). If decision fork, go to (3).
2. At each chance fork, calculate its expected value. Record this
value in bold next to the fork. This value is also the expected value
for branch leading into that fork.
3. At each decision fork, compare expected value and choose
alternative of branch with best value. Record choice by putting
slash marks through each rejected branch.
• Comments:
–
–
This is a backward induction procedure.
For any decision tree, such a procedure always leads to an optimal
solution.
Decision Analysis-32
Goferbroke Example (cont’d)
Decision Tree Analysis
Decision Analysis-33
Painting problem
• Painting at an art gallery, you think is worth $12,000
• Dealer asks $10,000 if you buy today (Wednesday)
• You can buy or wait until tomorrow, if not sold by then, can be yours
for $8,000
• Tomorrow you can buy or wait until the next day: if not sold by then,
can be yours for $7,000
• In any day, the probability that the painting will be sold to someone
else is 50%
• What is the optimal policy?
Decision Analysis-34
Drawer problem
• Two drawers
– One drawer contains three gold coins,
– The other contains one gold and two silver.
• Choose one drawer
• You will be paid $500 for each gold coin and $100 for each silver
coin in that drawer
• Before choosing, you may pay me $200 and I will draw a randomly
selected coin, and tell you whether it’s gold or silver and which
drawer it comes from (e.g. “gold coin from drawer 1”)
• What is the optimal decision policy? EVPI? EVE? Should you pay
me $200?
Decision Analysis-35
Utility Theory
Decision Analysis-36
Validity of Monetary Value Assumption
• Thus far, when applying Bayes’ decision rule, we
assumed that expected monetary value is the
appropriate measure
• In many situations and many applications, this
assumption may be inappropriate
Decision Analysis-37
Choosing between ‘Lotteries’
• Assume you were given the option to choose from two
‘lotteries’
– Lottery 1
50:50 chance of winning $1,000 or $0
– Lottery 2
Receive $50 for certain
.5
.5
1
$1,000
$0
$50
• Which one would you pick?
Decision Analysis-38
Choosing between ‘lotteries’
• How about between these two?
– Lottery 1
50:50 chance of winning $1,000 or $0
– Lottery 2
Receive $400 for certain
.5
.5
1
• Or these two?
– Lottery 1
50:50 chance of winning $1,000 or $0
– Lottery 2
Receive $700 for certain
$0
$400
.5
.5
1
$1,000
$1,000
$0
$700
Decision Analysis-39
Utility
• Think of a capital investment firm deciding whether or
not to invest in a firm developing a technology that is
unproven but has high potential impact
• How many people buy insurance?
Is this monetarily sound according to Bayes’ rule?
• So... is Bayes’ rule invalidated?
No- because we can use it with the utility for money
when choosing between decisions
– We’ll focus on utility for money, but in general it could be utility
for anything (e.g. consequences of a doctor’s actions)
Decision Analysis-40
A Typical Utility Function for Money
u(M)
4
3
What does
this mean?
2
1
0
$100 $250
$500
$1,000
M
Decision Analysis-41
Decision Maker’s Preferences
• Risk-averse
u(M)
– Avoid risk
– Decreasing utility for money
• Risk-neutral
M
u(M)
– Monetary value = Utility
– Linear utility for money
• Risk-seeking (or risk-prone)
M
u(M)
– Seek risk
– Increasing utility for money
• Combination of these
M
u(M)
…
M
Decision Analysis-42
Constructing Utility Functions
• When utility theory is incorporated into a real decision
analysis problem, a utility function must be constructed
to fit the preferences and the values of the decision
maker(s) involved
• Fundamental property:
The decision maker is indifferent between two alternative
courses of action that have the same utility
Decision Analysis-43
Indifference in Utility
• Consider two lotteries
p
1-p
$1,000
1
$X
$0
• The example decision maker we discussed earlier
would be indifferent between the two lotteries if
– p is 0.25 and X is …
– p is 0.50 and X is …
– p is 0.75 and X is …
Decision Analysis-44
Goferbroke Example (with Utility)
• We need the utility values for the following possible
monetary payoffs:
Monetary
Payoff
45°
u(M)
Utility
-130
-100
60
90
670
M
700
Decision Analysis-45
Constructing Utility Functions
Goferbroke Example
• u(0) is usually set to 0. So u(0)=0
• We ask the decision maker what value of p makes
him/her indifferent between the following lotteries:
p
1-p
700
1
0
-130
• The decision maker’s response is p=0.2
• So…
Decision Analysis-46
Constructing Utility Functions
Goferbroke Example
• We now ask the decision maker what value of p makes
him/her indifferent between the following lotteries:
p
1-p
700
1
90
0
• The decision maker’s response is p=0.15
• So…
Decision Analysis-47
Constructing Utility Functions
Goferbroke Example
• We now ask the decision maker what value of p makes
him/her indifferent between the following lotteries:
p
1-p
700
1
60
0
• The decision maker’s response is p=0.1
• So…
Decision Analysis-48
Goferbroke Example (with Utility)
Decision Tree
Decision Analysis-49
Exponential Utility Functions
• One of the many mathematically prescribed forms of a “closed-form”
utility function
M
 

u(M )  R 1  e R 


• It is used for risk-averse decision makers only
• Can be used in cases where it is not feasible or desirable for the
decision maker to answer lottery questions for all possible outcomes
• The single parameter R is the one such that the decision maker is
indifferent between
0.5
0.5
R
and
1
0
(approximately)
-R/2
Decision Analysis-50