Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 1 of 36
z
Flexibility Adds Value
– Enables adaptation to actual future conditions z
Flexibility Costs
– Money to create flexibility
– Complexity of design and management
– Time and Effort to create flexible plans and designs z
Thus, the Central Design issue is: How much
Flexibility should we incorporate into system?
z
The analytic question is: How do value flexibility?
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 2 of 36

z
What is an Option?
– LO: Concept and usefulness to designer z z z z z
Lattice Model
– LO: Analysis of time evolution of uncertainty
Option Valuation by Lattice (decision analysis)
– LO: How valuation proceeds through a lattice
Dynamic Programming
– LO: Use of this optimization analysis
Arbitrage Pricing of Options
– LO: Significance and Applicability of Concept
Black-Scholes Analysis
– LO: Use in Financial Markets
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 3 of 36
z z z z z
Merck and Kodak
– LO: Choice of Decision or ‘Options’ Analysis
Antamina
– LO: Use of Simulation for Valuation
Research Development at Ford
– LO: Hybrid Decision and ‘Options’ Analysis
Complex Engineering Systems
– LO: Screening Models to Identify Good Options
Hydropower Development
– LO: Path Dependency Issues
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 4 of 36

z z z z z z
To define technical concept of “Options”
– 4 step ‘Mantra’
To identify types of Options: ‘calls’ and ‘puts’
– Build upon their use in financial context
To present non-linear asymmetric returns
– Use of diagrams
To indicate drivers of value for options
– Uncertainty, Time increase value for American options
To distinguish range of applications
– Financial, “On” and “In” systems
Examples
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 5 of 36
z
An “Option” provides a formal way of defining flexibility z
An “Option” has a specific technical definition z
Semantic Caution:
– The technical meaning of an “option” is… much more specific and limited than … ordinary meaning of “option” in conversation... where “option” = “alternative” z
Pay careful attention to definition that follows!
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 6 of 36

An Option is...
z
A right, but not an obligation…
– Asymmetric returns; exercise only if advantageous
– Acquired at some cost z to take some action…
– to grow or change system, buy or sell something, etc, etc, z now, or in the future...
– May be indefinite
– But can be for a limited time after which option expires z for a pre-determined price (the “strike” price).
– Cost of action separate from cost of option
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 7 of 36
Spare tire on a car is an “option” because it provides z right, but not an obligation…
– Operator can use it or not
– Acquired at some cost – price of tire, loss of storage space z to take some action…
– to change the tire z now, or in the future...
– In this case, whenever desired z for a pre-determined price (the “strike” price).
– Time and effort of jacking up car, replacing tire, etc.
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 8 of 36

z
Two basic types of options
– CALL : right to take advantage of an opportunity
(such as ability to expand garage if demand is high)
– PUT : right to limit losses of a bad situation
(which is what an insurance policy provides) z
Options in design can be complicated
– NESTED, one following another (e.g.: research is option on development, which is option on production)
– SIMULTANEOUS (e.g.: research on fuel cells provides options on both hybrid cars and residential uses)
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 9 of 36
z
Focus first on financial options because
– Consequences easiest to present
– This is where options and valuation developed z
Financial Options
– Are tradable assets (see quotes finance.yahoo.com)
– Sold through exchanges similar to stock markets
– Are on all kinds of goods (known as “underlying assets”)
• Stocks
• Commodities (oil, meat, cotton, electricity...)
• Foreign exchange, etc., etc.
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 10 of 36

z
Two basic types of options
– Call: right to BUY asset for a set price
– Put: right to SELL asset for a set price z
The set price is known as the “strike” price z
Financial options can get very complicated
– In addition to Nested and Simultaneous
– Very exotic possibilities (“Asian”, “Bermudan”,
“caput”, “knock-out,” -- see www.enlexica.com)
– Not in this course !
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 11 of 36
z z
S = fluctuating market price of “underlying asset”
S* = S at the moment owner of option takes advantage of right (“when option is exercised”) z
K = Strike price z z
PAYOFF = owner’s net from a having the option, this is consequence we need to understand
Let’s examine what payoff looks like…
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 12 of 36

z
Call Option gives person right to buy an asset for a predetermined “strike” price, K z
Only rational to exercise right when price of asset is greater than “strike” price: S > K z
If exercised, option owner pays agreed price of K to get asset worth S* => Payoff = S* - K z
If unexercised, Payoff = 0 z
Formally, payoff = Maximum of either 0 or S* − K
= Max [0, S* − K]
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 13 of 36
Payoff
($)
Price of
Asset
S
S - K
0
0 K
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Payoff of
Option
Asset Price ($)
Richard de Neufville 
Option Concepts Slide 14 of 36

z
Example: A Call Option to
– buy 100 shares of AT&T (shares of AT&T constitute the
“underlying asset”)
– at $20 per share (this is the “strike price”)
– through Jan. 20, 2006 z
On Oct. 27, 2004, prices were (finance.yahoo.com)
– 1 share of AT&T = $ 16.42
– option to buy 1 share = $ 0.50
z
Here’s what the payoff diagram looks like…
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 15 of 36
Payoff
Per share
($)
Current
AT&T
Price
0
0
Price of
AT&T
S
S - K
Payoff of
Option
K= $20 Price of AT&T share ($)
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 16 of 36

z
Example: A PUT Option to
– SELL 100 shares of AT&T
– at $20 per share z
Only rational to exercise right when price of asset is LOWER than “strike” price: S < K z
If exercised, option owner GETS agreed price of K for asset worth S* => Payoff = K - S* z
If unexercised, Payoff = 0 z
Net payoff for put = Max [0, K - S*]
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 17 of 36
Payoff
($)
Asset
Price
S
K K-S
0
0
K
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Payoff of
Option
Asset Price ($)
Richard de Neufville 
Option Concepts Slide 18 of 36

z
For Call Option, if asset price > strike price… owner then makes profit – that could be unlimited z
Owner not required to exercise option, so…
Loss limited to cost of buying option
( $0.50/share in AT&T example) z
Value of option not symmetric
– For owner of call option: All gain, No pain
– (For seller of call option: all pain, no gain) z
Asymmetry is key to option value
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 19 of 36
z
How much should you pay to acquire an option?
z
Payoff diagrams show for a given strike price
–
–
Call payoff increases with asset price
Put payoff decreases with asset price z
Payoff generally does not reflect full value of option z
Why is that?
–
–
–
Owner exercises only when advantageous
In general, owner can wait for a higher price
Value is Max[ immediate exercise, waiting]
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 20 of 36

Some logical boundaries on value of call option that can be exercised any time (American)
Payoff
($)
S Price > 0
Otherwise buy option immediately
Upper Bound:
Call Value Equals
Asset Price
Price < S
Option yields S*- K
Option value < S
Lower Bound:
Call Value Equals
Payoff
Price > S - K
Or buy and exercise immediately
K Stock Price ($)
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 21 of 36
z
Consider an option whose current asset price equals strike price: (S = K) (it is “at the money”)
Immediate exercise payoff is zero
However, if you wait :
Payoff might be higher
Worst is zero payoff Payoff ($)
EV[S]
S - K
Value of waiting not reflected in immediate exercise, to be added
= value of option
0
0 K Asset Price ($)
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 22 of 36

z
Value exceeds immediate exercise payoff z z
Asymptotically nears immediate payoff for increased S
If no upside: value (expectations) = 0 = value (option)
Payoff($)
S - K
Value
0
0
K
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Asset Price ($)
Richard de Neufville 
Option Concepts Slide 23 of 36
z
Two “at the money” options (S=K) on different assets
Both have immediate payoff = 0
Asset A with greater volatility has higher P(larger net payoffs) => higher expected value
Asymmetric value favors high variation (limited losses)
Payoff ($) Asset A Payoff ($) pdf
Asset B
S-K S-K pdf
0
0 K Asset Price ($)
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
0
0 K Asset Price ($)
Richard de Neufville 
Option Concepts Slide 24 of 36

z
Increasing time to expiration increases value of options you can exercise any time (“American”)
–
–
Ability to wait allows option owner to benefit from asymmetric returns
Longer- term contains shorter-term options + more time cannot be worse, can only be better z
Compare a 3 and 6 month American call
–
–
–
Can exercise 6 month call at same time as 3 month
Can wait longer with 6 month
Which is more valuable? Must be longer one...
z
Time impact not obvious for options with fixed date for exercise (“European”)
–
Could miss out on profitable opportunities
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 25 of 36
z z
For a set strike price, value increases with
– Stock price increases
– Volatility
– Time
Payoff
($)
Increased strike price
– Reduces likelihood of payoffs
– Reduces call option value
Value increases with volatility and time to expiration
S - K
Asset Price ($)
0 K
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 26 of 36

z
“Real” because they refer to project and systems
– Contrast with financial options that are contracts z
Real Options are focus of interest for Design
– They provide flexibility for evolution of system z
Projects often contain option-like flexibilities
– Rights, not obligations (example: to expand garage)
– Exercise only if advantageous z
These flexibilities are “real” options z
Let’s develop idea by looking at possibilities…
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 27 of 36
z
Examples: z
Lease equipment with option to buy at end of lease
– Action is to buy at end of lease (or to walk away)
–
Lease period defined up-front (typically 2-3 years)
–
Purchase price defined in lease contract z
Flexible manufacturing processes
–
–
–
–
Ability to select mode of operation (e.g. thermal power by burning either gas or oil)
Switching between modes is action
Continuous opportunity (can switch at any time)
Switching often entails some cost (e.g.: set-up time)
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 28 of 36

z
Call-like
–
–
–
Capture benefits from increases in project value
Exercise typically involves putting money into project
Exercise when expectations of positive return increase z
Put-like
–
–
–
Insure against losses from decreased project value
Exercise may involve short-term costs or salvage value
Exercise when expectations of losses z
Compound (nested)
–
–
Projects might contain multiple options
Exercise decisions based on overall profit maximization z
In detail…
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 29 of 36
z
Waiting to Invest
–
–
–
–
A project might be profitable today, but better tomorrow
Leaving investment opportunity open ~ holding a call
Deciding factors: uncertainty resolution; foregone profits
Choice based on: Max [immediate investment, waiting, 0] z
Expand -- Accelerate effort or level of involvement
–
–
–
Allows greater participation in upside
Cost of expansion is like strike price
Choice based on: Max [status quo, expanded project z
Restart Temporarily Closed Operations
–
–
Similar to waiting to invest or expand (a special case)
Choice based on: Max [remain closed, re-open]
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 30 of 36

z
Abandon
–
–
–
Ability to halt investment eliminates further losses might include shut-down costs and salvage values
Choice based on: Max [continuing, abandoning] z
Contract -- Decelerate or narrow involvement
–
–
–
Reduces participation level and exposure to losses
Often incurs short-term scale down costs
Choice based on: Max [status quo, contracted] z
Temporarily Shut Down Operations
–
–
–
A special case of contraction
Eliminates losses, but can incur shut-down costs
Choice on: Max [status quo, temporarily shut-down ]
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 31 of 36
z
Combinations of Options
–
–
–
–
Many real options exist simultaneously
E.g.: those to abandon, contract, or temporarily shut down
Complex problem: value of multiple options are often interdependent and in general not additive
Exercise may make others valueless (abandon ends project) z
Switching Between Modes of Operation (example: dual fuel burner case)
–
–
–
Flexible systems contain an infinite series of options
Allow continual switching between modes of operation
If switching modes has a cost, it acts like a strike price z
For compound options, must value as system
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 32 of 36

z
Those “real” because, in contrast to financial options, they concern projects, they are “ON” projects
– E.g.: the option to open a mine (Antamina case)
– These do not concern themselves with system design
– Most common in literature z
Those “real” because they concern the design elements of system, they are “IN” projects
– EX: options for staging of system of communication satellites
– These require detailed understanding of system
– Most interesting to system designers
Financial options
Options ON projects
Options IN projects
These need knowledge of system
Real Options
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 33 of 36
z
These are financial options, but on technical things z
They treat technology as a “black box” z
Example: Antamina mine (detailed discussion later)
– option to open the mine after a two-year exploration period
– Uncertainty concerns: amount of ore and future price
=>uncertainty in revenue and thus in value of mine
– Option is a Financial Call Option (on Mine as asset) z
Differs from normal Financial Option because
– Much longer period -- financial option usually < 2 years
– Special effort needed to model future value of asset, it can’t be projected simply from past data (as otherwise typical)
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 34 of 36

z
These create options by design of technical system z
They require understanding of technology z
Example: Communications Satellites
– Designers can create options for expansion of capacity by way they configure original satellites
– Technical skill needed to create and exercise option z
Differ from other “real” Options because
– Special effort needed to model feasible flexibility within system itself (e.g.: modeling of ‘trade space” for design of communication satellites)
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 35 of 36
z z z z
Options embody formal concept of flexibility
Options are not “alternatives”
4 step Mantra “right, but not obligation, to act…”
“Calls” for opportunities, “puts” for downside risks z
Options have asymmetric returns => source of value z z z
Options are Financial and “Real”
Many “real” options available to system designers
“Real” Options “on” and “in” systems
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 36 of 36

z
Slides Summarizing lessons learned from
‘Garage’ and ‘Communications Satellite’ Cases z
Slides showing how flexibility affects valuation z
Slides illustrating impracticality of using decision analysis to analyze flexibility practice
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 37 of 36
z z
Typically focuses on design to specifications
Example: communications satellite system
– The System was designed to meet some specified level of performance (or “specs”)
– These specs decided outside the engineering practice
(for example, by market and or financial analyses)
– Note that design is a complex optimization effort
Cost
Best design to specification
Capacity
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 38 of 36

z z z
Why is this?
Because technological, market and other conditions uncertain
Example: communications satellite system:
– Profitability depends on market size for system
Profit
Loss
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Design spec
Market demand
Richard de Neufville 
Option Concepts Slide 39 of 36
z z z
Outcomes vary in probability
Consequences of outcomes x probability => pdf
(probability distribution function)
Example: communications satellite system:
Probability distribution
Loss
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Profit
Richard de Neufville 
Option Concepts Slide 40 of 36

z z
To change the distribution of probability distribution to increase, maximize value
Key means of doing this: flexibility that permits adaptation of design to circumstances
Probability distribution
Shift
Loss
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Profit
Richard de Neufville 
Option Concepts Slide 41 of 36
z
Accentuate the positive -- take advantage of opportunities (also known as “call options”)
Original pdf
New pdf
Loss
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Profit
Richard de Neufville 
Option Concepts Slide 42 of 36

z
Minimize the negative -- avoid big losses (as with insurance) (also known as “put” options)
New pdf
Original pdf
Loss
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Profit
Richard de Neufville 
Option Concepts Slide 43 of 36
z z z
Represents a real change in concept of design and management of engineering systems over time
Why is this?
Because: instead of designing to a spec, we design for a range of possible levels of performance
Cost
Design to specification
Design for expansion
Performance
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 44 of 36

Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 45 of 36
z
Case of Flexible Burner on Power Plant z
Turbines for electric power generation can be powered by
– Gas burners
– Oil burner
– Flexible, dual use burner (accepts either oil or gas) z
Fixed technologies (gas or oil) cost less to acquire than more complex flexible burner z
Under what conditions might flexible systems be valuable?
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 46 of 36

z
Based on Kulatilaka and Marcus paper z z
Price of gas: fixed at $1 per energy unit
Price of oil: increases over time
– In Year 1 oil costs $0.75 per energy unit
– Price increases 5% per year z z z z
Discount cash flows at 10%
Installation occurs in Year 0
Operations start in Year 1
Revenues are independent of technology z
What is the NPV for each burner?
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 47 of 36
z
Oil burner cheaper to operate until Year 6
Oil and Gas Prices
Price
$1.25
$1.15
$1.05
$0.95
$0.85
$0.75
0
Gas
5
Year
Oil
10
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 48 of 36

Year
Gas Plant
Revenue
Cost
PV Net
Cash Flow
NPV
Oil Plant
Revenue
Cost
PV Net
Cash Flow
NPV
Flexible
Plant
Revenue
Cost
PV Net
Cash Flow
NPV
0 1 2 3 4 5 6 7 8 9 10
1.16
1.21
1.27
1.34
1.40
1.47
1.55
1.63
1.71 1.79
2.50
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00 1.00
-2.50
0.15
0.17
0.20
0.23
0.25
0.27
0.28
0.29
0.30 0.30
1.16
1.21
1.27
1.34
1.40
1.47
1.55
1.63
1.71 1.79
2.50
0.79
0.83
0.87
0.91
0.96
1.01
1.06
1.11
1.16 1.22
-2.50
0.34
0.32
0.30
0.29
0.27
0.26
0.25
0.24
0.23 0.22
3.00
1.16
1.21
1.27
1.34
1.40
1.47
1.55
1.63
1.71 1.79
0.79
0.83
0.87
0.91
0.96
1.00
1.00
1.00
1.00 1.00
-3.00
0.34
0.32
0.30
0.29
0.27
0.27
0.28
0.29
0.30 0.30
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 49 of 36
z
Ranking of technologies
– Oil -- Flexible -- Gas z
Oil burner captures early cost advantages over gas
– Time value of money means early gains more significant than later losses z
Oil burner also better than flexible
– Both capture cost advantages early-on
– Flexible advantageously switches to gas in Year
– Extra cost of acquiring flexible > later gains z
Critical assumption: input prices are predictable
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 50 of 36

z
What if oil could follow one of three price paths?
Oil Prices
$1.80
$1.40
Price
$1.00
$0.60
$0.20
0
High
Medium
Low
5
Year
10
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 51 of 36
Year
Oil Plant
Revenue
Cost (High) p=0.3
Cost (Medium) p=0.4
Cost (Low) p=0.3
Cost (Avg.)
PV Net Cash
Flow
NPV
Flexible Plant
Revenue
Cost (High) p=0.3
Cost (Medium) p=0.4
Cost (Low) p=0.3
Cost (Avg.)
PV Net Cash
Flow
NPV
0
2.50
2.50
2.50
2.50
-2.50
3.00
3.00
3.00
3.00
-3.00
0.63
.
1.16
1.00
0.79
0.39
0.73
0.39
1
1.16
1.18
0.79
0.39
0.79
0.34
2
1.21
1.24
0.83
0.41
0.83
0.32
3
1.27
1.30
0.87
0.43
0.87
0.30
4
1.34
1.37
0.91
0.45
0.91
0.29
1.21
1.00
0.83
0.41
0.75
0.38
1.27
1.00
0.87
0.43
0.78
0.37
1.34
1.00
0.91
0.45
0.80
0.37
5
1.40
1.43
0.96
0.47
0.96
0.28
6
1.47
1.51
1.01
0.50
1.00
0.26
7
1.55
1.58
1.06
0.52
1.05
0.25
8
1.63
1.66
1.11
0.55
1.11
0.24
9
1.71
1.74
1.17
0.58
1.16
0.23
10
1.79
1.83
1.23
0.61
1.22
0.22
1.40
1.00
0.96
0.47
0.83
0.36
1.47
1.00
0.50
0.85
0.35
1.55
1.00
1.00
0.52
0.86
0.36
1.63
1.00
1.00
0.55
0.86
0.36
1.71
1.00
1.00
0.58
0.87
0.36
1.79
1.00
1.00
0.61
0.88
0.35
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 52 of 36

z
Rank of technologies
– Flexible -- Oil -- Gas z
Flexible technology enabled advantageous switching
– For high oil price: dual technology better than oil only
– For high gas price: dual better than gas alone
– Benefits accrue early on when uncertainty in prices is considered
– Operating cost savings outweigh additional acquisition costs
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 53 of 36
z z z z z
Real Situation can be VERY complicated
Prices Change Rapidly
– They may go up and down in many pathways
– The switch between fuels can be exercised often
Decision Analysis can be Impractical
– Simple Example Situation Already Difficult
Analysis assumed fixed Discount Rate, But
– Discount rate should reflect volatility of prices
– As risk changes, so should discount rate
– Cannot change discount rate in decision analysis
Another Method Required!!! => Options Analysis
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 54 of 36

Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 55 of 36
z
Start R&D project for $100,000 (0.1M) z
$1,100,000 (1.1 M) more to complete development
–
–
Commercial feasibility determined by initial R&D results
Plan to sell (license) technology to highest bidder z
Revenue estimate
–
–
50% chance to sell technology for $2000,000 (2M)
50% chance to sell for $100,000 (0.1M) z
Assume constant 10% discount rate applies z
Fund project?
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 56 of 36

1 2 Year 0
Initial Cost (0.1)
Development
License
Revenues
Present
Value
(0.1)
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
(1.1)
(1)
0.5*2
0.5*0.1
0.868
Richard de Neufville 
Option Concepts Slide 57 of 36
z z
NPV = - 232
Project should be rejected
Fund (0.1)
Do Not Fund
Com Good (1.1/1.1)
0.5
Com Bad (1.1/1.1)
0.5
2/1.1^2
0.1/1.1^2
0
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 58 of 36

z
Develop only if $2000 license is expected
Year 0 1 2
Initial Cost (0.1)
Development
License
Revenues
Present
Value
(0.1)
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
0.5*(1.1)
(0.5)
0.5*2
0.5*0
0.826
Richard de Neufville 
Option Concepts Slide 59 of 36
z z
NPV = + 226
Should accept project
Fund
(0.1)
Do Not Fund
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Good
0.5
Bad
0.5
Commit
(1.1/1.1)
Abandon
Com (1.1/1.1)
Abandon
2/1.1^2
0
0
0
0.1/1.1^2
Richard de Neufville 
Option Concepts Slide 60 of 36

z
Ability to abandon project has significant value
–
–
Limits downside
Continue only if advantageous z
Standard NPV misses option value completely
– Fails to consider effect of intelligent management decisions z
Standard NPV distorts value when there is risk
– Assumes that: NPV with expected values = expected NPV
– “Flaw of the Averages” see article, also Exercise 2
– However: Consequences of scenarios have asymmetries
– Example, production costs often not linear with volume z
Decision analysis has the advantage of recognizing value of flexibility
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 61 of 36
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 62 of 36

z
Analysis may be too complicated
– Situation may change too often so that analysis too confused
– Example: Prices for Basic Resources fluctuate rapidly up and down z
Inadequate basis for Choosing discount rate
– When nature of risk constantly changing,
– … discount rate should also be changing
– … No single discount rate adequately covers situation
– See Presentation on Valuation for details
Engineering Systems Analysis for Design
Massachusetts Institute of Technology
Richard de Neufville 
Option Concepts Slide 63 of 36