L1B Garage Case As Example of F in Design 2012-3

Garage Case
Dr. Richard de Neufville

Professor of Engineering Systems and
Civil and Environmental Engineering
Massachusetts Institute of Technology
Garage Case Concepts / RdN
Background!
Garage case is a prototype example.
It is a model for application portfolio.
The technology is simple enough so we can
see through it and gain insights into the how
flexibility in design increases expected value.
Case demonstrates several ways that
Flexibility improves design
Outline of presentation
1: Motivation: examples of real cases
=> HCSC building, Tufts Dental School, etc
2: The Case itself

=> Logic of Analysis; description of case
=> Analysis Structure and Results
3: Interpretation of Results
=> Better quantitatively and qualitatively
=> win-win solution
4: Decision rules
5: Implementation Issues
1. Motivation
The Garage Case is representative of a
broad class of infrastructure designs
1: Health Care Services Corporation Bldg.
in Chicago
2: Tufts Dental School in downtown Boston
3: Major Bridges: George Washington, NYC
Ponte 25 de Abril, Lisbon, Portugal
4: Parking Structure, Blue Water Mall, UK
References on web site
Vertical Phasing as a Corporate Real

Estate Strategy and Design Option,
A. Guma, J. Pearson, K. Wittels, R. de
Neufville, and D. Geltner, Journal of
Corporate Real Estate, 11(3), 144-157, 2009.
A. Guma thesis
J. Pearson and K. Wittels thesis
http://ardent.mit.edu/real_options/Common_course_materials/papers.html
Images: www.chicagoarchitectureinto/Building/1235/Blue-Cross-BlueShield-Of-Illinois-Tower.php
HCSC Building
(aka Blue Cross Blue Shield Tower)
Original Building erected in 1997

with 30 stories
27 story expansion for 2010
Original design had to have
Strength to carry double load

Empty space for possible future elevators
Planning permission from City, etc.
2.3 Million Sq.ft., 2nd largest in Chicago

after Sears (now Willis) tower
Before and after picture
Source: Goettsch Partners, 2008 and Pearson and Wittels, 2008.!

Tower under construction

and finished
Source: Wayne Lorentz/Artefaqs Corp.
Tufts Dental School
10 story building completed in 1973

But School knew it would want more and
built in structural strength
5 story addition ready end of 2009
Construction on top of building in middle
of city obviously problematic
In downtown Boston theater district
Pics: dental.tufts.edu/1176818445781/TUSDM-Page-dental2w_
119271296754.html
Story: www.healthcarebuildingideas.com/article/going-vertical
Tufts dental under construction
Source: nerej.com
Tufts dental school doubled
Source: bostonrealestateobserver.com
Ponte 25 de Abril, Lisbon
Built under dictatorship, 1966
With strength for second deck, allowances

for possible rail service (pre-built station)
Situation 30 years later very different
Portugal is in European Union

Receives money from community,
especially for Metro
New Conditions lead to new solutions,
within flexible framework
25 de Abril Bridge, Lisbon
Sources:
Estudio Mario
Novais
Biblioteca de
Arte-Fundao
Calouste
Gulbenkian
Photo c 1966
25 de Abril Bridge in 2009
de Neufville photo!
Blue Water Shopping Centre
Blue Water one of biggest in UK,
Southeast of London, in Kent
Parking facility made of precast

reinforced concrete
Built for possible addition of several

extra levels
Could be expanded rapidly using

precast columns, panels
Aeriel view of site
Source: webaviation.co.uk
Take-away from these cases
Vertical phasing of major infrastructure

is a serious business
Practical, not academic idea
Therefore
Although garage case is simple, for

clarity of presentation.
It is definitely not simple-minded!
2. Garage Case!
Reference on web:
Valuing Options by Spreadsheet: Parking
Garage Case Example,
R. de Neufville, S. Scholtes, and T. Wang,
American Society of Civil Engineers,
Journal of Infrastructure Systems, Vol.12,
no. 2, pp. 107-111, 2006
http://ardent.mit.edu/real_options/Common_course_materials/
papers.html
Logic of Analysis
Consider engineering base case design
for fixed objective (mission or specifications)
Recognize reality of uncertainty

different
values, due to system non-linearities

Different designs are also possible, likely
Incorporate flexibility
A
design with high expected value

Avoids downside losses, takes opportunities
Win-win solutions
Parking Garage Case

Major garage serving mega-mall
Actual demand necessarily uncertain:
Population
growth, demographics speculative

Mall success chancy (in case of Blue Water, a
major competitor opened up nearby)
Competition from other parking facilities
Engineering design assumes a fixed

forecast
Parking Garage Case details

Demand
At
start is for 750 spaces

Over next 10 years is expected to rise
exponentially by another 750 spaces
After year 10 may be 250 more spaces
Annual revenue/space used = $10,000

The discount rate is taken to be 12%
Parking Garage details (Cont)

Costs
annual operating costs (staff, cleaning, etc.)
= $2,000 /year/space available
(note: spaces used < spaces available)
Annual lease of the land = $3.6 Million
construction cost = $16,000/space + 10%
for each level above the ground level
Can accommodate 200 cars per level

Engineering Base Case

Demand growth as predicted, no variability
Year
Demand
Capacity
Revenue
Recurring Costs
Operating cost
Land leasing cost
Cash flow
Discounted Cash Flow
Present value of cash flow
Capacity costs for up to two levels
Capacity costs for levels above 2
Net present value
$3,600,000
1
2
3
750
893
1,015
1,200
1,200
1,200
$7,500,000 $8,930,000 $10,150,000
$2,400,000
$3,600,000
$1,500,000
$1,339,286
$2,400,000
$3,600,000
$2,930,000
$2,335,778
$2,400,000
$3,600,000
$4,150,000
$2,953,888
19
20
1,688
1,696
1,200
1,200
$12,000,000 $12,000,000
$2,400,000
$3,600,000
$6,000,000
$696,641
$2,400,000
$3,600,000
$6,000,000
$622,001
$32,574,736
$6,400,000
$16,336,320
$6,238,416
Optimizing the base case

Optimization = find highest value for
all designs
Solution:
consider each major design

alternative
In
this case, number of floors

Typically, a sweet spot
What does graph of floors vs. value

look like?
Optimal design for base case

(no uncertainty) is 6 floors
EXPECTED NPV ($, MILLIONS)
10
0
2
-5
-10
-15
NUMBER OF LEVELS
TRADITIONAL NPV
Recognizing Uncertainty
Many things uncertain
Costs
may be easier to estimate,

contractors may give fixed bids
Focus on Demand, more uncertain,

farther in future
Assume:
could be 50% off the projections,

either way;
Annual volatility for growth is 10%
Distribution of Outcomes
Recognizing Uncertainty => implies
many possible future scenarios
We calculate possible value of
system for each possible scenario
We obtain a distribution of outcomes
(as indicated in ESD70)
Also Cumulative distributions or
target curves
Target Curves
Represent cumulative chance of getting
a result below any specific level
Going
from below lowest value (no chance)

To at or below the highest value (100% chance)
Allows read on Value at Risk (VAR) :

Definition:
VAR is A loss that will not be

exceeded at some specified confidence level
We are p percent certain that we will not lose
more than V dollars for this project.
Target curves graphically

Cumulative Probability
CDF
100%
80%
60%
40%
20%
0%
-400
-200
200
NPVA
NPVB NPV
90%VAR for NPVB
10% Probability
400
600
90% VAR for NPVA
Look at distribution of NPV of designs A, B:

90% VAR for NPVA -$91; for NPVB, $102
A few notes on VARG

VAR is a common financial concept
It stresses downside losses, risks
However, designers also need to look
at upside potential: Value of Gain
So we expand VAR to VARG
Value
at Risk and Gain
Effect of uncertainty on analysis

Changes results Why?
Non-linearities
in model
Lowers results in this case -- Why?

Capacity
constraints systematically limit

ability to profit from good opportunities
Changes design Why?

Above
encourages lower investment

Simulated results uncertainty

Why is Right-hand side gone?
Lower demand => Loss ;
Higher demand => Gain limited by garage size
600
5-floor design
Frequency
500
Simulated Mean
400
6-floor design
300
Deterministic
Result
200
100
0
-17.8 -15.6 -13.5 -11.3
-9.2
-7.0
-4.9
-2.7
-0.6
1.6
3.7
5.9
8.0
NPV Target Curve (CDF)

Compare Actual (5 Fl) with unrealistic fixed 6 Fl design
1
0.9
Probability
0.8
0.7
0.6
CDF for Result of
0.5
Simulation Analysis (5-
0.4
floor)
Implied CDF for
0.3
Result of
0.2
Deterministic NPV
Analysis (6-floor)
0.1
0
-20
-15
-10
-5
10
Recognizing uncertainty =>

different design: 5 floors
EXPECTED NPV ($, MILLIONS)
10
0
2
-5
-10
-15
NUMBER OF LEVELS
TRADITIONAL NPV
RECOGNIZING UNCERTAINTY
Recognizing uncertainty =>

different value!
Value calculated for deterministic case
= $6.2 million
Value with uncertainty much different,
= $3.5 million
deterministic value is FALSE
Conclusion: deterministic analysis has
given both wrong design and value!
Introduce flexibility into design

How can we make garage design flexible?
What is flexibility?
Ability
to adjust project to needs, opportunities

According to how future develops
We exercise intelligent management of
development of system over time
How do we do this here?

Make
it possible to add more levels as needed

Stronger columns, foundations
More cost than for inflexible garage with same
number of floors.
Result greater expected value

Year
Demand
Capacity
Decision on expansion
Extra capacity
Revenue
Recurring Costs
Operating cost
Land leasing cost
Expansion cost
Cash flow
Discounted Cash Flow
Present value of cash flow
Capacity cost for up to two levels
Capacity costs for levels above 2
Price for the option
Net present value
1
820
800
$8,000,000
2
3
924
1,044
800
1,200
expand
400
$8,000,000 $10,440,000
$1,600,000
$3,600,000 $3,600,000
$2,400,000
$3,600,000
$3,200,000
$3,600,000
$3,200,000
$3,600,000
$4,440,000
$3,160,304
$8,390,000
$974,136
$9,200,000
$953,734
$1,600,000
$3,600,000
$8,944,320
$2,800,000 -$6,144,320
$2,500,000 -$4,898,214
19
1,519
1,600
20
1,647
1,600
$15,190,000 $16,000,000
$30,270,287
$6,400,000
$7,392,000
$689,600
$12,878,287
Including Flexibility => Another, better design:

4 Fl with stronger structure enabling expansion
Summary of design results

from different perspectives
Note: good result
from deterministic
case is a sham
Perspective
Deterministic
Recognizing Uncertainty
Simulation
No
Yes
Option Embedded
No
No
Incorporating Flexibilty
Yes
Yes
Design
6 levels
5 levels
4 levels with strengthened
structure
Estimated Expected NPV

$6,238,416
$3,536,474
$10,517,140
Why is the optimal design much better when

we design with flexibility?
Multi-dimensional valuation
For uncertainty, 1 dimension is not enough
Design
Initial Investment
Expected NPV
Minimum Value
Maximum Value
Design with Flexibility Thinking Design without Flexibility thinking

(4 levels, strengthened structure)
(5 levels)
$18,081,600
$21,651,200
$10,517,140
$3,536,474
-$13,138,168
-$18,024,062
$29,790,838
$8,316,602
Comparison
Better with options
Better with options
Better with options
Better with options
Everything is better! How did it happen?

Root cause: change the framing of design problem
From: focus on a (mythical) forecast or set of specs
To: managing (realistic) uncertainties by flexibility
3. Interpretation of results
Why do Flexible designs
systematically provide better value?
What is the insight?
Does Flexibility cost?
Why it can be win-win
Sources of value for flexibility

Cut downside ; Expand Upside
Avoid downside ; Profit from Upside
Cumulative Probability
Expand upside potential

Original
distribution
Distribution with
flexibility
Cut downside risks
Value

1) Minimize exposure to downside risk!
1
0.9
Probability
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
-20
-15
-10
-5
5-Floor Design
10
4-Floor Design

2) Maximize potential for upside gain!
100.0%
90.0%
Mean for NPV

without Flexibility
80.0%
Probability
70.0%
CDF for NPV

with Flexibility
60.0%
50.0%
40.0%
CDF for NPV

without Flexibility
30.0%
Mean for NPV

with Flexibility
20.0%
10.0%
0.0%
-20
-15
-10
-5
10
15
20
25
30
35
What is cost of flexibility?

Often said: Flexibility costs
In what sense is this true?
Clearly 4-story garage with strength to
add floors cost more than an inflexible
one that does not have this capability
Is flexible design more expensive?
What is fair comparison?

To be fair, compare relative cost of
design with flexibility with one without
So, if you had to make a 1-time
decision, what is best choice?
5 or 6 story garage!!
Fair comparison is between
flexible design 4-stories
Inflexible 5 or 6 stories
Flexible Design as Win-Win

Flexible design not only increases value,
it saves money 3 ways. You can:
1. Build smaller dont have to meet
future needs right at start
2. Defer costs, and thus decrease their
NEGATIVE Present Value
3. Avoid costs completely some
projected needs will not occur
4. Decision Rules
1: What is different about analysis
for garage case?
2: How we implement flexibility in
spreadsheet analysis
What is different?
Here is image of a spreadsheet for garage case:
How does it differ from standard format?

Figure D.17: Spreadsheet for Garage Case (Source: Flexibility in Design)
Difference subtle but crucial

Difference not immediately visible
Difference is in formulas behind
number:
Rules for exercising flexibility
The
way we implement flexibility in

spreadsheet
If statements that trigger actions based on
past situations
Implementation of a Rule
In cell E6 (which has expand in it:
IF(AND(D4<MAX_CAP,
MIN(D4,D5) +
MIN(E4,E5) = D5+E5), expand, )
Interpretation
IF(AND
=> two conditions to be respected

(D4<MAX_CAP => number of spaces in
previous year less than maximum capacity
MIN(D4,D5) + MIN(E4,E5) = D5+E5 =>
demand in past 2 years more than current
capacity of garage
Then expand, triggers expansion cost in E11,
for more capacity next year or
How many rules ?

What is limit on number of rules?
No
theoretical limit
Practical limit we and eventual audience
for analysis may get confused
Rules need to cover important possible

decisions
Expand
or not?
Close?
Raise
prices? Etc.
Can rules vary over time?

Often should, to reflect changing
situations
Examples for garage case?
Expansion
at end of project may not make

sense. Not enough time to make profits to
pay for costs.
May want to include consideration of
changing construction costs
Is it right rule?
What is criterion for right rule?
Two perspectives:
Descriptive
(you do): Rule reflecting managers

choices (might not be optimal!)
Normative (you should): Rule to maximize
Expected Net Present Value (ENPV)
Not obvious how to determine optimal

Can
try several possibilities

Reference
Cardin, M.-A., Facing Reality: Design and
Management of Flexible Engineering
Systems, Master of Science thesis, MIT
Technology and Policy Program, May,
2007.
http://ardent.mit.edu/real_options/Real_Opts_papers/
Cardin_SM_Thesis.pdf
5. Implementation
Caution: We need to make sure that
flexible designs can be implemented
when needed.
Example: Bluewater model for this case
What happened?
Ignorance
of new owners
Ready collaborators (designers, suppliers)
Compatible regulations they can change
Implementation plan needed

More on this later
Basics concept is that you have to
keep flexibility alive:
Knowledge
of Possibilities
Collaboration of stakeholders
Monitoring, anticipating regulatory changes
An implementation plan necessary

part of good design
Summary of Concepts
Flexibility Design is practical and used
by top professionals
It increases expected value by
Reducing downside losses

Increasing upside gains
Thus improving on many dimensions
It can be a win-win solution

It requires thoughtful implementation
Decision rules are secret sauce for
analysis of flexibility

L1B Garage Case As Example of F in Design 2012-3

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

L1B Garage Case As Example of F in Design 2012-3

Uploaded by

Copyright:

Available Formats

Garage Case

Dr. Richard de Neufville

Garage Case Concepts / RdN

=> HCSC building, Tufts Dental School, etc

2: The Case itself

References on web site

Vertical Phasing as a Corporate Real

J. Pearson and K. Wittels thesis

Original Building erected in 1997

27 story expansion for 2010

Original design had to have

Strength to carry double load

2.3 Million Sq.ft., 2nd largest in Chicago

Before and after picture

Source: Goettsch Partners, 2008 and Pearson and Wittels, 2008.!

Tower under construction

Source: Wayne Lorentz/Artefaqs Corp.

Garage Case Concepts / RdN

Tufts Dental School

10 story building completed in 1973

Tufts dental under construction

Garage Case Concepts / RdN

Tufts dental school doubled

Garage Case Concepts / RdN

Ponte 25 de Abril, Lisbon

Built under dictatorship, 1966

With strength for second deck, allowances

Situation 30 years later very different

Portugal is in European Union

25 de Abril Bridge, Lisbon

Garage Case Concepts / RdN

25 de Abril Bridge in 2009

Garage Case Concepts / RdN

Blue Water Shopping Centre

Blue Water one of biggest in UK,

Southeast of London, in Kent

Parking facility made of precast

Built for possible addition of several

Could be expanded rapidly using

Aeriel view of site

Garage Case Concepts / RdN

Take-away from these cases

Vertical phasing of major infrastructure

Although garage case is simple, for

Garage Case Concepts / RdN

for fixed objective (mission or specifications)

Recognize reality of uncertainty

values, due to system non-linearities

design with high expected value

Parking Garage Case

growth, demographics speculative

Engineering design assumes a fixed

Parking Garage Case details

start is for 750 spaces

Annual revenue/space used = $10,000

Parking Garage details (Cont)

Can accommodate 200 cars per level

Engineering Base Case

Garage Case Concepts / RdN

Optimizing the base case

consider each major design

this case, number of floors

What does graph of floors vs. value