ESD.36 System Project Management
+
-
Project Risk Management
Instructor(s)
Prof. Olivier de Weck
Lecture 11
October 16, 2012
+
Outline
-

Layers of Risk and Classical Risk Management

Review: Project NPV, Value at Risk (VAR) Concept

Example: Garage Case
- ESD.36J SPM
+
-
Layers of Risk
Classical Risk Management
- ESD.36J SPM
3
+


Projects entail many types of risks
Projects bring together many risks
In energy (petroleum) field …..









Subsurface (e.g. Macondo well)
Fiscal
Market
Technology
Partner, contractor
Project execution
Operability
Project Manager authority and influence limited relative to
set of risks that matter
Nature, degree of manageability and, therefore, desired
mental model , differs by risk type, nature of outcome
- ESD.36J SPM
4
+
-
Layers of Risk
Natural Risks
Market Risks
• Geology
• Weather
Country/Fiscal
Industry/Competitive
Technical/
Project Risks
• Construction
• Operations
• Partner/ally
• Technical
• Project management
• Industry evolution
• Demand, growth rates
• Supply conditions
• Competition
• Infrastructure
High Influence
Context: Energy Industry
- ESD.36J SPM
• Commodity Prices
• Exchange rates
• Political stability • Interest rates
• Terrorism
• Risk premium
• Financial,
economic stability,
inflation
• Regulatory stability
or intervention
• Contract
enforcement
• Legal stability
Low Influence
Source: D. Lessard
5
+
-
Risk Categories
Technical
Risk
Market/Threat
Change
Programmatic
Risk
Cost Risk
Schedule Slips
- ESD.36J SPM
Schedule
Risk
6
+
-
1992
1996
1996
1996
1998
1998
1999
1999
2001
2003
2003
2003
2005
2007
2012
Mission Success and Failure
Mars Observer
US
Mars Global Surveyor US
Mars 96
USSR
Mars Pathfinder
US
Nozomi
Japan
Mars Climate Orbiter US
Mars Polar Lander
US
Deep Space 2 Probes US
Mars Odyssey
US
Mars Express Orbiter/ ESA
MER Spirit
US
MER- Opportunity
US
MRO
US
Phoenix Mars Lander US
Mars MSL Curiosity US
Failure
Lost prior to Mars arrival
Success More images than all Mars Missions prior
Failure
Launch vehicle failure
Success Technology experiment lasting 5 times plan
Failure
No orbit insertion; fuel problems
Failure
Lost on arrival
Failure
Lost on arrival
Failure
Lost on arrival (carried on Mars Polar Lander)
Success High resolution images of Mars
Success/Failure Orbiter imaging Mars and lander lost
Success Operating lifetime of more than 15 times
Success Operating lifetime of more than
Success Returned more than 26 terabits of
Sucesss Landed in North Polar Regions
Success Landed large RTG powered rover at Gale Crater
Despite recent string of successes less than 50% of pre2001 missions successful
- ESD.36J SPM
7
+
-
A Risk Management Framework
Correct
deviations
Anticipate
what can
go wrong
Control
Identify
Track
actions
Communicate
Track
Analyze
Decide
what is
important
Plan
Plan to take action
- ESD.36J SPM
8
+
Risk ID/Assessment
-
N
5
Reqmnts
4
Product
Cost
ID Risks and Score
3
1
2
2
Environment
Schedule



1
2
3
4
5
Probability that a particular event will occur
Impact or Consequence if the event does indeed occur
Aggregate Into Categories


3
Brainstorm Risks


1
Rule of Thumb Limit @ N20
Score (Based on Opinion & Data)
Involve All Stakeholders
- ESD.36J SPM
9
+
Quantitative Assessment
-
Risk = Probability * Consequence
PROBABILITY
CONSEQUENCES
CATASTROPHIC
FREQUENT
(HI GH)*
0 . 7 <P< 1 .0
PROBABLE
(M EDI UM )*
0 . 4 <P< 0 .7
I M PROBABLE
(LOW )*
I M POSSI BLE
0 <P<0 .4
P =0
0 . 9 HIGH
0.7
0.4
0.0
0.8
0 . 6 M EDIUM
0.3
0 . 0 NONE
0.6
0.4
0 . 2 L OW
0.0
0.3
0.2
0.1
0.0
1 . 0 - 0 .9
CRI TI CAL
0 . 8 - 0 .7
M ARGI NAL
0 . 6 - 0 .4
NEGL IGABLE
0 . 3 - 0 .0
* Ad d it io n a l t e rm i n o l o g y , n o t in US Ai r Fo rc e Gu i d e o n S o ft wa re Ris k Ab a te m e n t
No t e : Ri s k ra ti n g is c o n s is t e n t wit h R = P*C
Often used to track risk history over project
- ESD.36J SPM
10
+
Concept Question 1

Which of the following risks are most
challenging to deal with?






Low Probability – Low Impact
Low Probability – High Impact
High Probability – Low Impact
High Probability – High Impact
Medium Probability – Medium Impact
They are all difficult
- ESD.36J SPM
11
+
Risk Sector Plot (NASA)
-
L e ve l
V a lu e
C rite ria
5
N e a r ce rta in ty
E ve ryth in g p o in ts to th is b e co m in g a p ro b le m , a lw a ys h a s
4
V e ry like ly
H igh ch a n ce of th is b e co m in g a p rob lem
3
L ike ly (5 0 /5 0 )
T he re is an e ve n ch a n ce th is m a y tu rn in to a p ro b lem
2
U n like ly
R isk like th is m a y tu rn in to a p ro b lem o n ce in a w h ile
1
Im p rob ab le
N o t m u ch ch a n ce th is w ill b e co m e p ro b lem
Probability
A ttrib u te : P ro b a b ility
5
4
3
2
1
2
2
3
3
6
5
9 12
8 11
1
1
1
1
2
2
1
2
4
3
2
3
7 10
5 8
3 5
4 5
Impact
A ttrib u te : Im p a c t
L e ve l
V a lu e
T e ch n ica l C rite ria
C o st C rite ria
S ch e d u le C rite ria
5
C a ta stro p h ic
C a n ’t co n tro l th e ve h icle
OR
C a n ’t p e rfo rm th e m issio n
> $ 1 0 M illio n
S lip to le ve l I m ile sto n e s
4
C ritica l
L o ss o f m issio n , b u t
a sse t re co ve ra b le in tim e
$ 1 0 M  X < $ 5 M illio n
S lip to le ve l II m ile sto n e s
3
M o d e ra te
M issio n d e g ra d e d b e lo w
n o m in a l sp e cifie d
$ 5 M  X < $ 1 M illio n
S lip to le ve l III m ile sto n e s
2
M a rg in a l
M issio n p e rfo rm a n ce
m a rg in s re d u ce d
$ 1 M  X < $ 100 K
L o ss o f m o re th a n o n e
m o n th sch e d u le m a rg in
1
N e g lig ib le
M in im u m to n o im p a ct
M in im u m to n o im p a ct
M in im u m to n o im p a ct
- ESD.36J SPM
12
+
Threshold Risk Metric (NASA)
-
12
PROBLEM DOMAIN
RISK*
10
8
MITIGATION
DOMAIN
Pessimistic
Transition Thresholds
6
Expected
Optimistic
4
2
Note: *from risk table
WATCH DOMAIN
Event #1
Feb 96
2
3
Mar 96
Accept
4
Apr 96
5
6
May 96
Time
- ESD.36J SPM
13
+
Application to Hum Log DC

Risks in HumLog Distribution Center Project

Technical




Cost



Cost overrun by Contractor
Donor do not fulfill $ commitments
Schedule


Fire
Blocked Transportation
Over or under Capacity
Delay in Host Government Approval
Programmatic


Contractor Non-Performance
Political Unrest, Looting, Theft
- ESD.36J SPM
14
+
Risk Matrix for HumLog DC Project
-
5
Theft
Probability
4
Unrest
3
Approval
Delay
2
1
Fire
Overrun $
Capacity
1
Donor $
Transport
2
3
4
5
Impact
- ESD.36J SPM
15
+
-
Risk Mitigation Plan
Risk
Risk
Score
Mitigation
Fire
20
Sprinkler System, Hydrant …
Political Unrest
16
Fence, Evacuation Plan
Donor $ Withdrawal
10
Weekly Project Updates, $ Risk Pool
Theft
8
Security System, Inventory Tracking
Approval Delay
6
Government Liaison Officer
Cost Overrun
6
EVM, Firm Fixed Price Contract
Transport Blocking
3
Alternate Transport Path (Sea)
Under-Capacity
1
Lease extra capacity locally
- ESD.36J SPM
16
+
-
Review of Project NPV, VAR
- ESD.36J SPM
17
+
NPV as the main criterion

Net Present Value (NPV)







Is typically the key criterion for deciding go/no-go on projects
Treats undertaking as an investment project
Often used to select among alternative (competing) projects
The project will cause expenditures: conceptual design, detailed
design, manufacturing, implementation, operations
The project will yield revenues: sales from products and services
Both expenditures and revenues occur over time and must be
discounted.
Other Financial Metrics



Payback period
Internal Rate of Return (IRR)
Return on Investment (ROI)
- ESD.36J SPM
18
+
-
Net Present Value (NPV)
• Measure of present value of various cash flows in different
periods in the future
• Cash flow in any given period discounted by the value of a
dollar today at that point in the future
– “Time is money”
– A dollar tomorrow is worth less today since if properly
invested, a dollar today would be worth more tomorrow
• Rate at which future cash flows are discounted is
determined by the “discount rate” or “hurdle rate”
– Discount rate is equal to the amount of interest the
investor could earn in a single time period (usually a
year) if s/he were to invest in an equally risky investment
- ESD.36J SPM
19
+
Net Present Value (NPV)
-
T
Ct
 (1  r )
NPV 
t
t0
1500
500
29
27
25
23
21
19
17
15
13
11
9
7
5
3
0
1
Cashflow, Pt [$]
1000
Cashflow
DCF (r=12%)
-500
-1000
-1500
Program Time, t [yrs]
- ESD.36J SPM
20
+




Discounted Cash Flow (DCF)
Forecast the cash flows, C0, C1, ..., CT of the
project over its economic life
 Treat investments as negative cash flow
Determine the appropriate opportunity cost of
capital (i.e. determine the discount rate r)
Use opportunity cost of capital to discount the
future cash flow of the project
Sum the discounted cash flows to get the net
present value (NPV)
- ESD.36J SPM
21
+
DCF example
-
Period
Discount F actor
Cash F low
Present Value
0
1
-150,000
-150,000
1
0.935
-100,000
-93,500
2
0.873
+ 300000
+ 261,000
Discount rate = 7%
- ESD.36J SPM
NPV =
$18,400
22
+
Concept Question 2

The cash flows on a project are expected to
be -$10M (year 0), $5M (year 1), $5M (year
2) and $5M (year 3). What is the NPV with a
discount rate of r=10%?






+$5M
-$4.5M
+$2.2M
-$1.2M
$0M
Please … no more concept questions ;{
- ESD.36J SPM
23
+

Development Cost Model
Cashflow profiles based on beta curve:
c ( t )  Kt

(1  t )
 1
Typical development time ~1-6 years
Learning effects captured – span, cost
0.06
Support
0.05
Tool Fab
normalized cost

 1
0.04
Tool Design
ME
0.02
Engineering
0.01
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53
normalized time
- ESD.36J SPM
24
+
-
Risk-Adjusted Discount Rate
•
DCF analysis assumes a fixed schedule of cash flows
•
•
•
What about uncertainty?
Common approach: use a risk-adjusted discount rate
The discount rate is often used to reflect the risk
associated with a project: the riskier the project, use a
higher discount rate
Typical discount rates for technology programs: 12-20%
Issues with this approach?
•
•
- ESD.36J SPM
25
+
Value at Risk Concept


Value at Risk (VAR) recognizes
fundamental reality: actual value of any
design or project can only be known
probabilistically
Because of inevitable uncertainty in



Future demands on system
Future performance of technology
Many other market, political factors
- ESD.36J SPM
26
+
Value at Risk Definition

Value at Risk (VAR) definition:



A loss that will not be exceeded at some
specified confidence level
“We are p percent certain that we will not
loose more than V dollars for this project.”
VAR easy to see on cumulative probability
distribution (see next figure)
- ESD.36J SPM
27
+
VAR Cumulative Distribution Function

Look at distribution of NPV of designs A, B:


90% VAR for NPVA is
90% VAR for NPVB is
-$91
$102
Cumulative Probability
CDF
100%
80%
60%
40%
20%
0%
-400
-200
0
200
NPVA
NPVBNPV
90%VAR for NPVB
10% CDF Probability
- ESD.36J SPM
400
600
90% VAR for NPVA
28
+
VAR and Flexibility

VAR is a common financial concept

It stresses downside losses, risks


However, designers also need to look at
upside potential: “Value of Gain”
Flexible design provides value by both:



Decreasing downside risk
Increasing upside potential
See next figure
- ESD.36J SPM
29
+
-
Sources of value for flexibility
Cut downside ; Expand Upside
C u m u la tive P ro b a b ility
E xp a n d u p sid e p o te n tia l
O rig in a l
d istrib u tio n
D istrib u tio n w ith
fle xib ility
C u t d o w n sid e risks
V a lu e
Value-at-Risk-and-Gain (VARG)
- ESD.36J SPM
30
+
-
Parking Garage Project Example
Valuing Options by Spreadsheet: Parking Garage Case Example
Richard de Neufville, Stefan Scholtes and Tao Wang -- ASCE Journal
of Infrastructure Systems, Vol.12, No.2. pp. 107-111, 2006
- ESD.36J SPM
31
+



Intended “Take-Aways”
Design project for fixed objective (mission or
specifications) is engineering base case
Recognizing variability => different design
(because of system non-linearities)
Recognizing flexibility => even better design
(it avoids costs, expands only as needed)
- ESD.36J SPM
32
+



Inspiration - Comments
This is a simplified educational case
But similar real world situations exist
Applicable to general technical projects
Bluewater development in England
(http://www.bluewater.co.uk/)
Image by John Winfield http://bit.ly/16Eala4> on
Wikimedia Commons. License: CC-BY-SA
- ESD.36J SPM
33
+
Parking Garage Case

Garage in area where population expands


Actual demand is necessarily uncertain


Demand drives capacity for # of parking spots
Design Opportunity: Strengthened structure



New commercial/retail opportunities
enables future addition of floor(s) (flexibility)
costs more initially (flexibility costs)
Design issue: is extra cost worthwhile?
- ESD.36J SPM
34
+
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 maybe 250 more spaces
could be 50% off the projections, either way;
Annual volatility for growth is 10%

Consider 20 years






Average annual revenue/space used =
$10,000
The discount rate is taken to be 12%
- ESD.36J SPM
35
+


Parking Garage details (Cont)
Costs
 annual operating costs (staff, cleaning,
etc.) = $2,000 /year/space available
(note: spaces used is often < spaces
available)
 Annual lease of the land = $3.6 Million
 construction cost = $16,000/space + 10%
for each level above the first level
Site can accommodate 200 cars per level
- ESD.36J SPM
36
+
-
Step 1: Set up base case
Demand growth as predicted, no variability
Year
D em and
C apacity
R evenue
R ecurring C osts
O perating cost
Land leasing cost
C ash flow
D iscounted C ash F low
P resent value of cash flow
C apacity costs for up to tw o levels
C apacity costs for levels above 2
N et present value
- ESD.36J SPM
0
$3,600,000
$32,574,736
$6,400,000
$16,336,320
$6,238,416
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
$622,001
capex= capital expenditures=initial investment
37
Optimal design for base case
(no uncertainty) is 6 floors
+
-
optimal
under-capacity
(lost revenue)
over-capacity
(too much capex)
Expected NPV ($, Millions)
10
5
0
2
3
4
5
6
7
8
9
-5
-10
-15
Number of Floors
Traditional NPV
- ESD.36J SPM
38
+
Step 2: Simulate uncertainty
-
Lower demand => Loss
600
Higher demand => Gain limited by garage size
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
- ESD.36J SPM
-9.2
-7.0
-4.9
-2.7
-0.6
1.6
3.7
5.9
8.0
39
+
-
NPV Cumulative Distributions
Compare Actual (5 Fl) with (unrealistic) fixed 6 Fl design
1
0.9
P r oba bility
0.8
0.7
0.6
C D F fo r R esu lt o f
0.5
S im u latio n A n alysis (5-
0.4
flo o r )
Im p lied C D F fo r
0.3
R esu lt o f
0.2
D eter m in istic N P V
A n alysis (6-flo o r )
0.1
0
- 20
- ESD.36J SPM
- 15
- 10
-5
0
5
10
40
Recognizing uncertainty =>
different design (5 floors)
+
Expected NPV ($, Millions)
-
10
5
0
2
3
4
5
6
7
8
9
-5
-10
-15
Number of Floors
Traditional NPV
- ESD.36J SPM
Recognizing uncertainty
41
+
-
Step 3: Introduce flexibility into
design (expand when needed)
Year
0
D em and
C apac ity
D ec is ion on ex pans ion
E x tra c apac ity
R evenue
R ec urring C os ts
O perating c os t
Land leas ing c os t
E x pans ion c os t
1
820
800
$8,000,000
2
3
19
20
924
1,044
800
1,200
ex pand
400
$8,000,000 $10,440,000
1,519
1,600
1,647
1,600
$15,190,000 $16,000,000
$1,600,000
$3,600,000 $3,600,000
$1,600,000
$3,600,000
$8,944,320
$2,400,000
$3,600,000
$3,200,000
$3,600,000
$3,200,000
$3,600,000
$2,800,000
$2,500,000
-$6,144,320
-$4,898,214
$4,440,000
$3,160,304
$8,390,000
$974,136
$9,200,000
$953,734
C as h flow
D is c ounted C as h F low
P res ent value of c as h flow
C apac ity c os t for up to tw o levels
$30,270,287
$6,400,000
C apac ity c os ts for levels above 2
P ric e for the option
N et pres ent value
$7,392,000
$689,600
$12,878,287
Including Flexibility => Another, better design:
4 Floors with strengthened structure enabling expansion
- ESD.36J SPM
42
Summary of design results
from different perspectives
+
-
Perspective
Simulation
Option Embedded
Design
Estimated Expected NPV
Deterministic
No
No
6 levels
$6,238,416
Recognizing Uncertainty
Yes
No
5 levels
$3,536,474
Incorporating Flexibilty
Yes
Yes
4 levels with strengthened
structure
$10,517,140
Why is the optimal design much better
when we design with flexibility?
- ESD.36J SPM
43
-
Sources of value for flexibility:
1) Minimize exposure to downside risk
1
0.9
0.8
Probability
+
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
-20
-15
-10
-5
5-Floor Design
- ESD.36J SPM
0
5
10
4-Floor Design
44
+
Sources of value for flexibility:
-
2) Maximize potential for upside gain
100.0%
90.0%
Mean for NPV
without Flexibility
80.0%
CDF for NPV
with Flexibility
Probability
70.0%
60.0%
50.0%
40.0%
CDF for NPV
without Flexibility
30.0%
4
5
Mean for NPV
with Flexibility
20.0%
10.0%
0.0%
-20
- ESD.36J SPM
-15
-10
-5
0
5
10
15
20
25
30
35
45
Comparison of designs
with and without flexibility
+
-
Design
Design with Flexibility Thinking
Design without Flexibility thinking
Comparison
(4 levels, strengthened structure)
(5 levels)
Initial Investment
$18,081,600
$21,651,200
Better with options
Expected NPV
$10,517,140
$3,536,474
Better with options
Minimum Value
-$13,138,168
-$18,024,062
Better with options
Maximum Value
$29,790,838
$8,316,602
Better with options
Wow! Everything is better! How did it happen?
Root cause: change the framing of the problem
• recognize uncertainty
• add in flexibility thinking
- ESD.36J SPM
46
+
Summary from Garage Case

Sources of value for flexibility




Cut downside risk
Expand upside potential
VAR chart is a neat way to represent the
sources of value for project flexibility
Spreadsheet with simulation is a powerful
tool for estimating value of flexibility
- ESD.36J SPM
47
+
-
Real World meets Academia
Screenshot of the news article, "Logan Parking Squeeze Could Reach Crisis Level," from Boston Business Journal removed due to
copyright restrictions. Please see http://www.bizjournals.com/boston/stories/2001/11/05/story3.html for further reference.
- ESD.36J SPM
48
+
Summary


Layers of Risk
Classical Risk Management




One way to aggregate risks is NPV


Some issues such as system safety need to be tracked in
addition to NPV
NPV is uncertain



Risk = Probability of Event X Impact of Event
Risk Management Cycle (identify, Analyze, Plan, Track, Control)
Risk Identification is not enough, must do something
Compute Value-at-Risk (VAR) curves
Shape VAR curves by embedding flexibility
Real Options embody formal concept of flexibility
- ESD.36J SPM
49
+
Acknowledgements

Prof. Don Lessard


Concept of the Layers of Risk
Prof. Richard de Neufville


Garage Case Study
Tao Wang, ESD PhD
Real Options "in" Projects and Systems Design -- Identification of
Options and Solutions for Path Dependency, Doctoral Dissertation,
MIT, May, 2005
- ESD.36J SPM
50
MIT OpenCourseWare
http://ocw.mit.edu
ESD. 6\VWHP3URMHFW0DQDJHPHQW
Fall 2012
For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.