1. + University
of the
Philippines
Distinguished
Systems Thinking in Alumni
Lecture
Dynamic Planning of College of Series
Energy Systems Engineering
Diliman,
Jose B. Cruz, Jr. Quezon City
The Ohio State University Engineering
Columbus, Ohio USA Theatre
17 July 2009
2. +
Why So Much Focus on Energy?
Rapid depletion of fossil fuels
during the past 100 years, from
accumulations of fossils during the
past hundreds of millions of years.
Negative impact of burning fossil
fuels on the environment – global
warming.
2
3. +
How Did This Arise?
Engineering inventions, and
advances in science and
technology during the 20th
century have transformed
society into one with intense
and pervasive use of electrical
energy.
3
4. NAE List of 20th Century Greatest
+
Engineering Achievements
In rank order
1 Electrification
2 Automobile
3 Airplane
4 Water Supply and
Distribution
5 Electronics
4
5. NAE List of 20th Century Greatest
+
Engineering Achievements
In rank order
6 Radio and Television
7 Agricultural Mechanization
8 Computers
9 Telephone
10 Air Conditioning and
Refrigeration
5
6. + NAE List of 20th Century Greatest
Engineering Achievements
In rank order
11 Highways
12 Spacecraft
13 Internet
14 Imaging
15 Household Appliances
6
7. + NAE List of 20th Century Greatest
Engineering Achievements
In rank order
16 Health Technologies
17 Petroleum and
Petrochemical Technologies
18 Laser and Fiber Optics
19 Nuclear Technologies
20 High-performance Materials
7
8. +Consequences of Our Modern
Way of Living
Electrification led to other great
engineering achievements.
Each of the 19 other
achievements is closely coupled
to the availability of electricity.
Each implies greater use of
energy.
8
9. +Consequences of Our Modern
Way of Living
The total energy demanded by
the technologically transformed
world can not be met without the
availability of high energy density
fossil fuels (coal and oil).
9
10. +Consequences of Our Modern
Way of Living
Continued high rate of burning of
fossil fuels using current
technologies releases carbon
dioxide and other gases,
contributing to global warming,
and placing earth and humanity
at great risk.
10
11. +
Energy Expenditures (USA)
(From EIA AEO 2008)
11
13. +
Energy Production by Fuel
(From EIA AEO 2008)
13
14. +
Energy Consumption by Fuel
(From EIA AEO 2008)
14
15. + Characteristics of Large-Scale
Systems
Presence of more than one
stakeholder and decision-
maker
Presence of dynamics (next
state depends on current state
and current action)
15
16. +
What Is a Dynamic System?
Simplest Class: Modeled by an ordinary differential equation,
where the independent variable is time.
Example: Vertical motion of an automobile tire moving on a rough road
d2y dy
m +d + ky = f . This is usually written
dt 2
dt
in vector-matrix form as
dx ⎡ 0 1 ⎤ ⎡ 0 ⎤
=⎢ ⎥x + ⎢ ⎥
dt ⎢ −k / m −d / m ⎥
⎣ ⎦ ⎢ f /m
⎣ ⎥
⎦
⎡ x ⎤
where x = ⎢ 1 ⎥ , x = y, x = dy
⎢ x ⎥ 1 2
dt
⎣ 2 ⎦
when simulating in MATLAB 16
17. +
Discrete-time Dynamics
Difference equations rather than
differential equations are used.
Example: Autoregressive Moving
Average (ARMA)
yk + an−1 yk −1 + ... + a0 yk −n =
bmuk + bm−1uk −1... + b0uk −m
Vector-Matrix Representation:
x k +1 = Axk + Bu
17
18. + What is Game Theory?
Game Theory is a body of knowledge concerning
decision-making in a system with two or more
Decision Makers.
A player is a decision maker or a controller.
A player chooses a decision, strategy, or control.
A decision choice is based on available information.
Associated with each player is a cost or pay-off
function.
A cost depends on one or more decisions.
Much of game theory deals with how a player
selects a decision.
18
19. Very Brief History of Game Theory
+
Mathematical foundation of game theory:
John von Neumann and O. Morgenstern, Theory of
Games and Economic Behavior, Princeton University
Press, 1944 [1].
Game theory cuts across multiple disciplines of
mathematics, operations research, economics, political
science, control theory, and engineering.
For a recent brief history see
Jose B. Cruz, Jr. and Xiaohuan Tan, Dynamic
Noncooperative Game Models for Deregulated Electricity
Markets, Nova Publishers, 2009 [2,Section 2.1].
19
20. +
Different Solution Concepts in Game
Theory
Players are assumed to be rational.
Zero-sum, min-max, max-min:
Pure strategies (deterministic choice)
Mixed strategies (choice of random
distribution)
Nash equilibrium for nonzero sum games.
Pareto optimality.
Stackelberg equilibrium.
20
21. +
Dynamic Game Theory
1 1 1
For DM1 :Find {u ,u ,...,u 0 1 N −1
}
to "optimize" J1
2 2 2
For DM2 :Find {u ,u ,...,u 0 1 N −1
}
to "optimize" J2
N −1
i
Ji = L (xN ) +
N ∑L i
k
1 2
(xk ,u ,u )
k k
k =0
21
22. + For Energy Systems, What
Game Concept Is Appropriate?
• The government is one of the DMs and
it will be a dominant player.
• The dynamic Stackelberg strategy
(Leader-Follower) is appropriate.
• The dominant player is the Leader and
announces its sequence of strategies first.
22
23. + Static Stackelberg Game
• Let there be two players, Player 1 and Player 2.
• ui is the decision variable of Player i,
ui ∈Ui , i = 1, 2.
• Ji (u1,u2 ) is the scalar cost function of Player i,
i = 1, 2.
• One player, called the Leader, declares its
decision strategy first.
• The other player is called the Follower.
• H. von Stackelberg, The Theory of the Market Economy,
Oxford University Press, English translated ed., 1952 [3].
23
24. + Static Stackelberg Game -2
• Reaction Set of Player 1: D1 = {(u1,u2 ) ∈U1 × U2 :
T: U2 → U1, u1 = Tu2 , J1(Tu2 ,u2 ) ≤ J1(u1,u2 ) for
all u1 ∈U1, for each u2 ∈U2 }.
• Stackelberg strategy pair with Player 2 as Leader,
Player 1 as Follower:
(u1S 2 ,u2S 2 ) ∈{(u1S 2 ,u2S 2 ) ∈D1 : J2 (u1S 2 ,u2S 2 ) ≤ J2 (u1,u2 )
for all (u1,u2 ) ∈D1 }.
• Similarly Player 1 may be the Leader and Player 2
the Follower.
24
25. + Historical Roots of Dynamic
Games
4. R. P. Isaacs, Differential Games: a Mathematical
Theory with Applications to Warfare and Pursuit,
Control and Optimization. New York: John Wiley and
Sons, 1955. First book on dynamic games.
5. Y.C. Ho, “Differential Games, Dynamic Optimization,
and Generalized Control Theory,” Journal of
Optimization Theory and Applications, Vol. 6, No. 3,
1970. Clarified connections of control theory to
dynamic game theory.
6. T. Basar and G.J. Olsder, Dynamic Noncooperative
Game Theory, 2nd Edition (revised), the Society for
Industrial and Applied Mathematics, 1998.
Comprehensive and extensive treatment of dynamic
games.
25
26. + Dynamic Games
The evolution of a discrete-time dynamic system is modeled by difference
equations x(k + 1) = f (x(k),u1(k),u 2 (k),k), where x(k) is the state vector,
ui (k) is the control or decision vector of Player i, and k is discrete time or
time stage, k = 0,1, 2,...,N, and f is a mapping from x, u1, u 2 , and t to the
space of x. The sequence {x(k)} describes the evolution of the state as a
consequence of the application of decisions u1, u 2 applied at preceding time
stages. A continuous time model is described by a set of ordinary
differential equations x = f (x(t),u1(t),u 2 (t)), t ∈[t0 ,tf ], x(t0 ) = x0 and the
symbols are defined similarly. An open loop control is a time sequence
ui = {ui (0),ui (1),...,ui (N − 1)}, starting at a given state x(0) = x0 .
A closed loop control is a sequence
{ui (k,x(k))} = {ui (0,x(0)),ui (1,x(1)),...,ui ((N − 1),x(N − 1))}.
26
27. Stackelberg Strategies for Dynamic Games*
+
First considered by Chen and Cruz, and Simaan and Cruz:
7. C.I. Chen and J.B. Cruz, Jr., “Stackelberg
Solution for Two-Person Games with Biased
Information Patterns,” IEEE Trans. on Automatic
Control, Vol. AC-17, No. 6, December 1972, pp.
791-798.
8. M. Simaan and J.B. Cruz, Jr., “On the
Stackelberg Strategy in Nonzero-Sum Games,”
Journal of Optimization Theory and Applications, Vol.
11, No. 5, May 1973, pp. 533-555.
9. M. Simaan and J.B. Cruz, Jr., “Additional Aspects
of the Stackelberg Strategy in Nonzero-Sum Games,”
Journal of Optimization Theory and Applications, Vol.
11, No. 6, June 1973, pp. 613-626.
27
28. + Dynamic Stackelberg Games
• A player, say Player 2, called the Leader, commits to a strategy
for the entire horizon of the game and announces it before the start
of the game.
• The other player, Player 1, called the Follower is aware of the
Leader's commitment and as a rationale decision maker proceeds
1 2
to optimize its cost function J1(u ,u ) with respect to its choice for a
control sequence u1, taking into account the Leader's commitmnt to
2
a specific control sequence u .
• The Leader notes how the Follower will react, and chooses u to 2
optimize J2 (u1,u 2 ) under the condition that u1 is a reaction to u 2 .
28
29. + Dynamic Stackelberg Games - 2
• Define the reaction set of the Follower as the set of of u1
sequence for each possible announced commitment of u2
sequence by the Leader:
R1 = {(u1,u 2 ) : J1(u1,u 2 ) ≤ J1(v 1,u 2 ) for all v 1 ∈U 1 and
for each u 2 ∈U 2 }
• If Player 1 is the Leader a similar reaction set for the
Follower is defined:
R 2 = {(u1,u 2 ) : J2 (u1,u 2 ) ≤ J2 (u1,v 2 ) for all v 2 ∈U 2 and
for each u1 ∈U 1 }
• The Leader selects its decision from the reaction set of the
Follower that results in the minimum of its cost function. 29
31. Determining a Stackelberg
+
Closed Loop Strategy
At
the initial state x = 1, each Player chooses a
decision of 0 or 1.
At
time 1, state x = 2, each player chooses a
decision of 0 or 1.
At
time 1, state x = 1, each player chooses a
decision of 0 or 1.
At
time 1, state x = 0, each player chooses a
decision of 0 or 1.
Each decision maker or Player has 16 choices
31
33. Reaction Sets
+
R1c = {(c115 ,c 21 ), (c18 , c 22 ), (c113 , c 23 ), (c16 , c 24 ), (c111, c 25 ),
(c14 , c 26 ), (c19 , c 27 ), (c12 , c 28 ), (c115 , c 29 ), (c116 , c 210 ),
(c15 , c 211 ), (c111, c 212 ), (c16 , c 213 ), (c112 , c 214 ), (c11, c 215 ),
(c12 , c 216 )}
R2c = {(c11, c 211 ), (c12 , c 211 ), (c13 , c 2 ), (c14 , c 211 ), (c15 , c 211 ),
(c16 , c 211 ), (c17 , c 211 ), (c18 , c 211 ), (c19 , c 211 ), (c110 , c 23 ),
(c111, c 211 ), (c112 , c 23 ), (c113 , c 211 ), (c114 , c 23 ), (c115 , c 211 ),
(c116 , c 23 )}
For example, for u 2 = c 26 (0,1,0,1) Player 1 minimizes J1 and
gets u1 = c14 ,(0, 0, 1 ,1). This is repeated for each u 2 = c 2 j ,
thus obtaining R1c . 33
35. + Stackelberg Example
There are two closed loop Stackelberg controls with Player 2 as
Leader, (c15 , c 211 ) and (c16 , c 212 ), both leading to J1S 2 = 7 and
c
J2S 2 = 2 and to the same trajectory x(1) = 2 and x(2) = 1. At time
c
t = 1, the remaining controls are (u1,u 2 ) = (1,0) and the remaining
costs are J1 = 2 and J2 = 5. Supose that Player 2 considers
violating its commitment made at time t = 0 regarding its control
at time t = 1. Its closed loop Stackelberg strategy for a game
starting at t = 1 and x(1) = 2 is (u1,u 2 ) = (0,1) leading to J1 = 6 and
J2 = 3. It will be tempted to violate its commitment to reduce its cost.
35
36. + Stackelberg Example -2
Thisexample shows that the closed loop
Stackelberg strategy violates Bellman’s
principle of optimality. That is, the
continuation of a previously announced
closed loop strategy, starting at a later time,
is not necessarily a closed loop Stackelberg
strategy for a new game starting at the later
time.
This example is in Simaan and Cruz [Ref 9].
36
37. + Stackelberg Example -3
Forthe same example in [9] it was shown
that the open loop strategy for a game
starting at t=0, x=1 violates the principle
of optimality.
Ifa Leader violates its commitment made at
an earlier time and changes its strategy at a
later time with a new commitment, there will
be a credibility problem. The Follower may not
believe a subsequent commitment by the
Leader.
37
38. + Stackelberg Example - 4
This
violation of the principle of optimality is
known as time-inconsistency in economics.
The credibility problem and the time-
inconsistency problem suggest that a
Stackelberg-like closed loop strategy that
satisfies the principle of optimality would be an
acceptable suboptimal alternative.
Such a strategy, called Feedback
Stackelberg was introduced in [7], precisely
defined and fully described in [9]. It is a
suboptimal closed loop Stackelberg-like
strategy. But it is time-consistent. 38
39. + Feedback Stackelberg
Strategies
Principal
property: The principle of
optimality holds. (Time-consistency holds).
Dynamic programming can be applied.
Optimal Cost-to-go at stage k is the sum of
the incremental cost at stage k plus the
optimal cost-to-go at the next stage k+1,
where the optimization is performed stage
by stage starting with the last stage, in the
sense of Stackelberg.
39
40. + Dynamic Stackelberg
Strategies
Feedback Stackelberg
strategies proposed for the
first time in Simaan and Cruz
1973 are suboptimal but time-
consistent and widely used in
macroeconomics.
The same methodology can
be applied to energy systems.
40
41. Dynamic Stackelberg Strategies
+
Are Pervasive in Macroeconomics
Kydland and Prescott published a paper in
1977 showing that the government
strategy is time-inconsistent (violates the
principle of optimality of Bellman’s Dynamic
Programming). This paper revolutionized
the entire field of macroeconomics.
Kydland published a more theoretical
paper in 1975, used as a reference in
Kydland and Prescott, 1977).
41
42. Dynamic Stackelberg Strategies
+
Are Pervasive in Macroeconomics
Kydland and Prescott won the Nobel
prize in economics in 2004.
Kydland, 1975 referred to Simaan and
Cruz, 1973, where time-inconsistency is
proved. Simaan and Cruz, 1973 is a
reference in Kydland’s Ph.D. dissertation
supervised by Prescott at Carnegie
Mellon University in 1974.
42
43. Related Developments in
+
Economics
12. Finn Kydland, “Equilibrium Solutions in Dynamic
Dominant Player Models,” Journal of Economic
Theory, 15, 307-325, 1977. Kydland states that
the dominant solution, open loop or closed loop
are time-inconsistent. Suggests that a feedback
solution is self-enforcing. Cites Simaan and Cruz
[8,9].
13. Guido Taballini, “Finn Kydland and Edward
Prescott’s Contribution to the Theory of
Macroeconomic Policy,” Scand. J. of Economics,
107(20), 203-216, 2005. Taballini notes that
Kydland [12] cites Simaan and Cruz [8,9].
43
44. Related Developments in Economics
+
10. Finn Kydland, “Noncooperative and Dominant
Player Solution in Discrete Dynamic Games,
“International Economic Review, Vol. 16, No. 2,
June 1975, pp. 321-335. Cites Simaan and Cruz:
“The dominant player problem, on the other hand,
has only recently received a little attention in the
game literature, and the two interesting papers by
Simaan and Cruz [24,25] should be mentioned.”
11. Finn E. Kydland and Edward C. Prescott, “Rules
Rather Than Discretion: The Inconsistency of
Optimal Plans,” The Journal of Political Economy,
Vol. 85 No. 3 (June 1977), pp. 473-492. This paper
is one of the bases for Kydland and Prescott to be
selected for the 2004 Nobel Prize in Economics. 44
45. + Introduction to Current Joint Work
with R R Tan and A B Culaba, DLSU
Energy consumption is closely coupled with both
economic growth and greenhouse gas emissions.
Despite the increasing popularity of renewables, the
world remains highly dependent on fossil fuels for
transportation, power generation and industrial use.
Various novel solutions are at inherent disadvantage
compared to entrenched technologies due to network
externalities.
Presented at the 29th APAMS, July 13 - 15, 2009
46. + Some Examples of Nascent
Energy Supply Chains
Biofuel production systems from dedicated energy
crops
Fossil-based electricity production with carbon
capture and storage
The “hydrogen economy”
Presented at the 29th APAMS, July 13 - 15, 2009
47. + Motivating Case
In the Philippines, Jatropha curcas has been
touted as a promising dedicated energy
crop for biodiesel production
However, investments in upstream (farm-
level) production capacity has not been
matched by corresponding growth in
downstream (oilseed pressing and
conversion) capacity
This imbalance in the J. curcas supply chain
is typical of nascent energy systems.
Presented at the 29th APAMS, July 13 - 15, 2009
48. +
The Basic Model
(Cruz et al., 2009)
Axt = yt Material and energy balances
of physical streams
xt+1 = B(zt – yt) + xt
Response of
where: production
capacity to
A = technical coefficient matrix deficits or
surpluses
xt = sectoral total output vector at t
yt = sectoral net output vector at t
B = influence matrix
Presented at the 29th APAMS, July 13 - 15, 2009
49. + Key Assumptions
MatrixA reflects scale-invariant physical
relationships such as process yields
Matrix B reflects econometrically determined
collective behavioral responses of supply chain
agents
Vector
x reflects total system outputs, including
intermediates
Vectoractual y reflects net system outputs,
while z gives the desired output level.
Productioncapacities are assumed to respond to
surpluses or deficits incurred in the previous
time interval.
Presented at the 29th APAMS, July 13 - 15, 2009
50. +
The Basic Model
(Cruz et al., 2009)
(I – BA) defines the
dynamic
xt+1 = (I – BA)xt + Bzt characteristics of
the system.
zt = Kxt + zo
Adaptive target
output level is
where: introduced
K = control matrix
(I – BA + BK) now
zo = baseline target output defines the
dynamic
characteristics of
the controlled
xt+1 = (I – BA + BK)xt + Bzo system.
Presented at the 29th APAMS, July 13 - 15, 2009
51. The Extended Model
Material and energy balances
of physical streams
Response of
production
capacity to
deficits or
Lagged influences surpluses
may be interpreted
probabilistically
Presented at the 29th APAMS, July 13 - 15, 2009
52. +
The Extended Model
The extended
model is thus
Denoting reduced to the
same form as the
previous one.
Presented at the 29th APAMS, July 13 - 15, 2009
53. Case Study 1
(Scenario 1, Cruz et al., 2009)
Energy crop
Biofuel
Land
Farming Biofuel
production
Presented at the 29th APAMS, July 13 - 15, 2009
54. +
Case Study 1
(Cruz et al., 2009)
Presented at the 29th APAMS, July 13 - 15, 2009
55. Case Study 2
(Scenario 4, Cruz et
al., 2009)
Presented at the 29th APAMS, July 13 - 15, 2009
56. Case Study 2
(Scenario 4, Cruz et
al., 2009)
Presented at the 29th APAMS, July 13 - 15, 2009
57. Case Study 3
(Scenario 5, Cruz et
al., 2009)
Farm
Farm capacity
capacity responds to
increases biodiesel
with oilseed surplus or
surplus deficit
Biodiesel production
capacity exhibits sluggish
response
Presented at the 29th APAMS, July 13 - 15, 2009
58. Case Study 4
We revisit Case 3, but
introduce time lags with:
( 1, 2, 3)
T = (0.3, 0.5,
0.2)T
Presented at the 29th APAMS, July 13 - 15, 2009
59. + Key Implications
Undesirable dynamic characteristics in nascent
energy supply chains may arise due to feedback
loops in physical linkages or information flows.
Control theory can be used to systematically
design interventions to suppress undesirable
system behavior.
Such interventions can come in the form of policy
instruments or economic incentives/disincentives.
Presented at the 29th APAMS, July 13 - 15, 2009
60. +
Conclusions
We have extended our dynamic input-output
model for nascent energy supply chains to
incorporate weighted time lags in the capacity
response.
This extension allows for added flexibility in
modeling real systems wherein changes in
production capacity may be subject to time
delays.
Presented at the 29th APAMS, July 13 - 15, 2009
61. +
Conclusions
We have extended our dynamic input-output
model for nascent energy supply chains to
incorporate weighted time lags in the capacity
response.
This extension allows for added flexibility in
modeling real systems wherein changes in
production capacity may be subject to time
delays.
Presented at the 29th APAMS, July 13 - 15, 2009
62. Pending Collaboration with UPLB
+
Virgilio T.
Villancio
Program Leader
Integrated R&D on Jatropha curcas for Biodiesel
UP Los Banos
63. +
OPPORTUNITIES
Growing demand for
biofuels
Unstable prices of
crude oil
Rising prices of
vegetable oils
Need for non-food
sources of oil
Higher value of by-
products as additional
source of revenue
65. +
It is locally known as Tubang
bakod, Tuba-tuba, Kasla, Tubang
aso, Tubang silangan, tawa-tawa
Planted in fences
for hedges, thus the
term Tubang bakod
Seeds are grounded
and used to poison fish
thus the term Tuba
Leaves are used as herbal
medicine for fractures
66. +
• 2,000‐5,000 kg/ hectare/ year
(depending on the quality of Jatropha
seed and soil)
• 0.3‐ 9 kg / tree seed producBon
• Can bear fruit throughout the year
• Oil yield 30 – 40% crude non‐edible oil
• 0.75 – 2 tons biodiesel / hectare
67. +
PROCESSING AND UTILIZATION
FEEDSTOCK PRODUCTION
Mechanical processing
Germplasm Management, EnzymaNc processing
Varietal improvement, seed Processing of by‐products
technology, provenance tesNng Waste management
Nursery development
Development of producNon GOALS
systems, prototype plantaNon BUSINESS AND
ENTERPRISE Rural employment
Soil FerNlity management DEVELOPMENT Income generaNon
Pest and diseases management Energy independence
Cleaner environment
Flowering and fruiNng MARKET DEVELOPMENT
physiology
Product development and
Post ProducNon management promoNon PNOC FUNDED
Technology promoNon Establishment of the value DOST-PCARRD
chain FUNDED
CHED FUNDED
SOCIAL ECONOMICS POLICY ENVIRONMENTAL
Capacity development
71. +
• 3 fruit clusters per branch
per fruiBng season
Matured
pods • 12 fruits per bunch
• 2.66 seeds per fruit
• 48 branches per tree
• 1,600 trees per hectare
• 1,400 seeds per kg
• 5,250 kg per hectare
74. +
JATROPHA PLANTATION AT ZAMBOANGUITA, DUMAGUETE
75. + R & D Plan for OSU,
DLSU, UPLB
Investigate Total Dynamic Supply Chain.
Investigate genetic reengineering of
Jatropha curcas for improved total supply
of biodiesel (oil content, continuous
harvesting, less water needs).
Develop strategies for various
stakeholders
Investigate dynamic policy interventions. 75
77. +
The Engineer of 2020
A Study by the National
Academy of Engineering
77
78. The premise
Past: Engineering and engineering education
were reactive, responding to change.
Today: Rapid change signals that it is time to
reverse the paradigm.
Premise: If we anticipate the future and are
proactive about changing engineering and
engineering education, we can shape a
significant, dynamic role for our profession.
79. The process
Phase I: Imagining the future
and the challenges it will
present to engineering: Woods
Hole Workshop.
Phase II: Considering how
engineering education should
prepare for that future:
Washington DC Summit.
National Academy of Engineering
80. Steering Committees
Phase I Phase II
Wayne Clough, Chair, Ga Tech Wayne Clough, Chair, Ga Tech
Alice Agogino, UC Berkeley Alice Agogino, UC Berkeley
George Campbell, Cooper Union Mark Dean, IBM
James Chavez, Sandia Labs Deborah Grubbe, DuPont
David Craig, Reliant Energy Randy Hinrichs, Microsoft
Jose Cruz, Ohio State Sherra Kerns, Olin College
Peggy Girshman, NPR Alfred Moye, H-P
Daniel Hastings, MIT Diana Natalicio, UT at El Paso
Michael Heller, UC San Diego Siman Ostrach, Case West Res
Deborah Johnson, U Virginia Ernest Smerdon, U Arizona
Alan Kay, H-P Karan Watson, Texas A&M
Tarek Khalil, U Miami David Wisler, GE Aircraft Engines
Robert Lucky, Telcordia Technologies
John Mulvey, Princeton
Sharon Nunes, IBM
Sue Rosser, Georgia Tech
Ernest Smerdon, U Arizona
81. Context for engineering
Breakthroughs in technology
Demographics
Challenges
Economic/societal forces
83. Demographics
8 billion people; a 25% increase since 2000.
Balance tipped toward urbanization.
Youth “bulge” in underdeveloped nations while
developed nations age.
If the world condensed to 100 people:
56 in Asia 7 in Eastern Europe/Russia
16 in Africa 4 in the United States
85. Economic/societal forces
High speed communications /
Internet
Removal of trade barriers
Terrorist attacks; wars in Iraq,
Afghanistan
Emergence of technology-
based economies in other
nations
Sustained investment in
higher education in countries
like China, India
86. Social, global and professional
context of engineering practice
Population is more diverse.
Social, cultural, political forces will shape and affect
the success of technological innovation.
Consumers will demand higher quality,
customization.
Growing imperative for environmental sustainability.
Increasing focus on managing risk and assessment
with view to security, privacy, and safety.
87. Aspirations for the Engineer of 2020
Engineering’s image
Public that understands and appreciates the
impact of engineering on socio-cultural systems.
Public that recognizes engineering’s ability to
address the world’s complex and changing
challenges.
Engineers will be well grounded in the
humanities, social sciences, and economics as
well as science and mathematics.
88. Aspirations for the Engineer of 2020
Engineering without boundaries
Embrace potentialities offered by creativity,
innovation, and cross-disciplinary fertilization.
Broaden influence on public policy and the
administration of government, nonprofits, and
industry.
Recruit, nurture and welcome underrepresented
groups to engineering.
89. Aspirations for the Engineer of 2020
Engineering a sustainable society
Lead the way toward wise, informed,
economical, and sustainable development.
Assist in the creating of an ethical balance
in standard of living for developing and
developed countries alike.
90. Aspirations for the Engineer of 2020
Educating the engineer of 2020
Reconstitute engineering curricula and related
educational programs to prepare today’s
engineering students for the careers of the future.
Create a well-rounded education that prepares
students for positions of leadership and a
creative and productive life.
91. Attributes of the engineer of 2020
Strong analytical skills
Practical ingenuity, creativity; innovator
Good communication skills
Business, management skills
High ethical standards, professionalism
Dynamic/agile/resilient/flexible
Lifelong learner
Able to put problems in their socio-technical
and operational context
Adaptive leader
92. To succeed
Attract best and brightest with
a forward-looking educational
experience – Phase II.
Educate them to be ready:
To implement new technology.
To focus on innovation.
To understand global trends.
93. Thoughts from the Phase II summit
Some needs have not changed:
A sound grounding in science
The learning experience of great lectures
Studio experiences with open-ended problem solving
Other things have really changed:
Access to IT creates challenge of coupling deep learning
with instant gratification
Means and ends of using computers to bring the world to
campus and enrich learning
Design tools and sophisticated instruments that enable
students to experience the excitement of engineering
Charles Vest
94. Thoughts from the Phase II summit
Research/co-op experience with real problems
Experience with real-world tools and teams
Encourage and recognize diversity
Social, ethical aspects of engineering
What students need to learn instead of what
we want to teach
Creative and practical thinking
Arden Bement
95. Highlights from Phase II summit
Break out of the present mold
Education, not just curriculum
Career, not just jobs
Multiple models, not just one
Leadership, not just teamwork
More coordination with industry
Cross-disciplinary emphasis
96. More highlights from Phase II summit
Emphasis on innovation
Systems approach
Larger context for engineering
and technology
Non-engineering career tracks
Global perspective
Market forces, macroeconomics
Sense of urgency
97. + References
The National Academies Summit on America’s
Energy Future: Summary of a Meeting, National
Research Council, 2008
http://www.nap.edu/catalog/12450.html
Electricity from Renewable Resources: Status,
Prospects, and Impediments, National Research
Council, 2009
http://www.nap.edu/catalog/12619.html
J. B. Cruz, Jr., R. R. Tan, A. B. Culaba, J-A. Ballacillo,
“A Dynamic Input-Output Model foe Nascent
Bioenergy Supply Chains,” Applied Energy, 2009.
78
98. +
References
The
Engineer of 2020: Visions of
Engineering in the New Century,
National Academy of Engineering,
2004.
Educating
the Engineer of 2020:
Adapting Engineering Education to the
New Century, National Academy of
Engineering, 2005.
79