Understanding Industrial Ecology Dynamics
and Competitive Enterprise Strategies in the
Large Commercial Aircraft Industry
HYBRID AGENT-BASED SYSTEMS DYNAMICS SIMULATION
AS A TOOL FOR ARCHITECTING THE EXTENDED ENTERPRISE
Sgouris Sgouridis
ESD PhD Candidate
Research Presentation
LAI PLENARY CONFERENCE
1000
900
800
700
600
San Antonio, TX
April, 20th 2006
500
400
300
200
100
1 96
1 96 0
1 96 2
1 96 4
1 96 6
1 97 8
1 97 0
1 97 2
1 97 4
1 97 6
1 98 8
1 98 0
1 98 2
19 4
8
1 98 6
1 99 8
1 99 0
1 99 2
1 99 4
19 6
9
2 00 8
2 00 0
2 00 2
4
0
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
1
Agenda
• The Aviation Industry Ecosystem
• Why model the aviation industry?
• Methods: existing models and
frameworks
• The hybrid agent-based System
Dynamics modeling approach
• Expected Value
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
2
A view of the Commercial
Aviation Industry Enterprise
Ecology
Business Passengers
Leisure Passengers
Airport Authorities
American
SouthWest
Boeing
Unions
Ryanair
Lufthansa
Airbus
?
…
JAL
Capital Markets
Leasing companies
Engine
Manufacturers
Suppliers
FAA
DOT
EPA
Congress
DOC
WTO
US Government
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
EU Government
3
A taxonomy of the extendedenterprise
Integral
Customer
Government
Unions
OEM
Capital Markets
Customer
Government
Suppliers
Unions
OEM
Suppliers
Capital Markets
Modular
Different enterprises interact with their environment differently
More details in Piepenbrock (2004)
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
4
Why Model the Aviation Industry?
From Hansman (2004)
Oscillatory behaviors.
Response amplification.
250
Aircraft % Change
8
US GDP Change
200
4 per. Mov. Avg. (US GDP Change)
6
150
4
100
2
50
19
6
19 1
6
19 2
6
19 3
6
19 4
6
19 5
6
19 6
6
19 7
6
19 8
6
19 9
7
19 0
7
19 1
7
19 2
7
19 3
7
19 4
7
19 5
7
19 6
7
19 7
7
19 8
7
19 9
8
19 0
81
19
8
19 2
8
19 3
8
19 4
8
19 5
8
19 6
8
19 7
8
19 8
8
19 9
9
19 0
9
19 1
9
19 2
9
19 3
9
19 4
9
19 5
9
19 6
9
19 7
9
19 8
9
20 9
0
20 0
0
20 1
0
20 2
0
20 3
04
0
0
-50
-2
-100
-4
What are the mechanics behind the observed
behavior?
What are the dominant causality drivers?
How can the response be damped?
Is damping desirable by all stakeholders?
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
5
Why Model the Aviation Industry?
GDP
+
250
Pax
Demand
+
Aircraft
Orders
8
Aircraft % Change
US GDP Change
200
4 per. Mov. Avg. (US GDP Change)
6
150
4
100
2
50
0
19
6
19 1
6
19 2
6
19 3
6
19 4
6
19 5
6
19 6
6
19 7
6
19 8
6
19 9
7
19 0
7
19 1
7
19 2
7
19 3
74
19
7
19 5
7
19 6
7
19 7
7
19 8
7
19 9
8
19 0
8
19 1
8
19 2
8
19 3
8
19 4
8
19 5
8
19 6
8
19 7
8
19 8
8
19 9
9
19 0
9
19 1
9
19 2
9
19 3
9
19 4
9
19 5
9
19 6
9
19 7
9
19 8
9
20 9
0
20 0
0
20 1
0
20 2
0
20 3
04
0
-50
-2
-100
-4
Number of
Competitors
+
-
Aircraft
Pricing
Airline
Profitability
+
+
+
GDP
+
Projected
Demand
+
+
Aircraft
Orders
+
Current Operating
Fleet
Pax
Demand
+
Fares +
Several layers of Complexity
-
Load
Factors
-
Service
Quality
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
6
Commercial Aviation Modeling
Overview
Lynn 1997
Aris 2002
Gillet 1994
Piepenbrock 2004
Bhadra 2003
Benkard 2004
Narrative-based
Framework-based
Agent-Based
Niedringhaus
2004
Econometrics
Hansen 1990
Adler and Berchman 2004
Game Theory
System Dynamics
/ Differential Eq.
Krugman 1987
Esty and Ghemawat 2002
Weil 1996
Lyneis 2000
Liehr, Grossler et al. 2001
Miller and Clarke 2004
Some examples of applications
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
7
Methodology Use in this Effort
Framework-based
Narrative-based
Econometrics /
Demand
Modeling
Hypothesis
Generation
Verification
Model Structure
Supporting
Functions
System Dynamics
Game Theory
Agent schemata
Agent-Based
Aviation Industry
Model
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
8
Model Agents and Schemata
PASSENGERS
Demand
Price
Lufthansa
American
…
Ryanair
SouthWest
JAL
AIRLINES
Leasing companies
Price
Demand
AIRCRAFT?
MANUFACTURERS
Airbus
Boeing
Government
Agents provide the
flexibility to model
individual behaviors
(schemata) that may
reflect :
• strategies,
• utility functions,
• path-dependencies
(e.g. effects of chance
events, learning, and
evolutionary behaviors
/ emergency).
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
9
Model Dashboard View (SD Version)
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
10
Some EEA questions for the model
to assist in …
• What are the long-term effects of aircraft
OEM pricing decisions on the aviation
industry as a whole ?
– Should the extended enterprise include the
airlines?
– Will seat overcapacity make point-to-point / lowcost airlines more prevalent?
• What are the critical parameters that
determine the aviation industry’s reaction to
shocks (factors / demand, disruptive
technologies, competition) and how can they
change in the future ?
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
11
Value
• Applied:
A Tool for long-term Enterprise Architecting
Supporting the consideration of impacts in
strategic decision-making by
Experimenting with the performance of different
strategies in various scenarios (forecasting)
Testing hypotheses of agent behavior drivers
(customers or competitors) against their revealed
actions (strategic understanding)
• Methodological:
– contributions in the applied modeling of enterprises
by developing the ABM/SD hybridization concept
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
12
Questions ?
Photo credits: Robert and Shana Parkeharrison; www.parkeharrison.com
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
13
Selected References
Adler, N. and J. Berechman (2001). "Evaluating optimal multi-hub networks in
a deregulated aviation market with an application to Western Europe."
TRANSPORTATION RESEARCH PART A-POLICY AND PRACTICE
35(5): 373-390.
Aris, S. (2002). Close to the sun: how Airbus challenged America's domination
in the skies. London, Aurum.
Benkard, C. (2004). "A dynamic analysis of the market for wide-bodied
commercial aircraft." REVIEW OF ECONOMIC STUDIES 71(3): 581-611.
Esty, B. and P. Ghemawat (2002). "Airbus vs. Boeing in Super Jumbos: A
Case of Failed Preemption." Harvard Business School Strategy Working
Paper Series.
Gillett, D. (1994). Strategy in the Commercial Aircraft Industry in the United
States: A Comparison of Decisionmaking by McDonnell-Douglas and
Boeing Aircraft Companies from 1977-1983. Industrial College of the
Armed Forces. Fort McNair, Washington, D.C., National Defense
University.
Hansen, M. (1990). "Airline Competition in a Hub-Dominated Environment - An
Application of Noncooperative Game-Theory." TRANSPORTATION
RESEARCH PART B-METHODOLOGICAL 24(1): 27-43.
Jiang, H. H. and R. J. Hansman (2004). An Analysis of Profit Cycles In the
Airline Industry. M. Massachusetts Institute of Technology Cambridge, MIT
International Center for Air Transportation. Report No. ICAT-2004-7.
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
14
Selected References
Krugman, P. (Fall 1987). "Is Free Trade Passe?" Journal of Economic
Perspectives.
Liehr, M., A. Grossler, et al. (2001). "Cycles in the sky: understanding and
managing business cycles in the airline market." SYSTEM DYNAMICS
REVIEW 17(4): 311-332.
Lyneis, J. (2000). "System dynamics for market forecasting and structural
analysis." SYSTEM DYNAMICS REVIEW 16(1): 3-25.
Lynn, M. (1997). Birds of prey: Boeing vs. Airbus: a battle for the skies. New
York, Four Walls Eight Windows.
Miller, B. and J.-P. Clarke (2004). Application of Real Options to Evaluate the
Development Process of New Aircraft Models. AIAA 4th Aviation
Technology, Chicago, IL.
Niedringhaus, W. (2004). "The Jet: Wise model of national airspace system
evolution." SIMULATION-TRANSACTIONS OF THE SOCIETY FOR
MODELING AND SIMULATION INTERNATIONAL 80(1): 45-58.
Piepenbrock, T. F. (2004). Enterprise design for dynamic complexity:
architecting & engineering organizations using system & structural
dynamics. Dept. of Civil and Environmental Engineering. Leaders for
Manufacturing Program. Sloan School of Management. Cambridge, MA,
Massachusetts Institute of Technology.: 2 v. (341 leaves).
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
15
Summary:
• My intention is to understand and forecast under
different scenarios the long-term behavior of the
aviation industry.
• I propose to create a fairly detailed model of the
aviation industry as an extended enterprise with a
focus on [long-term] decision making from the primary
agents (passengers, airlines, leasing companies, and
aircraft manufacturers and their suppliers) taking into
account government policies.
• The methodology that I am proposing is an AB/SD
modeling hybrid that would illustrate the differences of
modular vs. integral enterprises. Inputs from other
disciplines are by necessity abundant.
• As a by-product of this research, I am expecting that my
methodology can be generalizable to model other
industries.
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
16
Back-up
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
17
Methodology Use in this Effort
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
18
Pricing Decisions for Aircraft
List prices seem to be an almost perfectly linear
function of capacity.
Regression can be used to provide the basic
pricing model: e.g. :
PR = -18.4 + 0.45CAP + 5.05RN
But the critical number is the discount offered
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
19
Intersecting Architectures
Aircraft Capacity
and Range (787
vs A380)
Network Type
(HS vs PtP)
R&D Financing
WTO Rulings
Push vs. Pull
Manufacturing
and Marketing
Preliminary
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
20
Enterprise Bounded rationality
Are enterprises rational agents?
We know that for persons and everyday decisions the
assumption of rationality does not hold true in most
cases. People tend to use heuristics to narrow their
choice set and even then when utility cannot be
defined exactly they may make random choices.
This should not be the case for enterprises, since the
timeframe is longer and the capacity to calculate
probabilities of outcomes for uncertain events is
larger. Yet (i) the decision makers in the latter case
are still human and (ii) imperfect information and
discounting of risk may prevail.
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
21
Modeling Approaches
Waiting for internal structure to develop
(emergence)
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
22
Passengers
Pax choice set (given OD) =
{travel, connections, airline, price-range}
Pax utility =
f {price, connections, frequency, class}
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
23
Airlines
Airline choice set () =
{aircraft, OD-pair, price-setting, network
type}
{capacity, range, engine, specific
consumption, manufacturer, lifecycle cost,
lifecycle flexibility?}
Airline utility (from aircraft) =
f {revenue, operational cost, capital cost,
reliability}
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
24
Aircraft Manufacturers
Aircraft OEM choice set () =
{introduce new model, extend existing model,
product family, aircraft price, OD-pair, pricesetting, network type}
{capacity, range, engine, specific consumption,
manufacturer, lifecycle cost, lifecycle flexibility?}
Aircraft OEM utility () =
f {R&D cost, financial backing, sales revenue,
production cost}
© 2006 Sgouris Sgouridis, Lean Aerospace Initiative, Massachusetts Institute of Technology
25