See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/227267121

Jay Wright Forrester

Chapter · June 2011

DOI: 10.1007/978-1-4419-6281-2_20 · Source: OAI

CITATIONS READS
24 15,799

2 authors:

David C. Lane John Sterman

University of Reading Massachusetts Institute of Technology
121 PUBLICATIONS 3,672 CITATIONS 184 PUBLICATIONS 34,138 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by David C. Lane on 03 June 2014.

The user has requested enhancement of the downloaded file.

  

Profiles  in  Operations  Research:  

Jay  Wright  Forrester  

 
David  C.  Lane  
John  D.  Sterman  
 
 
 
 
 
 
 
Lane,  David  C.  and  John  D.  Sterman  
(2011)    Jay  Wright  Forrester.    
Chapter  20  in  Profiles  in  Operations  
Research:    Pioneers  and  Innovators.  S.  
Gass  and  A.  Assad  (eds.).  New  York,  
Springer:  363-­‐386.  
   
 
 
 
Jay Wright Forrester

Jay Wright Forrester’s distinguished career at the Massachusetts Institute of Technology

(MIT) began with pioneering work in servomechanisms and digital computation.
Drawing on that work, he then developed and founded the field of system dynamics. By
integrating concepts of feedback control theory and digital computation, he created a new
approach to simulating the behavior of social systems, explaining that behavior and
designing effective policies to improve system performance. He established system
dynamics as an academic discipline at MIT’s Sloan School of Management. He led path-
breaking research applying system dynamics to critical business and public policy
problems, from high-tech start-ups to urban policy and global development. System
dynamics is now one of the most widely used systems approaches in the world, with
academics and practitioners on every continent pursuing work in diverse fields. His
writings continue to inspire the field and he remains actively involved in its development.

He is a Member of National Academy of Engineering and a Fellow of the Institute of

Electrical and Electronic Engineers (IEEE). His honors include: U.S. National Medal of
Technology; Medal of Honor, (IEEE); Pioneer Award, IEEE Aerospace and Electronic
Systems Society. He was inducted into the International Federation of Operational
Research Societies (IFORS) Operational Research Hall of Fame, and has honorary
degrees from nine universities.

Frontier Years and Beyond

Jay Wright Forrester was born on July 14, 1918 on a cattle ranch near Climax, Nebraska,
to Ethel Pearl Wright Forrester (1886-1958) and Marmaduke (Duke) Montrose Forrester
(1883-1975). Their daughter, Barbara Francis, was born in 1921. Both parents attended
Hastings College, Nebraska. They were the original homesteaders of this land close to the
American frontier. When they arrived in Nebraska around 1910, both worked as country
schoolteachers. Jay was taught at home by his mother for his first two years of schooling.
After that, he rode his horse one and a half miles to a one-room school house. There, for
the first two years, he was taught by his father.

Jay developed an early interest in electricity, tinkering with doorbells, batteries and
telegraphs. He recalls that being raised on a Nebraska cattle ranch offered plenty of
opportunities to get his hands dirty finding practical
solutions to real problems such as building a wind-powered
generator to provide the first electricity to the ranch
(Forrester 1992). He was offered a scholarship to an
agricultural college, but decided that the life bucolic was not
for him and, instead, enrolled in the University of Nebraska
to study electrical engineering.

After earning a bachelor’s degree in electrical engineering in

1939, Jay moved to MIT. He worked as a research assistant
with Gordon Brown, a pioneer in servomechanism theory
and applications (Brown and Campbell 1948). During World War II, Jay worked on
feedback control systems and servo-control systems for radar. For his master’s thesis, he
designed and built a servo to stabilize radar antennae on naval ships. In 1943, the
prototype was installed on the aircraft carrier Lexington and Jay subsequently traveled to
Pearl Harbor to ensure its continued functioning. Though a civilian, he volunteered to
stay on board when the fleet was ordered to sea to make sure the servo (and thus the
ship’s radar) worked. During the mission, the Lexington participated in the retaking of
the Marshall Islands and survived a torpedo strike. He received an S.M. degree in
Electrical Engineering from MIT in 1945; his thesis was titled “Hydraulic
Servomechanisms Developments.”

On the Lexington From 1940-1951, Jay was Associate

Director of MIT’s Servomechanism
Laboratory and then Head of the Digital
Computer Division in MIT’s Lincoln
Laboratory (1951-1956). He directed the
ASCA project (Airplane Stability and
Control Analyzer), aimed at developing
flight simulators to test new aircraft
designs. Originally envisioned as an
analog computer, Jay realized that the
real-time requirements of the planned
simulator could not be met with analog
components. Jay learned of digital
computation through MIT alumnus Perry
The hydraulic servo-mechanism to stabilize radar antennae built by Crawford. He then visited the computing
Forrester and Brown. The prototype was installed on the centers at Harvard and the University of
Lexington. This one is on display in the Forrester Conference
Room at the MIT Sloan School of Management. Photo: John Pennsylvania’s Moore School of
Sterman. Electrical Engineering where the ENIAC
was being built. There he met the
Princeton mathematician John von
Neumann, who was helping to design the Moore School’s next generation computer, the
EDVAC, and J. Presper Eckert, one of the developers of the ENIAC. These visits
convinced Jay that the ASCA project would be based on digital computation, a bold
decision given that all existing digital computers were far too slow and limited to meet
the requirements of ASCA. As director (from 1951) of the MIT digital computer
laboratory, Jay led the development of the Whirlwind computer, which was, for years, the
only machine fast enough for real-time simulation of complex dynamical systems such as
an aircraft.

Whirlwind became the central element of the SAGE (Semi-Automatic Ground

Environment) system and became the first computer produced in volume. SAGE was
built to defend North America from Soviet bomber attack and consisted of a network of
digital computers and long-distance communication systems that sent target tracking
information from radar stations to computers. The Whirlwind-based computers in each
center processed the data and computed flight plans for interceptor aircraft and missiles, a
demanding real-time application requiring high reliability (Jacobs 1986). With roughly
80,000 vacuum tubes in each of the approximately three dozen SAGE centers, reliability
was an immense technical challenge. Jay’s legendary drive for quality and reliability led
to design improvements and manufacturing standards yielding unprecedented results -
when the last of the SAGE centers was decommissioned in 1983, the system-wide uptime
over their roughly 25 years of service was 99.8%, making it probably the most reliable
military command and control system ever implemented. Jay’s colleagues and students
during this period went on to major accomplishments. Robert Everett, Jay’s second in
command in the Whirlwind project, went on to lead the MITRE Corporation (originally,
MIT Research and Engineering), created by the spinoff of the digital computer division
of MIT’s Lincoln Laboratory. Kenneth Olson, one of Jay’s students, went on to found
and lead the Digital Equipment Corporation.

Computer memory became a major bottleneck early in the development of Whirlwind.

Memory cost $1 per digital bit per month. Electrostatic tubes provided some storage, but
were expensive and unreliable. Mercury delay lines were tried, with the bits stored as
sound waves traveling down a tube of mercury (Slater 1987). Jay even considered leasing
a microwave relay line between Boston and Buffalo, New York, to store bits in a pulse
train of electromagnetic radiation. The need for fast, high-capacity data storage spurred
Jay to create coincident-current magnetic core memory in 1949 (Forrester 1951, 1953;
US Patent Office 1956). Core memory was cheap, stable, and reliable. It was the
industry standard for decades (Evans 1983) and was the memory device that flew to the
moon on the Apollo missions. For its invention, the IEEE awarded Jay its Medal of
Honor in 1972. In 1979, he was made a member of the US National Inventors’ Hall of
Fame, and, in 1989, along with Robert Everett, was awarded the Presidential Medal of
Technology.

Whirlwind

Whirlwind: Jay (centre, standing) and on his left Robert Everett at Whirlwind I test control in
1950. Used with the permission of The MITRE Corporation. Copyright © The MITRE
Corporation. All Rights Reserved.
National Medal of Technology Core Memory

In 1989, along with Robert Everett, Forrester received the National Medal
of Technology, the nation’s highest award for technical achievement. Photo:
George Bush Presidential Library and Museum. Coincident Coil Magnetic Core Memory: Jay
holding a 64x64 core memory plane, 1954.
Picture used with the permission of The MITRE
Corporation. Copyright © The MITRE
Corporation. All Rights Reserved.

As described in Slater (1987), Jay’s achievements during the period of 1940-1955 were
extraordinary (for Jay’s perspective, see Forrester 2007a). Jay’s experiences proved of
great value in the next chapter of his scientific life—the development of his major work,
the field of system dynamics and its application to critical problems in business and
public policy. Naturally, his life also had personal dimensions of no less significance. It
was early in this period that Gordon Brown introduced Jay to Susan Swett. They married
on July 27, 1946 and went on to have three children. In 1952, they moved into a brown-
shingled house in Concord, Massachusetts, their home until 2007.

The Emergence of System Dynamics

By the mid-1950s, Jay felt that “the pioneering days in computers were over” and, ever
seeking new frontiers, was looking for new challenges (Forrester 1992, 343). His work
with servomechanisms, digital computation, and SAGE had provided extensive
experience in the management of complex organizations and large-scale high technology
projects. He relates a conversation with the then-president of MIT James Killian, who

… brought a group of visiting dignitaries to see us at the

Lincoln Laboratory. While walking down the hall with
Killian, he told me of the new management school that
MIT was starting, and suggested that I might be interested.
The Sloan School of Management had been founded in
1952 with a grant of 10 million dollars from [MIT
alumnus] Alfred Sloan, the man who built the General
Motors Corporation. The money was given on the
 expectation that a management school in a technical
environment like MIT would probably develop differently
from one in a liberal arts environment like Harvard,
Columbia, or Chicago. Maybe better, but in any case
different, and it was worth 10 million dollars to run the
experiment (Forrester 2007a, 347).

Jay joined the Sloan faculty in 1956. He spent the first year considering what contribution
digital computation and control theory might make to management. A 1956 memo to the
faculty research seminar titled “Dynamic models of economic systems and industrial
organizations” laid out his initial thinking, and became the first in a series of D-Memos
(dynamic modeling memos). Jay organized an industrial dynamics group and he and its
members began to log their models, reports, class assignments, papers, and musings in
the D-memo series, which continued through the 1990s, when computers and the internet
made it unnecessary to keep such paper files. The D-Memos, now in the MIT archives,
are a remarkable record of the evolution of a new field (most are available on a DVD
distributed by the System Dynamics Society).

Jay’s first dynamic model arose through chance conversations with executives at the
General Electric Corporation (GE) (Forrester 1992, 2007a). GE managers were puzzled
by large fluctuations in production, inventories, employment, and profit. These
oscillations endured despite the managers’ best efforts, and were attributed to outside
forces, specifically business cycle fluctuations in incoming orders. By talking to the
managers and observing how the different departments were run, Jay elicited an account
of how individual managers, from the retail level, through distribution channels, to the
factories, responded to the information locally available to them as they tried to control
their piece of the organization. Rather than attribute the fluctuations to exogenous events,
he saw the production and distribution of appliances as a system of interacting units. The
managers in each link in what today is called a supply chain were responding in a locally
rational fashion to the incentives and information they faced; for example, the need to
provide good customer service while avoiding excessive inventories. The resulting
changes in orders, production, hiring, and other decisions then fed back to alter
inventories, backlogs, prices, and advertising, creating a system consisting of multiple
feedback loops, just as a servomechanism consisted of a closed-loop control system.
Managers at each link of the supply chain altered the orders they placed with suppliers to
compensate for variations in orders and inventories, just as his antenna stabilization servo
adjusted the position of the antenna to compensate for the pitch, roll, and yaw of the
Lexington. Where, however, the servo damped out the variations in the environment, the
feedback structure of the supply chain amplified them into persistent cyclical swings.

In building this first model, Jay retained several vital features of the situation, including
an explicit stock and flow network for resources such as inventories and labor, the long
time delays between actions and outcomes such as shipping and production delays, and
nonlinearities such as the impact of inventory on shipments and nonnegativity constraints
on production. Retaining these features meant that the system was not analytically
tractable. Simulation was required. Jay carried out the first simulation of this system by
hand, calculating production, shipments, hiring, and other flows from inventory, work in
progress, workforce, and other system states, then updating these stocks, week by
simulated week. The results, recorded in a lab notebook, showed how the management
policies of the firm generated robust oscillations even when demand was constant.
Although inventories and backlogs are intended to absorb temporary fluctuations in
orders so that costly production changes can be minimized, Jay found that the firm’s own
policies, sensible and rational from the perspective of the managers at each decision
point, led to substantial amplification of perturbations in orders, and instability for the
system as a whole, a phenomenon now known as the bullwhip effect.

Jay soon moved to computer simulations of this problem. Further work showed how
feedback control theory could be adapted to understand puzzling, counterintuitive
behavior in a range of management and human systems (Forrester 1956). He called the
approach industrial dynamics (Forrester 1958).

Industrial Dynamics

Jay recruited talented MIT undergraduates as research

assistants—Willard Fey, Jack Pugh, Edward Roberts and
others—and spent the next years developing applications
and laying out a vision of the contribution that they could
make to management (Forrester 1959, 1960). These ideas
led to his classic book, Industrial Dynamics (Forrester
1961).

As described in Richardson (1991), systems concepts

including feedback control, mutual causality, deviation-
correcting and deviation-amplifying processes were in
the air during the middle of the last century. Jay’s unique
contribution, detailed in Industrial Dynamics, was to
develop ideas about systems, feedback, control, and
dynamics that were previously restricted to engineering and physical contexts into a
rigorous yet practical method for enterprise design, a method designed to “find
management policies and organizational structures that lead to greater success” (Forrester
1961, 449). Richardson (1991) documents the connections and parallels between Jay and
other pioneers of cybernetics, systems theory, and operations research (OR). Many of the
principles Jay articulated to guide effective modeling and policy design for complex
systems had no precedent in the work of others in systems theory and OR, and were
revolutionary in their implications—on principles for modeling (Forrester 1960, 1968b,
1968c, 1987), on the design of corporations (Forrester 1965), and on the counterintuitive
behavior of social systems (Forrester 1971b).

Industrial Dynamics contributed a set of four principles for effective modeling of

complex systems: counter-intuitive system behavior is driven by system structure,
structure involves non-linear relationships, computer simulation is necessary to explore
behavior, and that applying the previous three ideas provides a rigorous yet pragmatic
way for managers to improve the design of organizations.

Jay’s first principle was that the puzzling (counter-intuitive) behavior of companies,
economies, indeed, all systems, whether physical, physiological, economic, or social,
emerged endogenously from their structure. That structure includes physical elements
such as stocks of inventory, labor, capital, order backlogs; information systems that
determined what information was available to each decision maker and the extent to
which that information is delayed, smoothed,
aggregated, biased, or otherwise corrupted by Policies and Decisions
processes of measurement, reporting, and
“… understanding of decision making has
subjective adjustment; and, most importantly, been greatly handicapped by the
by the policies and decision processes of the presumption that it is a more subtle and
actors at each decision point in the system. He more sophisticated process than it actually
stressed the importance of discovering and is…. It is my feeling that in a dynamic
representing the mental model of the decision information-feedback system the human
decision maker is usually using a great deal
maker. Similar to, but largely independent of, less than the total amount of information
the view of Herbert Simon and his colleagues available to him. Furthermore, the
(Simon 1957, Cyert and March 1963), Jay information available to him is a great deal
emphasized that effective models of human less than that commonly presumed. In
systems must capture the bounded rationality general, his actions with respect to any
given decision stream will be almost
of the agents’ decision processes. Decision entirely conditioned by less than ten
making should be represented in models as it information inputs.”
is, warts and all, and not presumed to be the
fully rational optimizing behavior of the (Forrester 1961, 100)
mythical homo economicus.

The interaction of the physical structure, information flows, and decision processes
creates a network of feedback loops that generates the dynamics of the system. People
use information about system states such as inventory, labor, order backlogs, and the
company’s reputation for service quality to make decisions; those decisions then
condition production, shipments, hiring, orders, and other rates of flow that alter the
system states. These processes form closed loops, some of which constitute control
processes (negative feedbacks) such as the loop whereby excessive inventories led firms
to cut production below shipments, thus lowering inventory levels. Some form self-
reinforcing processes (positive feedbacks), such as the loop whereby customers react to
an increase in supplier lead times by increasing their safety stock targets and ordering
farther ahead, actions that further deplete supplier inventories and swell order backlogs,
causing still longer delivery delays, a process known today as phantom ordering and one
that played a major role in the tech bubble of the late 1990s (Sterman 2000).

The second principle is that nonlinearity plays a central role in the dynamics of complex
systems. Jay knew from his experience with electromechanical systems that
nonlinearities decisively conditioned their structure and behavior. Jay saw that economic
and social systems were also intrinsically nonlinear and could not be adequately
approximated with linear methods. Structurally, nonlinearities abound in the real world.
Product shipments are generally determined by orders, until inventory is depleted, at
which point shipments necessarily fall to zero. Production increases with work hours up
to a point, then peaks and falls as fatigue cuts productivity, boosts errors, and triggers
accidents. Behaviorally, linear systems cannot exhibit locally unstable behavior and
global stability, cannot exhibit bifurcations, endogenous shifts in their modes of behavior,
and cannot evolve. Yet, with few exceptions, such as the Lotka-Volterra predator-prey
model (Murray 1989), OR, economics, and dynamical theory were dominated at that time
by linear models. Linear theory dominated because it was analytically tractable. Even
after the computer became widely available, nonlinearity was slow to penetrate these
disciplines. Jay designed his modeling method from the start to incorporate nonlinearities
easily and intuitively. Subsequent developments have shown the prescience of Jay’s
focus on nonlinearity (Strogatz 1994, Mosekilde 1996). Physical and social scientists
now recognize the intrinsically nonlinear character of physical, biological, and socio-
economic systems. Though the terms chaos and self-organization were unknown at the
time, Jay’s early models are among the first models of human behavior ever developed to
exhibit phenomena such as deterministic chaos, self-organization, and increasing returns.

Jay’s third principle, that simulation was needed to explore system behavior, led to the
development of a practical computer simulation methodology for business, economic and
social systems. He and his first students created tools for simulating complex human
systems that included diagramming conventions and general purpose computer
simulation languages. The first compiler for such simulations, developed by Richard
Bennett, was dubbed SIMPLE (Simulation of Industrial Management Problems with Lots
of Equations). SIMPLE was followed by DYNAMO (DYNAMic MOdeling), which
remained the standard for system dynamics modeling for several decades. Subsequently,
the personal computers and graphical user interfaces triggered rapid growth in the
number of software packages for dynamic modeling, including iThink, Powersim,
Vensim, and many others.

The fourth and perhaps most radical of Jay’s innovations was his focus on system
dynamics as both a rigorous tool to develop scientific knowledge and a practical tool to
improve the performance of organizations. He consistently argued that senior managers
should build models to understand and improve their organizations. He believes that a
manager’s role is not merely captain of the ship but designer of the ship (Keough and
Doman 1992). This corporate designer role was an innovative approach to both modeling
and management, one he has advocated throughout his career. To carry out rigorous
scientific modeling and develop models that made a difference required engagement with
the mental models of managers and other stakeholders; while understanding might
develop without the active participation of key decision-makers, implementation of new
policies could not. There are practical and theoretical links here with recent developments
in OR (Lane 1994, 1999). Jay’ focus on implementation and the need to engage the
decision makers in the modeling process underlies subsequent work on organizational
learning (Forrester 1971c, Senge 1990) and the development of protocols for group
model building (Lane 1992, Richmond 1997, Vennix 1996, Vennix, Richardson and
Andersen 1997).
Management laboratories for enterprise The model versus a modeling process
design
“In any real-life applications of modeling to the
“Industrial dynamics is the investigation of the generation of policy ... the models are always in a
information-feedback character of industrial continuous state of evolution. Each question, each
systems and the use of models for the design of reaction, each new input of information, and each
improved organizational form and guiding difficulty in explaining the model leads to modification,
policy. clarification, and extension.

“It is only through costly experience and errors “I believe we are proposing the 'Process' of modeling
that managers have been able to develop rather than particular frozen and final models. The
effective intuitive judgment. We need to difference in viewpoint becomes especially important
expedite this learning process. Other professions as we move into the implementation phase. It seems to
in similar circumstances have turned to me that the average person will be greatly concerned if
laboratory experiments .... Controlled laboratory he feels that the future and alternatives are being frozen
experiments on industrial and economic once and for all into a particular model. Instead, we are
situations are now possible with computers to do suggesting that models will help to clarify our
the work .... The manager, like the engineer, can processes of thought: they will help to make explicit
now have a laboratory in which he can learn the assumptions we are already making and they will
quickly and at low cost the answers that would show the consequences of the assumptions. But as our
seldom be obtainable from trials in real understanding, our assumptions, and our goals change,
organizations. so can the models.

“Industrial dynamics is an approach that should “Rather than stressing the single-model concept, it
help in important top-management problems ... appears that we should stress the process of modeling
The attitude must be one of enterprise design ... . as a continuing companion to, and tool, for, the
The goal should be to find management policies improvement of judgment and human decision
and organizational structures that lead to greater making.”
success.”
(Forrester 1971c)
(Forrester 1961; 13, 43, 449)

Developing a Discipline

Throughout the 1960s, Jay and his students applied system dynamics to a growing range
of problems through teaching, research, consulting, and practical management
applications (Roberts 1978a, Richardson 1996, Sterman 2007). Jay, as a member of the
original DEC board of directors, built a series of models examining the growth of high
technology start-ups and used them to inform his position on key issues facing the
company (Forrester 1964, 1968a, 1975). DEC became the second largest computer firm
in the world; much of its early success can be attributed to the policies that Jay—
informed by his models—advocated as a member of the board. The corporate growth
model included production capacity, inventory, and shipments, financial management,
product development and the other key tangible assets and processes. A key aspect of the
model, however, was its portrayal of intangible elements such as the knowledge, skills
and attitudes of managers, engineers, and salespeople; standards for product quality and
the pressures altering them; organizational routines for pricing and resource allocation;
and the ability of top management to project its goals throughout the organization
(Forrester 2007a). Such intangibles are now known as dynamic capabilities and constitute
an active focus of research in strategy (Teece, Pisano, and Shuen 1997). The story of
DEC’s later demise (long after Jay left the board) is a fascinating tale consistent with
many of the dynamics Jay described his early corporate growth models (Schein 2003).

Jay’s co-workers went on to contribute to the spread of these ideas. Will Fey taught
industrial dynamics as a professor at Georgia Institute of Technology. Ed Roberts became
the David Sarnoff Professor at the MIT Sloan School of Management, where he
contributed important work applying system dynamics to the management of technology,
health care, and public policy (Roberts 1978a, 1978b). He and Jack Pugh founded Pugh-
Roberts Associates, the first of many consulting firms to apply industrial dynamics.

Urban Dynamics
Towards the end of the 1960s, Jay and his students began to address public policy issues
and the more general term system dynamics replaced industrial dynamics. Urban
Dynamics (Forrester 1969) was a study of the processes underlying the development,
stagnation, decline, and recovery of cities. The project began when the mayor of Boston,
John F. Collins, chose not to run for a third term and became a visiting professor at MIT
with an office next to Jay’s. In the former mayor’s struggles with urban problems, Jay
recognized the same policy resistance and unintended consequences he had so often
observed in corporate contexts. He suggested to Collins that they develop a systems
dynamics model of the problem situation. Characteristically, the model was developed
not merely by reference to theory, but in conjunction with Collins and others with first-
hand experience managing large cities. The model endogenously generated the dynamics
of urban growth and stagnation over several hundred years. As the simulated city
evolved, population growth, crowding and aging of the housing stock and industrial base
gradually shifted the city from an engine of upward mobility to a poverty trap. The model
explained why so many policies implemented during the 1960s and 1970s to alleviate
urban poverty failed, and in some cases made the problems of the cities and their citizens,
particularly the poor, worse. Subsequent events have shown this analysis to be largely
correct, but it was enormously controversial at the time. The hostile reactions to the
work, as well as examples in which support arose from unexpected quarters, are
discussed in Forrester (1992). The key seemed to be spending sufficient time with the
model to understand its assumptions and the source of its dynamics and policy insights.
The account of Jay’s testimony to a U.S. House of Representatives sub-committee on
urban growth gives an idea of how he went about explaining his ideas; an edited version
was published as “Counterintuitive behavior of social systems” (Forrester 1971b).

Counterintuitive testimony to Congress

“The nation exhibits a growing sense of futility as it repeatedly attacks deficiencies in our social system
while the symptoms continue to worsen. Legislation is debated and passed with great promise and hope.
But many programs prove to be ineffective. Results often seem unrelated to those expected when the
programs were planned. At times programs cause exactly the reverse of desired results.
 “It is now possible to explain how such contrary results can happen. There are fundamental reasons why
people misjudge the behavior of social systems. There are orderly processes at work in the creation of
human judgment and intuition that frequently lead people to wrong decisions when faced with complex
and highly interacting systems.
“People would never attempt to send a space ship to the moon without first testing the equipment by
constructing prototype models and by computer simulation of the anticipated space trajectories. No
company would put a new kind of household appliance or electronic computer into production without
first making laboratory tests. Such models and laboratory tests do not guarantee against failure, but they
do identify many weaknesses which can then be corrected before they cause full-scale disasters.
“Our social systems are far more complex and harder to understand than our technological systems. Why,
then, do we not use the same approach of making models of social systems and conducting laboratory
experiments on those models before we try new laws and government programs in real life? The answer
is often stated that our knowledge of social systems is insufficient for constructing useful models. But
what justification can there be for the apparent assumption that we do not know enough to construct
models but believe we do know enough to directly design new social systems by passing laws and starting
new social programs?”
(Forrester 1971b, 52-53).

World Dynamics
In 1970, Jay began work with the Club of Rome to apply system dynamics to perhaps the
most important issues of social policy: the dynamics of global development. (“The Club
of Rome is independent of any political, ideological and religious interests. Its essential
mission is ‘to act as a global catalyst for change through the identification and analysis of
the crucial problems facing humanity and the communication of such problems to the
most important public and private decision makers as well as to the general public’ ”
(Club of Rome 2009). Jay developed a model capturing feedbacks among population,
natural resources, pollution, agricultural and industrial production, capital investment,
and quality of life.

The resulting book, World Dynamics (Forrester 1971a), posed sharp questions about the
relationship between growth and quality of life, generating heated discussion in popular
and scholarly forums worldwide. World Dynamics led to a more detailed modeling study
directed by Dennis Meadows, who had just received his Ph.D. under Jay and had joined
the MIT faculty. Described in The Limits to Growth (Meadows et al. 1972), that study
triggered worldwide controversy and debate (Forrester, Low, and Mass 1974). More
important, because the world models were fully documented and easily replicated, they
led to a wide range of critiques and extensions (Meadows, Meadows, and Randers 1992,
2004; Meadows, Richardson, and Bruckman 1982).
 Jay’s first sketch of the world model (German edition of World Dynamics).

With this stream of work, Jay launched the field of global modeling (de Steiguer 1997),
an activity that also provoked storms of criticism. His willingness to endure it seems to
stem from a refusal born of intellectual courage to deny the consequences of one’s
analysis. Whether, for example, asking hard questions about the role of churches in
influencing population growth (Forrester 1973), or questioning the assumption that
unending economic growth is both desirable and necessary (Forrester 1971a, 1971b).
Consequently, Jay drew the wrath of left and right, conservatives and progressives, often
at the same time. His work has been instrumental in shaping public thinking on the
interactions among the environment, development, pollution, and natural resources. The
economist, Paul Ormerod, commenting on the contribution of World Dynamics and
Limits to Growth, observed that its “true and lasting significance ... was the development
of a fundamentally different approach to understanding the workings of the economy to
that of orthodox economics” (Ormerod 1994, 36).

Developments in the more than 35 years since the publication of World Dynamics have
only underscored the importance of Jay’s insights. Perpetual growth of population and
material production on a finite planet is impossible. Long time delays in the response of
the economy and technology to resource scarcity and environmental degradation cause
human activity to overshoot the carrying capacity of the planet. Research, unavailable
when Jay formulated his world model, now shows clearly that humanity has already
overshot the global carrying capacity and is rapidly consuming and degrading the natural
capital stocks upon which our civilization depends, from groundwater to soils to fish
stocks to the climate (Wackernagel et al. 2002; Meadows, Meadows, and Randers 2004).
An article (Wall Street Journal 2008), headlined, “New limits to growth revive
Malthusian fears,” observed that, “the resource constraints foreseen by the Club of Rome
are more evident today than at any time since the 1972 publication of the think tank's
famous book, The Limits of [sic] Growth.” Yet Jay’s most important insight in World
Dynamics is not about how much oil remains in the ground, how much CO2 we can dump
into the atmosphere, or the potential for technology to find alternative energy sources or
reduce pollution. It is that there is no purely technical solution to the challenge of
creating a sustainable society. Technological innovation, market forces, and government
policies are all aimed at ameliorating the symptoms of stress—pushing back the limits to
growth by finding more energy, reducing greenhouse gas emissions, irrigating marginal
lands and designing new cultivars to boost food production, thus allowing growth to
continue until another limit is reached. In a series of email posts to the system dynamics
community discussion list (Forrester 2008), Jay stated “obvious and self-evident courses
of attacking symptoms rather than underlying causes will be futile….[T]reating one
symptom can unleash a different overwhelming reaction.” “[G]rowing population and
industrialization will overwhelm the short-term efforts if we do not restrain these forces
that are exceeding the carrying capacity of the earth.”

To The Present

System dynamics began to coalesce into an academic field in the 1970s. Programs were
started at universities in the U.S. and around the world. Conferences were organized and
textbooks written. The System Dynamics Society was created in 1983, with Jay its first
president. A dedicated journal was created, the modern form of which, The System
Dynamics Review, appeared in 1985. In the following year, IBM’s Thomas Watson, Jr.
endowed the Jay W. Forrester Chair in Management at MIT. In his book, The Fifth
Discipline, Peter Senge (1990) explored the relationship between system dynamics and
organizational learning, attracting a new generation of managerial interest. The MIT
System Dynamics Group continues research into a wide range of complex systems issues,
from organizational change to climate change. System dynamics is one of the most
popular electives at the MIT Sloan School of Management, attracting over 400 students
per year (compared to an MBA program of about 375 per year). Jay’s students and those
he inspired, including the authors of this profile, went on to found or lead academic
programs in system dynamics around the world (Sterman 2007).

Over the years, Jay remained active and began a large modeling study of economic
dynamics, which integrated endogenous accounts of business cycles, inflation and
stagflation, the growth of government, and the great waves of economic expansion and
depression (Forrester 1977, 1979, 1980; Forrester, Mass, and Ryan 1976; Sterman 1985).

Jay formally retired from the MIT Sloan School in 1989, an event which he said “has had
no effect whatsoever on my work” (Forrester, 1997). Ever focused on the high leverage
points to foster enduring change, he has for some years devoted most of his time to
catalyzing the education of young people in the principles of systems (Forrester 1990,
1993). Interest in the education of young people and how they could learn systems
thinking is a long-established area of application for system dynamics (Roberts 1978b). A
new wave of experiments to develop the systems thinking and modeling capabilities of
young people began in the late 1980s when a then-retired Gordon Brown introduced
Tuscon, Arizona middle school teacher Frank Draper to system dynamics. The
enthusiastic response of Draper and his students was the creation of the K-12 project and
the introduction of dynamic modeling in schools throughout the United States (Creative
Learning Exchange 2009).

Consistent with his early focus on engaging managers in the modeling process, Jay not
only believes young people should and can learn system dynamics and modeling, but
calls for a revolution in pedagogy as well. He believes that effective education requires
learner-directed learning in which teachers are not the source of answers but guides and
coaches who help learners develop the inquiry skills they will need to become systems
citizens (Creative Learning Exchange 2009). The theory of political and social change
these beliefs represent is fundamentally optimistic, hopeful and empowering. It is a view
that, if begun early enough, everyone can gain an appreciation for the complex dynamics
of natural and human systems, and then use that insight to design policies to create a
better world.

Legacy and the Next Frontier

In person, Jay is quiet, imposingly tall and faultlessly courteous. He speaks slowly and
confidently, producing analyses of a complexity seldom found in conversation. He is
direct and unambiguous with both praise and criticism. He is also often hospitable and
convivial, happy to enjoy a joke and quick to share humorous stories himself.

He speaks of his parents, Gordon Brown, and his wife Susan as those to whom he feels
most indebted. His discharging of this debt has produced work the legacy of which is
immense.

Along with other pioneers of computer science, Jay’s innovations in hardware, software,
and computer simulation ushered in the digital age. Simulation is now used routinely
throughout the natural and social sciences, hailed as a third branch of science, standing
alongside theory and experiment as a unique and vital method to advance human
knowledge (Pool 1992).

The field of system dynamics is healthy and growing. System dynamics is increasingly
used in corporations, government and other organizations. It is taught in a growing
number of universities and schools. It is applied to issues from organizational change to
climate change, from physiology to fiscal policy. On a lighter note, Urban Dynamics
inspired video game designer Will Wright to create SimCity (Seabrook 2006). Only a few
years ago, simulation was difficult, expensive, and scarce. Today children in elementary
school routinely create and manage simulated worlds of stunning complexity through
interactive computer games. Of course system dynamics is much more than a method for
computer simulation, more than mathematical models grounded in control theory and
nonlinear dynamics. It is also a practical tool policy makers use to help solve important
problems. It is qualitative and quantitative, hard and soft, a theoretical discipline and a
pragmatic approach for group modeling and policy design. Key concepts of system
dynamics, including feedback, counterintuitive behavior, limits to growth, nonlinearity,
tipping points, and many others are now integrated into the discourse of management,
social theory and everyday life. Discussions of critical public policy issues routinely refer
to unintended consequences and policy resistance. Scientists, policymakers and the media
discuss the many positive feedbacks that can cause runaway climate change and debate
whether we have passed the tipping point leading to irreversible melting of polar ice
sheets.

Yet Jay, ever questing, ever focused on the important problems, is not satisfied. Speaking
at the 2007 International System Dynamics Conference celebrating the 50th anniversary
of the founding of the field, Jay, rather than reviewing the achievements of a half-
century, challenged the field to move boldly into the next frontier, to tackle the most
important problems no matter their difficulty:

The first 50 years of system dynamics have established an

introduction to the field. We have shown the importance of
achieving a better understanding of complex systems in
nature and human affairs. Now, the field is on a plateau
ready to launch the next great thrust forward. … We are
now at about the same state of advancement that
engineering was when MIT first opened its doors in 1865.
… System dynamics started 50 years ago with academic
programs that focused on the outside world with emphasis
on major issues outside of academia. However, the
pressures inherent in academic institutions are driving our
field back into academic journals and away from the public
that we should be serving. … System dynamicists must go
behind the symptoms of trouble and identify the basic
causes. At first, such arguments will be met with disbelief,
scorn, and ridicule. To prevail, the battle must be sustained
until public understanding begins to change” (Forrester
2007b, 359-360, 370).

No one should be surprised. From the Sand Hills of Nebraska to the MIT servo-
mechanism laboratory, from the Marshall Islands to the dawn of the computer age, from
Industrial Dynamics to World Dynamics, from corporate boardrooms to elementary
school classrooms, Jay Wright Forrester has lived his entire life on the frontier.
 On His Way
On the Frontier

Jay Wright Forrester, 1940 © JWF

Jay at a celebration of the 100th anniversary of the MIT

Electrical Engineering Department in 2003. He stands on a
Segway, a nonlinear dynamic system of sensors, servos, and
real-time digital feedback controls, directly descended from
innovations Jay helped pioneer. Photo: Dan Bricklin. All rights
reserved.
Honors and Awards

Jay is a Member of National Academy of Engineering, a Fellow of the IEEE, the

American Academy of Arts and Sciences, Academy of Management, and the Royal
Society of Arts (London). His honors include: Medal of Honor, IEEE; Systems, Man,
and Cybernetics Award for Outstanding Accomplishment, IEEE; Howard N. Potts
Medal, The Franklin Institute; Inventor of the Year, George Washington University; U.S.
National Medal of Technology; Pioneer Award, IEEE Aerospace and Electronic Systems
Society; Valdemar Poulsen Gold Medal, Danish Academy of Technical Sciences. He
was inducted into the IFORS’ Operational Research Hall of Fame (Lane 2006). Jay has
received honorary Doctorate of Engineering degrees from the University of Nebraska,
Newark College of Engineering, and the University of Notre Dame; and honorary
Doctorate of Science degrees from Boston University and Union College. He also
received honorary doctorate degrees from the University of Mannheim (Political
Science), State University of New York (Humane Letters), University of Bergen (Dr.
Philosophy), and the Universidad de Sevilla (Honoris Causa).

David C. Lane
John D. Sterman

References

Brown, G. S., D. P. Campbell. 1948. Principles of Servomechanisms. John Wiley &

Sons, New York.

Club of Rome. 2009. www.clubofrome.org (viewed February 19, 2009).

Creative Learning Exchange. 2009. http://clexchange.org (viewed February 19, 2009).

Cyert, R., J. March. 1963. A Behavioral Theory of the Firm. Prentice Hall, Englewood
Cliffs, New Jersey. (2nd ed., 1992, Blackwell, Cambridge, Massachusetts).

de Steiguer, J. E. 1997. The Age of Environmentalism. McGraw-Hill, London.

Evans, C. 1983. Conversation: Jay W. Forrester. Annals of the History of Computing 5(3)
297-301.

Forrester, J. W. 1951. Digital information storage in three dimensions using magnetic

cores. Journal of Applied Physics 22(1) 44-48.
Forrester, J. W. 1953. Coincident-current magnetic computer memory developments at
M.I.T. J. C. Chu, ed. Argonne National Laboratory Computer Symposium, August 3-5,
1953 . Argonne National Laboratory, Lemont, Illinois,150-158.

Forrester, J. W. 1956. Dynamic Models of Economic Systems and Industrial

Organizations. Note to the Faculty Research Seminar, MIT System Dynamics Group D-
Memo 0 (5 November). Reprinted as Forrester, J. W. 2003. System Dynamics Review
19(4) 331-345.

Forrester, J. W. 1958. Industrial Dynamics: A major breakthrough for decision makers.

Harvard Business Review 36(4) 37-66.

Forrester, J. W. 1959. Advertising: A problem in industrial dynamics. Harvard Business

Review 37(2) 100-110.

Forrester, J. W. 1960. The impact of feedback control concepts on the management

sciences (in Forrester1975, 45-60).

Forrester, J. W. 1961. Industrial Dynamics. MIT Press: Cambridge, Massachusetts.

Forrester, J. W. 1964. Modeling the dynamic processes of corporate growth. Proceedings

of the IBM Scientific Computing Symposium on Simulation Models and Gaming. 23-42.

Forrester, J. W. 1965. A New Corporate Design. Industrial Management Review (now the
Sloan Management Review) 7(1) 5-17.

Forrester, J. W. 1968a. Market growth as influenced by capital investment. Industrial

Management Review 9(2) 83-105.

Forrester, J. W. 1968b. Industrial Dynamics - After the first decade. Management Science
14(7) 398-415.

Forrester, J. W. 1968c. Principles of Systems. MIT Press, Cambridge, Massachusetts.

Forrester, J. W. 1969. Urban Dynamics. MIT Press, Cambridge, Massachusetts.

Forrester, J. W. 1971a World Dynamics. Pegasus Communications, Waltham,

Massachusetts.

Forrester, J. W. 1971b. Counterintuitive behavior of social systems, Technology Review

73(3) 52-68.

Forrester, J. W. 1971c. “The” model versus a modeling “process.” Dept. memo D1621-1,
system dynamics group, MIT. Reprinted 1985 System Dynamics Review 1(1) 133-134.
Forrester, J. W. 1973. Churches at the transition between growth and world equilibrium.
D. L. Meadows, D. H. Meadows, eds. Towards Global Equilibrium. Wright-Allen Press,
Cambridge, Massachusetts, 337-353.

Forrester, J. W. 1975. Collected Papers of Jay W. Forrester. Wright-Allen Press:

Cambridge, Massachusetts.

Forrester, J. W. 1977. Growth Cycles. De Economist 125(4) 525-543.

Forrester, J. W. 1979. An alternative approach to economic policy: Macrobehavior from

microstructure. N. M. Kamrany, R. H. Day, eds. Economic Issues of the Eighties. Johns
Hopkins University Press, Baltimore Maryland, 80-108.

Forrester, J. W. 1980. Information sources for modeling the national economy. Journal of
the American Statistical Association 75(371) 555-566.

Forrester, J. W. 1990. System Dynamics - Adding structure and relevance to pre-college

education. K. R. Manning, ed. Shaping the Future. MIT Press: Cambridge,
Massachusetts, 118-131.

Forrester, J. W. 1992. From the ranch to system dynamics. A. G. Bedeian, ed.

Management Laureates: A collection of autobiographical essays (vol. 1). JAI Press,
Greenwich, Connecticut, 335-370.

Forrester, J. W. 1993. System dynamics as an organizing framework for pre-college

education. System Dynamics Review 9(2) 183-194.

Forrester, J. W. 1997. Personal Communication.

Forrester, J. W. 2007a. A recollection of the history and reflections for the future at the
golden anniversary of the field. System Dynamics Review 23(2-3) 345–358.

Forrester, J. W. 2007b. System dynamics—the next fifty years. System Dynamics Review
23(2-3) 359-370.

Forrester, J. W. 2008. Posting to online System Dynamics Mailing List:

www.vensim.com/sdmail/sdmail.html, 7 April, SD 6994, and 14 April, SD 7034.

Forrester, J. W., G. W. Low, N. J. Mass. 1974. The debate on world dynamics: a response
to Nordhaus. Policy Sciences 5(2) 169-190.

Forrester, J. W., N. J. Mass, C. J. Ryan. 1976. The system dynamics national model:
Understanding socio-economic behavior and policy alternatives. Technological
Forecasting and Social Change 9(1-2) 51-68.
Jacobs, J. F. 1986. The SAGE Air Defense System: A Personal History. MITRE
Corporation, Cambridge, Massachusetts.

Keough, M., A. Doman. 1992. The CEO as organization designer: An interview with
Professor Jay W. Forrester, the founder of system dynamics. The McKinsey Quarterly
(2)3-30.

Lane, D. C. 1992. Modeling as learning: A consultancy methodology for enhancing

learning in management teams. European Journal of Operational Research 59(1) 64-84.

Lane, D. C. 1994. With a little help from our friends: How system dynamics and “soft”
OR can learn from each other. System Dynamics Review 10(2-3) 101-134.

Lane, D. C. 1999. Social theory and system dynamics practice. European Journal of
Operational Research 113(3) 501-527.

Lane, D. C. 2006. IFOR’s Operational Research Hall of Fame: Jay Forrester.

International Transactions in Operational Research 13(5) 483-492.

Meadows, D. H., D. L. Meadows, J. Randers, W. Behrens. 1972. The Limits to Growth.

New York, Universe Books.

Meadows, D.H., D. L. Meadows, J. Randers. 1992. Beyond the Limits. Chelsea Green
Publishing Company, Post Mills, Vermont.

Meadows, D. H., D. L. Meadows, J. Randers. 2004. Limits to Growth: The Thirty-Year

Update. Chelsea-Green, White-River Junction, Vermont.

Meadows, D. H., J. Richardson, G. Bruckmann. 1982. Groping in the Dark. John Wiley
and Sons, Chichester, England.

Mosekilde, E. 1996. Topics in Nonlinear Dynamics: Applications to Physics, Biology and

Economic Systems. World Scientific, Singapore.

Murray, J.D. 1989. Mathematical Biology. Springer, Berlin.

Ormerod, P. 1994. The Death of Economics. Faber & Faber, London.

Pool, R. 1992. The third branch of science debuts. Science 256 (April 3) 44-47.

Richardson, G. 1991. Feedback Thought in Social Science and Systems Theory.

University of Pennsylvania Press, Philadelphia.

Richardson, G. 1996. Modelling for Management (2 vols.). Dartmouth Publishing

Company, Aldershot, England.
Richmond, B. 1997. The strategic forum: Aligning objectives, strategy and process.
System Dynamics Review 13(3) 131-148.

Roberts, E., ed. 1978a. Managerial Applications of System Dynamics. Pegasus

Communications, Waltham, Massachusetts.

Roberts, N. 1978b. Teaching dynamics feedback systems thinking: an elementary view.

Management Science 24(8) 836-843.

Seabrook, John 2006. The Game Master. The New Yorker. November 6.
www.newyorker.com/archive/2006/11/06/061106fa_fact (viewed March 14, 2008.)

Schein, E. 2003. DEC is Dead, Long Live DEC: The lasting legacy of the Digital
Equipment Corporation. Berrett-Koehler, San Francisco.

Senge, P. M. 1990. The Fifth Discipline: The Art and Practice of the Learning
Organization. Doubleday/Currency, New York.

Simon, H. 1957. Administrative Behavior; a Study of Decision-Making Processes in

Administrative Organizations, 2nd ed. New York, Macmillan.

Slater, R. 1987. Portraits in Silicon. MIT Press, Cambridge, Massachusetts.

Sterman, J. D. 1985. A behavioral model of the economic long wave. Journal of

Economic Behavior and Organization 6(1) 17-53.

Sterman, J. D. 2000. Business Dynamics: Systems Thinking and Modeling for a Complex
World. Irwin/McGraw-Hill, New York.

Sterman, J. D. ed. 2007. Exploring the next frontier: System Dynamics at 50. Special
issue of System Dynamics Review 23 (2-3).

Strogatz, S. 1994. Nonlinear Dynamics and Chaos. Addison-Wesley, Reading,

Massachusetts.

Teece, D., G. Pisano, A. Shuen 1997. Dynamic capabilities and strategic management.
Strategic Management Journal 18(7) 509-533.

U.S. Patent Office. 1956. Multicoordinate Digital Information Storage Device, U.S.
Patent No. 2,736,880. U.S. Patent Office, Washington D.C.

Vennix, J. A. M. 1996. Group Model-building: Facilitating team learning using system

dynamics. Wiley, Chichester.

Vennix, J., G. Richardson, and D. Andersen (eds.) 1997. Group model building., System
Dynamics Review (special issue) 13(2).
 Wackernagel, M., N. B. Schulz, D. Deumling, A. C. Linares, M. Jenkins, V. Kapos, C.
Monfreda, J. Loh, N. Myers, R. Norgaard, and J. Randers. 2002. Tracking the ecological
overshoot of the human economy, Proceedings of the National Academy of Sciences
99(14) 9266-9271.

Wall Street Journal. 2008. New limits to growth revive Malthusian fears. (March 24,
2008).

Acronyms

ASCA Airplane Stability and Control Analyzer

EDVAC Electronic Discrete Variable Computer

ENIAC Electronic Numerical Integrator and Computer

D-Memos Dynamic Modeling Memos

DEC Digital Equipment Corporation

DYNAMO DYNAMic MOdeling

IEEE Institute of Electrical and Electronics Engineers

GE General Electric Corporation

MIT Massachusetts Institute of Technology

OR Operations Research/Operational Research

SAGE Semi-Automatic Ground Environment

SIMPLE Simulation of Industrial Management Problems with Lots of Equations

View publication stats