| |
Videos
Completed
Dissertations
ESD Working Paper Series
Engineering
Systems Symposia
Brunel Lecture
Series
Miller Lecture
Series
ESD
OpenCourseWare
|
|
List
of Papers for 2003:
(in reverse chronological order)
ESD-WP-2003-10-Epistemology
in Engineering Systems
by
Prof. Daniel D. Frey,
Massachusetts Institute of Technology
The engineering
systems division at MIT has adopted an official vision statement
-- “ESD will be a leader in understanding, modeling, predicting
and affecting the structure and behavior of technologically enabled
complex systems.” To fulfill this vision, I think it is
worthwhile for ESD faculty to reflect on epistemology and its
relationship to engineering systems. Epistemology is the branch
of philosophy concerned with the nature of knowledge, justification,
evidence, and related notions. By reflecting upon epistemology,
we may clarify in our own minds how we come to know something
about engineering systems and thereby improve our research methods.
In this white paper, I pose five questions related to epistemology
and engineering systems. I also discuss possible answers, but
my goal was primarily to spark discussion rather than solidify
a position.
>
top
ESD-WP-2003-09-ITS:
What We Know Now that We Wish We Knew Then: A Retrospective on
the ITS 1992 Strategic Plan
by
Prof. Joseph M.
Sussman, Massachusetts Institute of Technology
From September
1991 until June 1992, a core writing team, which included the
author, worked on what was the first Intelligent Transportation
Systems (ITS) strategic plan in the United States. This plan was
entitled, "A Strategic Plan for IVHS in the United States."
It served to define the ITS program at a national scale in a way
that has been characterized as seminal.
The plan,
by most accounts, served as the blueprint for the early development
of ITS in the U.S. and as the basis for the subsequent plans produced
by ITS America, the federal government, various states, and a
number of private-sector organizations.
This paper
explores numerous aspects of ITS retrospectively, contrasting
views from 11 years ago, when the Strategic Plan was produced,
with the current reality. Areas discussed include Advanced Traveler
Information Systems (ATIS), Advanced Transportation Management
Systems (ATMS), reliability, getting the ITS program off the ground
in the early 90s, strategic use of information, automated network
management, electronic toll collection (ETC), congestion pricing,
architecture, commercial vehicle operations (CVO), Advanced Public
Transportation Systems (APTS), and regions.
The paper
closes by comparing ITS with the Interstate, and finally by discussing
the upcoming reauthorization of the Transportation Efficiency
Act for the 21st Century (TEA-21) and what has been learned through
this retrospective about ITS-related issues on that reauthorization.
>
top
ESD-WP-2003-08-Lean
Transformation in the U.S. Aerospace Industry: Appreciating Interdependent
Social and Technical Systems
Lean practices
and principles build on a half-century of successive initiatives
aimed at transforming social and technical systems in organizations.
While they are seen as central to the revitalization of the U.S.
aerospace industry, there is great variation in the degree to
which lean initiatives emphasize just technical/manufacturing
systems versus additional social and enterprise dimensions. Based
on a national random sample survey of 362 U.S. aerospace facilities,
this paper examines factors that account for the incidence of
lean practices and the impact on outcomes relevant to key stakeholders.
While structural factors such as industry sector, facility size
and others have limited explanatory power, two process factors—organizational
learning and the value placed on intellectual capital —do
account for the increased presence of lean practices. In examining
employment outcomes, facilities higher just on the technical/manufacturing
aspects of lean have a significant and negative impact on job
growth, while facilities higher around the social systems associated
with lean have significant and positive employment growth. This
finding is consistent with the views of critics of the more narrow
technical, manufacturing-oriented approaches to lean as a threat
to employment and it validate proponents of a broader value-creating
approach to lean as a way of growing the enterprise. Enterprise
dimensions of lean (including both social and technical aspects
of lean) have a positive impact on productivity. Examining outcomes
relevant to multiple stakeholders and various factor inputs produces
a more complete understanding of the limitations and potential
for lean transformation in the aerospace industry.
>
top
ESD-WP-2003-07-A
Proposal to Improve the Health Care Systems for the Urban Poor
in the Squatter Settlements of the Developing Countries
by
Prof. Richard Larson
and Nebibe Varol, Massachusetts Institute of Technology
Rapid
urbanization and large scale population movements from rural to
urban areas have resulted in unprecedented health crises in the
developing countries. In addition to communicable diseases, respiratory
infections and malnutrition, psycho-social stresses due to marginalization
and exclusion from social activities and employment prospects
are also prevalent. Considering the rate of urban growth rate
and the rapid increase in the percentage of the poor living in
urban areas, the debilitating effects of health crises and urban
poverty are going to exacerbate if no precautions are taken. In
this respect, it is a critical point in time to come up with effective
health care strategies for the urban poor. This document provides
an insight into the reasons behind the current health problems
of the urban poor and the determinants of health in developing
countries, and proposes use of operations research to come up
with handling strategies for the major subdivisions of the health
problem in the developing world.
>
top
ESD-WP-2003-06-Role
of Technology in Manufacturing Competitiveness
by
Professor Thomas Eagar,
Thomas Lord Professor of Materials Engineering and Engineering
Systems, Massachusetts Institute of Technology, Christopher Musso,
Engineering Systems Division, Massachusetts Institute of Technology
A manufacturing
revolution has emerged in the past 50 years that is as significant
as the industrial revolution of the 19th century. From 1950 to
2000, the average productivity growth in manufacturing in the
United States was 2.8% per year, and this figure has been accelerating
for the past two decades as manufacturing productivity growth
has exceeded
the average of other sectors by more than one percent per year
(please see table below). Stated more simply, a US manufacturing
worker can produce four times as much per hour today as compared
with fifty years ago. This gain has resulted from competitive
pressures, the advent of new technologies, and a series of product
and process innovations. It has also resulted in a much higher
standard of living for Americans, as products become more useful
and more affordable. In order to utilize this new manufacturing
capacity, U.S. firms (and others) have expanded their marketing
abroad, creating rapid increase in global trade.
The perception
of a crisis in American manufacturing is the result of one of
the most difficult realities of large gains in productivity: additional
capacity almost always exceeds increased consumption. This results
in an inevitable shift of labor. Industries become more productive
as they mature, and competitive pressures increase. These two
factors require companies to decrease their workforce and often
result in movement of commodity industries overseas. The end result
is a loss of jobs in the United States. Displaced workers must
shift to new occupations, requiring new skills and abilities.
History has shown that this shift can be either detrimental or
beneficial to workers; the most important determinant of benefit
is the presence of innovative new industries, which, create high
value for their markets. The sustainability of growth in the U.S.
manufacturing sector is based on the ability of America to continue
to innovate. Innovation is the key to a vibrant U.S. manufacturing
base and continued generation of new jobs.
>
top
ESD-WP-2003-05-How
Useful is Quantitative Risk Assessment?
by
Professor George
E. Apostolakis, Massachusetts Institute of Technology
This article
discusses the use of Quantitative Risk Assessment (QRA) in decision-making
regarding the safety of complex technological systems. The insights
gained by QRA are compared with those from traditional safety
methods and it is argued that the two approaches complement each
other. It is argued that peer review is an essential part of the
QRA process. The importance of risk-informed rather than risk-based
decision-making is emphasized. Engineering insights derived from
QRAs are always used in combination with traditional safety requirements
and it is in this context that they should be reviewed and critiqued.
Examples from applications in nuclear power, space systems, and
an incinerator of chemical agents are given to demonstrate the
practical benefits of QRA. Finally, several common criticisms
raised against QRA are addressed.
>
top
ESD-WP-2003-04-Needs
and Possibilities for Engineering Education: One Industrial/Academic
Perspective
by
Christopher L. Magee,
Professor of the Practice of Mechanical Engineering and Engineering
Systems, Massachusetts Institute of Technology
This paper
reports a personal assessment of the readiness of new B.S. level
engineering graduates to practice engineering immediately upon
graduation. This assessment when reinforced by significant prior
work motivates a systemic analysis of the U.S. Engineering Education
System. The analysis is framed to address the implementation potential
of ideas for how educators might efficiently teach undergraduate
engineers “that engineering is more than differential equations”.
The concepts which seem best from this analysis are combinations
of aggressive intern opportunities combined with courses (starting
in the freshman year) that emphasize the creative engineering
process. These activities may be containable in the 4 year program
but the analysis also suggests that extension of engineering education
to 3 or more years beyond the B.S. would improve the possibility
of reaching key educational goals including teaching adequate
math and science fundamentals as well as engineering knowledge,
process and creativity. Such radical change will be difficult
and slow to occur (if at all) in this complex system. Moreover,
this system is understandingly resistant to change because of
significant perceptions of outstanding achievement. The driving
force for change that may be strong enough to overcome these barriers
is prospective students’ falling perceptions of engineering
education as a preferred option.
>
top
ESD-WP-2003-03-Metrics
Pilot Project for Military Avionics Sustainment: Experimental
Design and Implementation Plan
by
Kirkor Bozdogan, MIT Co-Lead of the Enterprise Integration Team
(EIT) of the LEAN SUSTAINMENT INITIATIVE (LSI), Benjamin M. Brandt
, Capt. Brandt (USAF) is a graduate student Research Assistant
and Candidate for the MS Degree in Technology and Policy at MIT,
Joseph M. Sussman,
MIT Co-Lead of the Enterprise Integration Team (EIT) of the LEAN
SUSTAINMENT INITIATIVE (LSI)
This working
paper outlines the design of an experiment, employing a pilot
project, for identifying and validating new metrics for managing
the US Air Force military avionics sustainment system. The paper
also presents a plan for implementing the pilot project. The experimental
design allows for the quantitifation of the effects of the new
metrics, while controlling for the effects of other factors impacting
the observed outcomes.
Underlying
the pilot project, and the proposed experimental design, are three
main hypotheses derived from earlier research: (a) currently used
metrics foster local optimization rather than system-wide optimization;
(b) they do not allow measures of progress towards the achievement
of system-wide goals and objectives, and, hence, do not allow
visibility into the impact of depot maintenance on the warfighter;
and (c) they are driving the “wrong behavior,” causing
suboptimal decisions governing maintenance and repair priorities
and practices and, as a result, undermining the efficiency and
effectiveness of the sustainment system, despite the fact that
the Air Force sustainment system has a dedicated and highly skilled
workforce supporting the warfighter..
>
top
ESD-WP-2003-02-Applying
STAMP in Accident Analysis
by
Nancy Leveson,
Mirna Daouk, Nicolas Dulac, and Karen Marais, Massachusetts Institute
of Technology
Accident
models play a critical role in accident investigation and analysis.
Most traditional models are based on an underlying chain of events.
These models, however, have serious limitations when used for
complex, socio-technical systems. Previously, Leveson proposed
a new accident model (STAMP) based on system theory where the
basic concept is not an event but a constraint. This paper shows
how STAMP can be applied to accident analysis using three different
views or models of the accident process and proposes a notation
for describing this process.
>
top
ESD-WP-2003-01.01-ESD
Internal Symposium: Incorporating Uncertainty Into
Conceptual Design of Space System Architectures
The environment
in which space systems are developed and operated can be classified
as nothing less than dynamic. However, it is clear that the methods
and tools relied on in conceptual design are based on static assumptions
and leave little room for anything more than snapshots of the
product and its environment. This paper introduces an approach
to challenge that model and instead quantify and compare space
system architectures around the central theme of uncertainty,
with emphasis on policy uncertainty, as well as, technical and
market uncertainty. Two cases of implementation are presented
and three generalized principles are proposed that flow from the
analysis: 1) engineering systems must be designed with uncertainty
as one of the central organizing principles, 2) since engineering
systems have management and social dimensions and thus involve
human interactions, there is an irreducible uncertainty associated
with these dimensions that will affect the design of the system,
and 3) uncertainty in use may allow the engineering system to
satisfy quite different missions from the original one intended.
>
top
ESD-WP-2003-01.02-ESD
Internal Symposium: An Attempt at Complex System Classification
This paper
searches for a useful taxonomy or classification scheme for complex
Systems. There are two aspects to this problem: 1) distinguishing
between Engineering Systems of interest to ESD (ES) and other
Systems, and 2) differentiating among Engineering Systems. The
first of these has been approached through general interaction
with other ESD faculty and use of the ESD definitions. This analysis
leads to a proposed specific set of ES which are human designed,
have high technical and human complexity and are real, open, dynamic,
have hybrid system states and have both autonomous and human-in–the
loop subsystems or elements.
The second
aspect has been approached by top-down and bottom-up analysis.
A topdown approach consists of reviewing past system classification
schemes starting with taxonomies proposed in the context of General
Systems Theory from the 1950’s and assessing their usefulnesswith
the proposed list of ES. Such schemes prove to be of limited value
in our search because they tended to be formulated from a mechanical
technology viewpoint and more importantly because they could not
anticipate the emphasis herein on systems with both technical
and human complexity.
The proposed
or testbed list is also useful in the bottom-up approach, since
it gives specific cases for qualitative and quantitative analysis
of various system attributes. The qualitative and preliminary
quantitative study indicates that functional types are the most
useful technical attribute for classification differentiation.
Information, energy, value and mass acted upon by various processes
are the foundation of the technical types building on prior work
byHubka, Pahl and Beitz and Van Wyk.
A meta-model
for Engineering Systems is suggested in the form of a multi-layer
network whose goal it is to fulfill human wants and needs by enabling
the flow of goods and services between sources and sinks. This
description essentially combines and extends the attributes suggested
by the bottom-up approach to be most useful in classification.
>
top
ESD-WP-2003-01.03-ESD
Internal Symposium: Physical Limits to Modularity
Architecture,
specifically the definition of modules and their interconnections,
is a central concern of engineering systems theory. The freedom
to choose modules is often taken for granted as an essential design
decision. However, physical phenomena intervene in many cases,
with the result that 1) designers do not have freedom to choose
the modules, or 2) that they will prefer not to subdivide their
system into as small units as is possible.
A distinction that separates systems with module freedom from
those without seems to be the absolute level of power needed to
operate the system. VLSI electronics exemplify the former while
mechanical items like jet engines are examples of the latter.
It has even been argued that the modularity of VLSI should be
extended to mechanical systems. This paper argues that there are
fundamental reasons, that is, reasons based on natural phenomena,
that keep mechanical systems from approaching the ideal modularity
of VLSI. The argument is accompanied by examples.
>
top
ESD-WP-2003-01.04-ESD
Internal Symposium: The Effect of e-Business on Supply
Chain Strategy
Internet
technology has forced companies to redefine their business models
so as to improve the extended enterprise performance - this is
popularly called e-business. The focus has been on improving the
extended enterprise transactions including Intraorganizational,
Business-to-Consumer (B2C) and Business-to-Business (B2B) transactions.
This shift in corporate focus allowed a number of companies to
employ a hybrid approach, the Push-Pull supply chain paradigm.
In this article we review and analyze the evolution of supply
chain strategies from the traditional Push to Pull and finally
to the hybrid Push-Pull approach. The analysis motivates the development
of a framework that allows companies to identify the appropriate
supply chain strategy depending on product characteristics. Finally,
we introduce new opportunities that contribute and support this
supply chain paradigm.
>
top
ESD-WP-2003-01.05-ESD
Internal Symposium: Patterns of Product Development
Interactions
Development
of complex products and large systems is a highly interactive
social process involving hundreds of people designing thousands
of interrelated components and making millions of coupled decisions.
Nevertheless, in the research summarized by this paper, we have
created methods to study the development process, identify its
underlying structures, and critique its operation. In this article,
we introduce three views of product development complexity: a
process view, a product view, and an organization view. We are
able to learn about the complex social phenomenon of product development
by studying the patterns of interaction across the decomposed
elements within each view. We also compare the alignment of the
interaction patterns between the product, process, and organization
domains. We then propose metrics of product development complexity
by studying and comparing these interaction patterns. Finally,
we develop hypotheses regarding the patterns of product development
interactions, which will be helpful to guide future research.
* This paper
also appeared in the proceedings of the International Conference
on Engineering Design, Glasgow, Scotland, August 2001.
>
top
ESD-WP-2003-01.06-ESD
Internal Symposium: Collected Views on Complexity
in Systems
by Joseph
M. Sussman, JR East Professor, Professor of Civil and Environmental
Engineering and Engineering Systems, Massachusetts Institute of
Technology, Cambridge, Massachusetts
The term
complexity is used in many different ways in the systems domain.
The different uses of this term may depend upon the kind of system
being characterized, or perhaps the disciplinary perspective being
brought to bear.
The
purpose of this paper is to gather and organize different views
of complexity, as espoused by different authors. The purpose of
the paper is not to make judgments among various complexity definitions,
but rather to draw together the richness of various intellectual
perspectives about this concept, in order to understand better
how complexity relates to the concept of engineering systems.
I
have either quoted directly or done my best to properly paraphrase
these ideas, apologizing for when I have done so incorrectly or
in a misleading fashion. I hope that this paper will be useful
as we begin to think through the field of engineering systems.
The
paper concludes with some short takes -- pungent observations
on complexity by various scholars -- and some overarching questions
for subsequent discussion.
>
top
ESD-WP-2003-01.07-ESD
Internal Symposium: The Concept of a CLIOS Analysis
Illustrated by the Mexico City Case
by Rebecca
Dodder, Doctoral Candidate, Technology, Management & Policy
Program and Joseph
M. Sussman, JR East Professor, Professor of Civil and Environmental
Engineering and Engineering Systems
The term
CLIOS (Complex, Large-scale, Integrated, Open Systems) was conceived
as way to capture the salient characteristics of a class of systems
that are of growing interest to researchers, decisionmakers, policy
makers and stakeholders. These systems range from an air traffic
control system to the global climate system, and from Boston’s
Big Dig to the eBay online trading system.
We start
by defining the primary characteristics of CLIOS. First, a system
is complex when it is composed of a group of interrelated units
(component and subsystems), for which the degree and nature of
the relationships is imperfectly known – with varying directionality,
magnitude and time-scales of interactions among the various subsystems.
Second, CLIOS have impacts that are large in magnitude, and often
long-lived and of large-scale geographical extent. Third, subsystems
within CLIOS are integrated, closely coupled through feedback
loops. Finally, by open we mean that CLIOS explicitly include
social, political and economic aspects (Sussman, 2000a).
Finally,
with CLIOS we are as concerned with the complexity of the organizational
and institutional parts of the systems as we are with the physical
system. In fact, understanding the organizational and institutional
structure and its interaction with the physical structure is one
of the key potential values of a CLIOS analysis.
>
top
ESD-WP-2003-01.08-ESD
Internal Symposium: The Evolving Role of Systems Analysis
in Process and Methods in Large-Scale Public Socio-Technical Systems
by David
H. Marks, Goulder Family Professor of Engineering Systems
and Civil and Environmental Engineering, Director, MIT Laboratory
for Energy and the Environment
The ESD
definition of Large-Scale Socio-Technical Systems is large-scale
and complex systems in which both human and non-human elements
interact where the social and/or management dimensions tend to
dominate. The word public has been added here to indicate that
subset which are quasi public systems, i.e. the problems of public
management of resources such as clean air and water or energy
in which public policy is needed to drive and set the context
for public investment and regulation which in turn influence private
individual and corporate decisions. Systems analysis plays an
important role in the formation of strategic policy for managing
these resources. The paradigm of systems analysis as applied to
large-scale open systems has not changed over the years. It is
still the mantra of Problem Identification, Systems Modeling,
Generation of Alternatives (Optimization), Evaluation and Implementation.
However, both the process by which systems analysis is carried
out, and the systems methods used in that process have evolved
significantly and for the better. This paper deals with a description
of these evolving methods and processes in the context of large-scale
energy and environmental systems. In particular, pathways to the
future in energy and environmental management are discussed as
long-term system analysis problems. Systems Analysis process changes
and methods changes, which have occurred and will need evolution
in the future, are identified.
>
top
ESD-WP-2003-01.09-ESD
Internal Symposium: Architecting/Designing Engineering
Systems Using Real Options
by Richard
de Neufville, Professor of Engineering Systems and of Civil
and Environmental Engineering
Everyone
concerned with engineering systems faces a common issue: How do
we design systems to perform well in a constantly evolving and
thus risky context? As professionals concerned with the system
(rather than its individual pieces), this design issues predominantly
relates to the overall configuration, the architecture of the
system. This paper presents an approach to this fundamental issue.
It suggests how we could architect flexible engineering systems
that can evolve optimally to meet new challenges and opportunities.
It suggests that the methods of “options analysis”--
that have revolutionized thinking about investments -- can provide
a conceptual basis for defining optimal configurations. When these
procedures are applied to design issues, they are generally known
as "real options analysis".
The fundamental
result of "real options analysis" is the determination
of the value of flexibility. It thus permits system designers
and managers to decide which flexible design elements, that permit
their system to evolve effectively over time, are worth their
cost. It thus provides a clear rationale for when to design specific
types of flexibility into the system.
>
top
ESD-WP-2003-01.10-ESD
Internal Symposium: A Control Engineering Approach
to Making Complex Infrastructures More Efficient and Reliable:
A Core Program for ESD
by Marija
D. Ilic, Professor, Engineering and Public Policy and Electrical
and Computer Engineering, Carnegie Mellon University
Many
of our national infrastructures, such as electric power, gas pipeline,
transportation and information/communication systems suffer from
common design, planning and operating problems. As a consequence
of these problems, the infrastructures cannot function at the
same time both efficiently and reliably. This presents a challenge
of national importance that can be met within our own ESD Program.
In this
paper, I present a research program using control engineering
and systems theory as a unifying theme for modeling each infrastructure
as a single complex dynamic system encompassing technical, economic,
policy and information processes. Based on these models, the research
program further seeks to develop controllers that force the infrastructure
to operate both efficiently and reliably; the controllers respond
to technical, economic and policy feedback. With these controllers
in place, the design and planning of each infrastructure will
naturally evolve to enhance efficiency and reliability. Since
the controllers respond to any change in system conditions, they
are equally as effective under malicious attacks. As such, they
can function as a means of providing secure infrastructures.
The controllers
I envision will operate naturally under regulated and deregulated
policy conditions. Further, they can themselves evolve as policy
conditions change so as to maintain reliable and efficient operation
of the infrastructure. Moreover, they can catalyze policy evolution
to support more reliable and efficient operation. Equally important,
they will not just be traditional controllers that act on feedback
signals to produce actuation signals. They will also be IT-based
decision making tools that implement flexible information flow-based
protocols between industry participants so as to support such
activities as market operation and participant learning. Combining
a systematic model-based approach to risk management with IT-intelligence
and distributed hardware is a real opportunity to provide a framework
for flexible dynamic robustness in complex systems. Neither IT
nor control engineering by themselves are sufficient to embark
on this tremendous challenge. One needs a very careful combination
of the data mining techniques and the more structured control
techniques to solve the problem.
In what
follows, I will explain my vision for the ESD Program in the context
of one infrastructure, namely the electric power system. This
is the system on which most of my research has focused. Nonetheless,
my vision for the program can extend to apply to the other infrastructures
named above.
>
top
ESD-WP-2003-01.11-ESD
Internal Symposium: Learning from Organizational Experience
by John
S. Carroll, MIT Sloan School of Management, Jenny W. Rudolph,
Boston College Carroll School of Management, and Sachi Hatakenaka,
MIT Sloan School of Management
Learning-in-action,
the cyclical interplay of thinking and doing, is increasingly
important for organizations as environments and required capabilities
become more complex and interdependent. Organizational learning
involves both a desire to learn and supportive structures and
mechanisms. We draw upon three case studies from the nuclear power
and chemical industries to illustrate a four-stage model of organizational
learning: (1) local stage of decentralized learning by individuals
and work groups, (2) control stage of fixing problems and complying
with rules, (3) open stage of acknowledgement of doubt and motivation
to learn, and (4) deep learning stage of skillful inquiry and
systemic mental models. These four stages differ on whether learning
is primarily single-loop or doubleloop, i.e., whether the organization
can surface and challenge the assumptions and mental models underlying
behavior, and whether learning is relatively improvised or structured.
The case studies illustrate how organizations learn differently
from experience, the details of learning practices, and the nature
of stage transitions among learning practices.
>
top
ESD-WP-2003-01.12-ESD
Internal Symposium: Defining Engineering Systems:
Investigating National Missile Defense
by Brian
Zuckerman
The MIT
Engineering Systems Division is currently building its intellectual
framework. There is not yet consensus within ESD as to which tools
and methods are central to the nascent engineering systems approach;
which questions it should address; or the extent to which qualitative
approaches should be incorporated into it. The goal of this paper
is to sharpen the debate by presenting multiple analyses of a
single engineering system. Presenting varying perspectives illuminates
issues such as:
- What
types of questions should engineering systems practitioners
ask when analyzing problems?
- Which
tools are fundamental, which are peripheral, and which lie outside
its purview?
- Is
there a trade-off between the analytical rigor of different
tools and the degree to which they can address questions the
approach considers important?
- Does
this approach suggest generalizable principles for analyzing
engineering systems?
This
paper uses national missile defense (NMD) as the analytical vehicle
for this approach. By any definition, NMD is an engineering system.
Moreover, the complexity of NMD facilitates the framing of analyses
on multiple levels, and provides a mechanism for exploring the
ramifications of different potential definitions of “engineering
systems” as a discipline. Finally, the issue is policy-relevant.
The United States is currently deciding how to build and deploy
NMD; the choice of system architectures may have important cost,
foreign policy, military readiness, and domestic political ramifications.
While there is considerable descriptive information about system
components, there is little hard data in the open literature regarding
system performance and costs. This paper draws upon the available
literature, while making estimates where necessary.
It is
important to state at the outset that this paper assumes two key
(and often-disputed) points. First, it is assumed that technologies
under development will be feasible. Second, it is assumed that
adversaries may build intercontinental ballistic missiles and
equip them with weapons of mass destruction (in addition to Russia
and China, who already possess them). The paper therefore should
be read as a vehicle for exploring issues at the heart of engineering
systems rather than as a policy analysis
>
top
ESD-WP-2003-01.13-ESD
Internal Symposium: System Dynamics: Systems Thinking
and Modeling for a Complex World
by
John D. Sterman,
MIT Sloan School of Management
Todays
problems often arise as unintended consequences of yesterdays
solutions. Social systems often suffer from policy resistance,
the tendency for well-intentioned interventions to be defeated
by the response of the system to the intervention itself. The
field of system dynamics, created at MIT in the 1950s by Jay Forrester,
is designed to help us learn about the structure and dynamics
of the complex systems in which we are embedded, design high-leverage
policies for sustained improvement, and catalyze successful implementation
and change. Drawing on engineering control theory and the modern
theory of nonlinear dynamical systems, system dynamics often involves
the development of formal models and management flight simulators
to capture complex dynamics, and to create an environment for
learning and policy design. Unlike pure engineering problems if
any exist, human systems present unique challenges, including
long time horizons, issues that cross disciplinary boundaries,
the need to develop reliable models of human behavior, and the
great difficulty of experimental testing. Successful change in
social systems also requires the active participation of a wide
range of people in the modeling and policy design process, people
who often lack technical training. In this paper I discuss requirements
for the effective use of system dynamics and illustrate with a
successful application to a difficult business issue.
>
top
ESD-WP-2003-01.14-ESD
Internal Symposium: Lean Enterprises – A Systems
Perspective
Becoming
a “Lean Enterprise” is increasingly being recognized
as an important strategy in achieving critical strategic goals
such as responsiveness, cycle time and cost across all phases
of the product life cycle. The concept of a lean enterprise is
not new. Many books address lean enterprise topics.1 For example,
The Machine That Changed the World2, the book that introduced
lean terminology, has a chapter on “Managing Lean Enterprises”.
Despite having much written on this subject, lean enterprises
are only starting to emerge in practice. Why has it taken so long
to transform organizations to lean enterprises? Lean enterprises
are complex, highly integrated systems comprised of processes,
products, organizations, and information, with multifaceted interdependencies
and interrelationships across their boundaries. Understanding,
engineering, and managing these complex social, technical, and
infrastructure processes are critical to becoming a lean
enterprise.
What then
are the attributes of a lean enterprise? Are there key fundamental
principles employed to achieve a lean enterprise? What are the
key concepts, architecture and interrelationships that comprise
the enterprise “system”? What is involved in “engineering”
a lean enterprise? This paper will address these questions along
with the critical issues involved in modeling, analyzing and understanding
the intricacies of complex enterprise systems.
>
top
ESD-WP-2003-01.15-ESD
Internal Symposium: Nano-technology: a Disruptive
Technology?
The term
"disruptive technology" as coined by Christensen (1997)
refers to a new technology having lower cost and performance measured
by traditional criteria, but having higher ancillary performance.
Christensen finds that disruptive technologies may enter and expand
emerging market niches, improving with time and ultimately attacking
established products in their traditional markets. This conception,
while useful, is also limiting in several important ways.
By emphasizing
only "attack from below" Christensen ignores other discontinuous
patterns of change which may be of equal or greater importance
(Utterback, 1994; Acee, 2001). Further, the true importance of
disruptive technology, even in Christensen's conception of it
is not that it may displace established products. Rather, it is
a powerful means for enlarging and broadening markets and providing
new functionality.
In this
paper nano-technologies will be considered in their roles as both
disruptive and more broadly discontinuous or radical innovation.
Various impacts will be assessed with emphasis on enlarged and
new markets that may be created
>
top
ESD-WP-2003-01.16-ESD
Internal Symposium: Complex systems: a review
by Seth Lloyd,
MIT Engineering Systems Division, MIT Department of Mechanical
Engineering, Santa Fe Institute, New England Complex Systems Institute
Engineers
have worked on complex systems ever since engineering began. But
the sciences of complexity have come in to their own in the last
few decades. Hoping to find common threads that weave their disciplines
together, researchers from the fields of physics, biology, chemistry,
math, computer science, economics, anthropology, linguistics,
et al. have banded together to try to develop unifying frameworks
for understanding complex systems. This paper reports on successes
and failures of these efforts..
>
top
ESD-WP-2003-01.17-ESD
Internal Symposium: Bits and Bucks:
Modeling complex systems by information flow
by Seth Llyod,
MIT and Thomas Lloyd, McKinsey Los Angeles
This paper
presents a general method for modeling and characterizing complex
systems in terms of flows of information together with flows of
conserved or quasi-conserved quantities such as energy or money.
Using mathematical techniques borrowed from statistical mechanics
and from physics of computation, a framework is constructed that
allows general systems to be modeled in terms of how information,
energy, money, etc. flow between subsystems. Physical, chemical,
biological, engineering, and commercial systems can all be analyzed
within this framework.
>
top
ESD-WP-2003-01.18-ESD
Internal Symposium: The Link between Cognition and
the Complexity of Engineering Systems Design
by John
R. Williams, Associate Professor, Engineering Systems Division
and Civil and Environmental Engineering Department, Massachusetts
Institute of Technology,
This paper
focuses on the role of human cognition in the design of large
complex systems. It contrasts the physical system that is the
product of the design with the cognitive model that is used by
the designer to “understand” the system. The complexity
of the system relevant to the designer is a function not only
of the physical system, but also of the cognitive model that the
designer holds in his mind. Furthermore, the level of cognitive
model available to an experienced designer depends on the state
of domain knowledge. To be useful in answering the question, “How
complex is this system to design?” the state of the domain
knowledge available to the designer must be assessed with respect
to the level at which the design problem is posed. The concept
of conceptual distance is introduced that depends on the disparity
between the present level of integrated knowledge and the conceptual
level of the design problem. This “distance” is a
measure of the complexity of the design task and is called the
cognitive complexity of the design. To investigate the concept
of cognitive complexity a model of human knowledge is proposed
along with a set of graphical abstractions. It is concluded that
the cognitive complexity of the design task is neither wholly
intrinsic (a property of the system) nor wholly subjective (a
property of the mind) but requires an objective evaluation of
the engineering problem with respect to present knowledge. It
is noted that the structure of knowledge in a specific domain
can be mapped and therefore a research program can be launched
to systematically determine the difficulty of various engineering
endeavors.
>
top
ESD-WP-2003-01.19-ESD
Internal Symposium: A New Accident Model for Engineering
Safer Systems
by Nancy
Leveson, Software Engineering Research Laboratory, Aeronautics
and Astronautics Dept., Massachusetts Institute of Technology
New technology
is making fundamental changes in the etiology of accidents and
is creating a need for changes in the explanatory mechanisms used.
We need better and less subjective understanding of why accidents
occur and how to prevent future ones. The most effective models
will go beyond assigning blame and instead help engineers to learn
as much as possible about all the factors involved, including
those related to social and organizational structures. This paper
presents a new accident model founded on basic systems and control
theory concepts. The use of such a model provides a theoretical
foundation for the introduction of unique new types of accident
analysis, hazard analysis, accident prevention strategies including
new approaches to designing for safety, risk assessment techniques,
and approaches to designing performance monitoring and safety
metrics..
>
top
ESD-WP-2003-01.20-ESD
Internal Symposium: ESD Symposium Comittee Overview:
Engineering Systems Research and Practie
by the ESD
Symposium Committee
This paper
briefly introduces the field of Engineering Systems, and highlights
its emergence from engineering practice and academic engineering.
This paper was prepared by the ESD Symposium Committee based upon
its own discussions, by an analysis of the other Internal Symposium
papers, and by interactions with their authors. This paper discusses:
- a framework
for describing the field of engineering systems, and emphasizes
a three-dimensional view
- the
challenges emerging in engineering practice that are associated
with the design of complex systems
- the
methods that address research and practice problems (most of
these methods currently exist, some must be developed)
- principles
and fundamentals of engineering systems
"Engineering
systems are increasing in size, scope, and complexity as a result
of globalization, new technological capabilities, rising consumer
expectations, and increasing social requirements. Engineering
systems present difficult design problems and require different
problem solving frameworks than those of the traditional engineering
science paradigm: in particular, a more integrative approach in
which engineering systems professionals view technological systems
as part of a larger whole. Though engineering systems are very
varied, they often display similar behavior. New approaches, frameworks,
and theories need to be developed to understand better engineering
systems behavior and design.”
>
top
ESD-WP-2003-01.21-ESD
Internal Symposium: The Impact of Instability on Complex
Social and Technical Systems
by Joel Cutcher-Gershenfeld
and Eric Rebentisch
Instability
is a pervasive phenomenon that has deep implications for virtually
all complex social and technical systems.
In engineering,
the identification and mitigation of various types of technical
instabilities is a well developed practice. This is a key focus,
for example, of engineers concerned about the prevention of potentially
destabilizing vibration in the frame of an aircraft or the mitigation
of sources of technical instability in the operation of a nuclear
reactor. However, the nature of instability in complex social
and technical systems is relatively unstudied and not well understood.
This is unfortunate because instability can have profound effects
on the performance of those systems as well as their ability to
improve their performance over time.
In this
paper, we present a conceptual framework for understanding instability
in sociotechnical systems. To illustrate what we mean by instability
in the context of complex engineering systems, we will draw on
data from the aerospace industry. In particular, we use two data
sets, to trace the impacts of various sources of instability.
One data set centers on instability and its impact on aerospace
programs, while the other centers on instability and its impact
on aerospace production and design facilities.
>
top
ESD-WP-2003-01.22-ESD
Internal Symposium: Isoperformance
An Alternative Design Methodology for Engineering Systems
by Olivier
L. de Weck, Assistant Professor Department of Aeronautics
and Astronautics and Engineering Systems Division (ESD)
Tradeoffs
between performance, cost and risk frequently arise during architecting
and design of complex Engineering Systems such as aerospace vehicles.
A paradigm shift is occurring from the pure performance optimization
approach of the past towards satisfying of performance targets
under concurrent risk and cost minimization. This paper proposes
“isoperformance” as a set based approach to designing
engineering systems by first identifying the acceptable performance
invariant set of designs from which a final design is chosen.
This is in contrast to a multiobjective cost-risk minimization
under performance equality constraints. This paper identifies
a number of issues associated with finding the desired performance
invariant set, I, given a deterministic or empirical system model
that maps design variables x to objective variables J. Isoperformance
is presented as a methodology that can quantify and visualize
the tradeoffs between determinants (independent design variables)
of a known or desired outcome. For deterministic systems the multivariable
performance invariant contours can be computed using sensitivity
analysis and a contour following algorithm, provided that a mathematical
system model of appropriate fidelity exists. In the case of stochastic
systems the isoperformance curves can be obtained via a regression
analysis, given a statistically representative data set. Once
isoperformance curves have been obtained, they are useful in extracting
a set of performance invariant solutions. Applying additional
objectives, other than performance, can then lead to a set of
pareto-optimal designs. Specific examples from opto-mechanical
space systems design and human factors are presented.
Definition:
Isoperformance is a methodology for obtaining a performance invariant
set of designs or problem solutions. These solutions approximate
performance invariant contours or surfaces based on an empirical
or deterministic system model. The word isoperformance by itself
is often shorthand for the isoperformance approach.
>
top
ESD-WP-2003-01.23-ESD
Internal Symposium: Bodies, Ideas, and Dynamics: Historical
Perspectives on Systems Thinking in Engineering
Today,
the idea that technology consists not simply of individual machines
but of systems of components and interconnections underlies much
of engineering theory and practice. Yet this idea is relatively
new in the history of technology; it evolved over a long period,
spanning more than a century, as engineers grappled with the implications
of machinery and collections of apparatus that spread over broad
geographical areas. A historical perspective on systems thinking
provides a critical background for contemplating new directions
in “engineering systems,” by highlighting the problems
that have constantly challenged engineers, as well as the new
puzzles posed by today’s world.
This paper
surveys the history of systems thinking in engineering in the
United States, from the nineteenth century to the late twentieth.
Throughout this period, engineers concentrated on certain kinds
of technical systems and developed various modes of systems thinking
to deal with them. Early in the 19th century, systems thinking
developed as coherent philosophies in specialized areas like manufacturing
and the military. Later in the centur y, the railroads emerged
as a large-scale system with diverse flows and materials. From
the late nineteenth century to World War II, systems thinking
in the electric power and telephone industries focused on interconnecting
disparate elements into larger wholes for systems spread over
large geographic areas. World War II led engineers to conceptualize
systems as integrated, dynamic entities, and to formalize methodologies
for managing the complex organizations to design and operate such
systems. These approaches flourished in the Cold War, although
its techniques are still with us today in selected areas. Late
in the twentieth century, engineers began to expand the boundaries
of technical systems to include not only their internal or organizational
dynamics, but also broader social and industrial contexts. Engineers
now also recognize that the complexity of these systems means
that accurate prediction or even simulation is not always possible.
>
top
ESD-WP-2003-01.24-ESD
Internal Symposium: Large Scale Infrastructure Systems
by Fred
Moavenzadeh, Massachusetts Institute of Technology
Highways,
bridges, office buildings, houses, etc. are typical large-scale
infrastructure systems of physical facilities that must be planned,
designed, built, operated, and maintained. In addition to physical
requirements, they have complex and often farreaching interactions
with the social, political, and economic systems they serve. Built
facilities that have long service life times and large size represent
a major commitment in terms of both capital expenditure and, equally
importantly, social and political structures.
>
top
ESD-WP-2003-01.25-ESD
Internal Symposium: The Anatomy of Large Scale Systems
Many theoretical
analyses of systems emphasize their behavior. In this paper we
shall emphasize the role of organizational structure in influencing
certain aspects of the behavior of systems, rather than the full
behavior of the systems. There are several historical examples
where structure was analyzed early on in order to gain a better
understanding of systems. In medicine, for example, anatomy was
studied well before we had a deep understanding of the role and
behavior of subsystems or infrastructures of the body, such as
the liver and blood flow. Different generic structures or architectures
provide different advantages and disadvantages in coping with
changes in the overall environment in which an evolving system
is expected to operate during its lifetime. We shall discuss some
of these advantages and disadvantages for various generic structures
or architectures. One difficulty in discussing systems issues
is the lack of a relatively precise language and concepts for
dealing with such issues. We propose that abstract algebra has
at least some of the needed concepts.
>
top
|
|
|
