Content uploaded by Brian Sauser
Author content
All content in this area was uploaded by Brian Sauser
Content may be subject to copyright.
Perspectives on Measuring Enterprise Resilience
Ozgur Erol, Devanandham Henry, Brian Sauser, Mo Mansouri
School of Systems and Enterprises, Stevens Institute of Technology Hoboken, NJ USA
Abstract- The new paradigm, known as “resilience engineering”,
emphasizes the importance of measuring resilience and suggests
the development of methodologies to analyze and prepare to
improve the resilience of enterprises. In this paper we review
existing resilience measurement methodologies, and propose new
modes of measurement. We define enterprise resilience as the
capacity to decrease vulnerability, the ability to change and
adapt, and the ability to recover quickly from disruption. Using
this definition, we identify metrics which evaluate, more
specifically: (1) an enterprise’s capability to decrease its level of
vulnerability to expected and unexpected events, (2) its ability to
change itself and adapt to changing environment; (3) its ability to
recover in the least possible time in case of a disruptive event.
Based on the discussed enterprise resilience metrics, we use
several examples and evaluate a set of illustrative responses to
common disruptions.
Keywords- Enterprise resilience, resilience engineering,
resilience metrics, enterprise resilience metrics
I. INTRODUCTION
The concept of resilience has been frequently discussed as
an essential strategy for the success and survival in today’s
turbulent business environment. Although, there is a good
amount of scholarly literature defining resilience in various
disciplines, there are only few which specifically focus on
“enterprise resilience”. Most notably, there is a gap in the
literature regarding a complete approach to creating,
measuring, and maintaining “enterprise resilience”. Given the
need for today’s enterprises to be resilient, it is important to
define appropriate metrics that would help us to estimate the
resilience of an enterprise, and also to compare enterprises.
This paper builds on the previous work [11] by the authors
where they have discussed enterprise resilience in depth and
defined it as an enterprise’s capability to decrease the level of
its vulnerability to expected and unexpected threats, its ability
to change itself and adapt to its changing environment, and its
ability to recover in the least possible time in case of a
disruptive event. In this paper, the authors review the existing
methodologies of measuring resilience and attempt to provide a
foundation to develop a comprehensive methodology for
measuring enterprise resilience.
II. ENTERPRISE RESILIENCE
A. Concept of Resilience
The concept of “resilience” has been frequently used and
discussed in the literature in a range of disciplines including
psychology, materials science, computer networks, ecology,
organizational theory, economics, risk management, sociology,
psychology, sociology, risk management, and network theory
[1-10]. Although each discipline provides a different definition
and a perspective on resilience, the common element among
these definitions is that resilience is a response to unexpected
or unforeseen changes and disturbances, and constitutes the
ability to adapt and respond to such changes [11]. Another
significant aspect of resilience is that it is often used in
interdisciplinary areas concerned with complex systems,
critical infrastructure systems, and ecosystems [8]. Therefore,
resilience can reliably be described as a fundamental
contributor to the stability of complex systems [9, 12-16] [17-
21].
B. Defining Enterprise Resilience
Gallopin [27] discusses enterprise resilience as an
enterprise’s adaptive capacity and its ability to cope with, adapt
to and recover after a disruption. He also articulates how well
an enterprise can decrease the level of its vulnerability to
expected and unexpected risks, how flexible it is in
reorganizing itself despite its changing environment, and how
effective it may be in recovering in the least possible time and
at the least possible expense [27]. He states that in order to
adjust to potential risks and tolerate disruptions, enterprises are
required to reduce the complexity of their infrastructures [27].
Key to being able to achieve this and assessing vulnerabilities
is understanding the interrelationships and interdependencies
between the business processes, information, and the
supporting technologies within the enterprises [28]. Sheffi and
Rice [29] describe creating a resilient enterprise as a strategic
initiative that changes the way an enterprise operates. This
results in enhanced competitiveness resulting in enhanced
competitiveness. They suggest that enterprise resilience can
be achieved by reducing vulnerability, by implementing
redundancy, and by increasing flexibility. Sheffi and Rice [29]
also describe resilience as a function of an enterprise’s
competitive position and the responsiveness of its supply chain.
III. CURRENT APPROACHES TO MEASURING RESILIENCE
The new paradigm known as “resilience engineering”
emphasizes the measurement of resilience. The practice of
resilience engineering suggests the development of tools and
methodologies to analyze, measure, and monitor the resilience
of organizations in their operating environment in order to
improve an organization’s resilience vis-à-vis the environment,
and to model and predict the short and long-term effects of
change and operational management decisions on resilience
[31]. In this approach, measurement is a means of supplying
the information to allow for better decisions.
Dalziell and McManus [16] discuss the importance of
measuring resilience as a key requirement to achieving
resilience within an organization and within a community.
They identify key requirements of this measurement scheme as
(1) the development of simple yet effective methodologies that
organizations can use to evaluate their resilience and strategies
for organizations to improve their resilience; (2) the need for
common terminology to facilitate dialogue and debate within
organizations about their resilience priorities, and to enable
communication between organizations about common issues
and interdependencies; (3) metrics for evaluating resilience.
These metrics must be both meaningful to decision makers
within organizations, and directly relevant to the overall goals
and objectives of the organization [16].
Although its importance has been frequently emphasized in
the literature, there is a lack of proposed methodologies
attempting to measure enterprise resilience. The challenge
resides in the certain characteristics of the concept of resilience.
First, measuring enterprise resilience is a difficult task which
requires a thorough understanding of the complex
interrelationships and interdependencies of an enterprise and its
environment. Second, resilience is an emergent attribute of an
enterprise, which may not be measured during the normal
operation of the enterprise. In order to measure resilience, it is
important to identify the inherent attributes of the enterprise
which will evolve and contribute to its ability to be resilient.
The dynamic rather than the static measurement of enterprise
attributes requires both new understandings of systems as well
as new methodologies of measurement. The next section
discusses some of the characteristics and possible approaches
to measuring enterprise resilience.
A. Systems View on Measuring Resilience
A majority of resilience-related literature encourages an
overall systems view of organizations as a basis to measuring
resilience. For example, Dalziell and McManus [16]
acknowledge the fact that organizations are highly complex
and dynamic entities. Thus, measuring resilience is not a matter
of simply identifying cause and effect relationships. Different
components and stakeholders in an organization are
interconnected by complex interactions. This means it can be
difficult to understand the impact a particular decision or action
may have on the overall system. They discuss the use of
modeling systems dynamics as a useful tool to obtain detailed
analysis in order to deal with such complexity. As a basis to
measuring resilience, it is important to look at the organization
in terms of its systemic properties, such as: (1) Articulation of
the system purpose, and from that, defining the system
boundary; (2) identification of the different components or
elements that the system requires in order to achieve the system
purpose; (3) examination of the relationships between these
different components and elements to understand how they
work together to achieve the system purpose; (4) review of
how the system interacts with its environment, its influence on
the environment, and how the environment effects change
within the system [16].
B. Resilience as an Emergent Feature of the System
There is a consensus in the literature that resilience is one
of the emergent features of a system [12, 32]. Haimes et al.
[12] define the emergent properties of systems as those system
features that are not designed in advance, but evolve based on
sequences of collected events that create the motivation and
responses for properties that ultimately emerge into system
features. How does this feature relate to the measurement of
resilience? Since resilience is an emergent feature of a system,
it cannot be directly measured within the as-is state of a
system; but should be understood as an evolving feature of
dynamic systems. Enterprise resilience is thus the interaction
of the characteristics and capacities of an enterprise, which will
eventually evolve in the case of a disruptive event so it can help
an enterprise to adapt to its changing environment and recover
from the impacts of the disruptive event. In order to measure
resilience from this perspective, it is important to identify the
inherent attributes of the enterprise which will evolve and
contribute to its ability to be resilient.
Haimes et al. [12] also discuss the emergent characteristic
of resilience in the context of a system of systems. They
indicate that the system of systems performs functions and
carries out purposes that do not reside in any component
system, and that these behaviors are emergent properties of the
entire system of systems and not the behavior of any
component system. They also emphasize that component
systems are typically designed independently (not as a part of a
larger system), controlled autonomously, and then integrated in
a distributed and loosely coordinated process. The emergent
properties of a system of systems are therefore measurable to
some extent, but only through knowledge of both component
systems and their integration. Pariès [32] also discusses the
concept of emergence and explains that the larger system may
show different characteristics than its subsystems. From this
point of view, measuring the resilience of an enterprise by
considering its extended environment (i.e. its supply chain)
requires a more comprehensive approach to understanding not
only the attributes of the enterprise but also the dynamics of the
environments surrounding it.
C. Inherent and Adaptive Characteristics of Resilience
Rose and Lio [14] distinguish two characteristics of
resilience as inherent and adaptive, the former referring to
resilience under normal operating conditions and the latter
referring to the deployment of ingenuity and extra effort in
crisis situations. They propose a mathematical optimization
model for measuring resilience. The main purpose of
distinguishing these two characteristics is to isolate them for
the purposes of measurement. In particular, adaptive resilience
had not yet been sufficiently understood and so Rose and Lio
paid special attention to this type of resilience. In their model,
they used simulated data and various scenarios to define and to
measure adaptive resilience. In their model, resilience refers to
post-disaster conditions, which are distinguished from pre-
disaster activities to reduce potential losses through mitigation.
Such an approach can be applied to measuring enterprise
resilience if certain attributes and metrics can be developed for
defining the pre-disaster activities which are aimed to reduce
potential losses. This approach also requires the development
and analysis of post-disaster conditions and scenarios. Their
model also extends to measuring resilience in an individual
company and the regional economy.
D. Resilience as a Continuous Process
Wreathall [33] proposes developing measurements for
resilience in terms of the processes necessary to build
resilience. Creating resilience is not a one-time event, but
rather spans over time from pre-event strategies to post-event
recovery [12]. Those processes related to resilience include the
functions and tasks to prevent, protect, respond, and recover.
According to Wreathall, through a comprehensive analysis of
“what” has to be done and “how” it can be accomplished
should be identified, performance measures should be attained.
So what is the implication of this approach for measuring
enterprise resilience? Enterprise resilience is the outcome of a
continuous processes including planning for resilience,
responding to threats in the case of disruptions, and taking
adaptive actions in order to recover. A thorough documentation
and attained performance metrics can be used to measure the
quality of outcome of those activities which lead to resilience
of enterprises.
E. Measuring Resilience Against The Disruptive Event
As discussed in the previous section, resilience-related
actions can occur proactively, concurrently, or as a response to
something that has already occurred [15]. Therefore resilience
becomes (1) the ability to prevent disruptive events, or (2) the
ability to prevent consequences of that disruptive event
becoming worse, or (3) the ability to recover from a disruptive
event that has happened. For each perspective, several metrics
can be identified For example, Rose and Liao [14] propose to
determine a quotient of failure probability, reduced
consequences from failure, and reduced time to recover.
Probability of failure is selected as a metric which indicates the
ability to prevent disruptive events, reduced consequences from
failure is a metric of the ability to prevent the consequences of
that disruptive event, and finally the reduced recovery time is
the metric for ability to recover from a disruptive event.
Westrum [15] classifies disruptive events based on their
predictability, their potential to disrupt a system, and the origin
of that disruptive event whether it is internal or external. A
vulnerability map here may best exemplify the relative gravity
of some disruptive events:
Figure 1 Vulnerability Map
Classifying the types of disruptive events or threats help to
create preventive actions and to model how a system will react
in case of that threat. Walker and Myers [34] use a similar
approach to create a database of threshold changes. Based on
past events, a classification of disruptive events and their
consequences can be used to model and to predict a resilience
measure. A similar approach is used by Sheffi [29]. He
proposes the use of vulnerability maps such as we see in Figure
1 help to visualize the relative likelihood of potential threats to
an organization and the company’s relative resilience to such
disruptions. These maps can be used to assess the resilience of
enterprises.
F. Measuring Resilience using Adaptive Capacity and Time
Dimension
Although the adaptive capacity of a system is a major
determinant of its resilience, it should not be considered in a
static manner. [31]; [16]. Ability to change and adapt can be a
much more meaningful measure if the time dimension is
considered [8]. For example, Dalziell and McManus [16]
propose defining key performance metrics and then measuring
resilience as a function of a system’s vulnerability and its
adaptive capacity within the desired time frame (Figure 2). A
similar approach can be applied to measure enterprise
resilience. Key performance indicators may have been
designed and measured for normal operations. However, the
challenge still resides in assessing the degradation as a result of
a disruptive event.
Figure 2 Resilience based on Adaptive Capacity and Vulnerability
Any system can sooner or later adapt to a changing
environment, but the time it takes for such adaptation is also
important from the perspective of resilience. Measuring
adaptive capacity can only be a meaningful measure for
resilience if the time dimension is considered, and the proper
chronological model is applied in each case. For example, the
time between the disruptive event and the system’s first
response to that event, or the time between the first impact and
full recovery. The time dimension of resilience based on the
phases of a disruptive event is illustrated by Sheffie’s [29]
model for phases of disruptive event (Figure 3).
Figure 3 Phases of a Disruptive Event
IV. PROPOSED METRICS FOR ENTERPRISE RESILIENCE
A. Recovery Time
Arguably the most significant metric in a dynamic and
diachronic model is recovery time, and in this section we will
discuss more concrete examples in order to demonstrate the
applicability of the measurement methodology we have been
discussing. Recovery time can be considered as the time taken
for an enterprise to overcome disruption and return to its
normal state. In order to measure recovery time, we need well
defined start and stop points. The start point could either be (a)
the occurrence of the disruption or (b) when the disruption
affects the enterprise – though in some cases, both these could
happen at the same instant of time (e.g. a factory being hit by
an earthquake). If it were (a), it also depends on the nature of
the disruption - some could be instant (e.g. earthquake) while
some others could be while some others could extend over a
period of time (e.g. drop in sales). If the start point were to be
(b), it brings in additional challenges in defining it properly.
The precise instant at which a disruption affects an enterprise
would depend on the nature of the disruption and also if the
effect is direct/primary (e.g. earthquake hitting an assembly
plant) or indirect/secondary (e.g. earthquake hitting the
manufacturing plant of a supplier). The stop point depends on
the definition of the recovered state, which can be considered
(in this section) to be equal to the normal state before the onset
of the disruption. Hence the stop point would be the instant of
time at which the enterprise reaches its original state.
Recovery time can therefore be calculated as the time
between the start and stop points, as defined above. We can
consider two similar enterprises A and B operating at their
respective normal levels of L1 and L2 respectively, as shown in
Figure 4 with time on the x-axis and level on the y-axis. Let us
assume that these two enterprises are affected by the same
disruption. Hence the start point for both enterprises is T0,
however they may be considered (a or b). The enterprises A
and B recover to their original levels at times T1 and T2
respectively. By our definition, stop points for the two
enterprises A and B would be T1 and T2 respectively. Hence
recovery times (RT) would be calculated as:
RTA = T1 – T0 RTB = T2 – T0
Figure 4 Comparison of Recovery Times
Therefore, from Figure 4, it can be observed that enterprise
A has a shorter recovery time than enterprise B. Recovery time
could be used as an indicator of resilience. If we were to
consider a shorter recovery time to indicate better resilience,
then we can conclude that enterprise A is more resilient than
enterprise B since T1 is lesser than T2. However, let us analyze
how the two enterprises behaved after the disruption occurred.
Enterprise A took a long time to get disrupted, but then it came
back quickly to reach its original state L1. On the other hand,
enterprise B got disrupted quickly and also came back as
quickly close to its original state (i.e. 80%) but it took a long
time to recover to its original level L2. Now, let us consider the
time instant T3, as shown in Figure 4. At that instant of time,
enterprise A is almost at its most disrupted state, while
enterprise B has recovered to about 80% of its original state
after being impacted by the disruption. Hence at the instant of
time T3, it can be said that enterprise B is more resilient than
enterprise A, which is different from our earlier conclusion. In
addition, if we had defined the stop point for the recovery time
calculation to be the time at which the enterprise reaches 80%
of its original state, then enterprise B would have been
identified to be more resilient than enterprise A. Similarly, the
manner in which start point gets defined, could also lead to
different conclusions. While comparing two enterprises, we
can possibly identify which enterprise is more resilient at that
instant of time. This conclusion is also applicable only at that
instant. In future, the enterprise that seemingly failed to recover
from a disrupted state may actually do so, but slowly. Does it
mean that the enterprise is not resilient? Is it ever too late to
exhibit resilience? This leads us to the issue of what an
acceptable or typical recovery time could be, in the context of
enterprise resilience. Primarily, this would depend on the
nature and intensity of the disruption and the damage it causes
to the enterprises. In addition, recovery time could depend on
many other factors like industry type (e.g. manufacturing or
service), production capacity (in a manufacturing enterprise)
and system size and complexity (in a manufacturing or
development enterprise). It can therefore be seen that in order
to use recovery time as an indicator of resilience, clear
definitions of start and stop points are required, which in turn
depend on the definition of the recovered state and of
resilience. In addition, when two dissimilar enterprises are
being compared, they must be brought to the same level
reference before their resilience can be compared. The
challenge still resides in identifying the stop and start points
effectively.
B. Level of Recovery
The level of recovery must also be accounted a fundamental
factor in assessing resilience. This metric can be used in
combination with the time factor to create a dimensional
understanding of predictable outcomes for a particular
enterprise or enterprise system. There are differences in the
definitions of resilience with respect to the level of recovery.
Some definitions indicate [35] that it is sufficient to recover to
a minimum state - this means that the recovered level is lower
than the original level. Some other definitions indicate that
recovery level is the same as the original level while some
others indicate that the recovered level is higher than the
original level [22]. It is evident that the level of recovery must
be defined appropriately before using it to indicate resilience of
an enterprise. The level of recovery could be specified by
comparing the recovered level to either the original level or the
disrupted level (the lowest level reached). We can consider the
scenario of two enterprises A and B as presented in Figure 5.
Time is on the x-axis, but is not considered in discussions here,
and the level of performance is on the y-axis. For enterprise A,
the levels of performance corresponding to the initial, disrupted
and recovered states are indicated by levels L1, L3, and L5. The
corresponding levels of performance for enterprise B are L2, L4
and L6. With respect to the original state, level of recovery
(RL) could be calculated as the difference between the
recovered level of performance and the original level of
performance, with resilience being indicated by value of this
difference. For the two enterprises A and B, recovery levels
could be calculated as:
RLA = L5 – L1 RLB = L6 – L2
From Figure 5 it can be seen that RLB would be lower than
RLA. Enterprise B was able to recover is closer to its original
level (i.e. 80%), while enterprise A could not recover very
close to its original level (i.e. 70%). Hence it could be
concluded that enterprise B is more resilient.
Figure 5 Comparison of Levels of Recovery
But when we calculate recovery levels with respect to the
disrupted state, it leads to a different conclusion. When
recovery level is calculated as the difference between the
recovered level of performance and the disrupted level of
performance, the values of enterprises A and B could be
calculated as:
RLA = L5 – L3 RLB = L6 – L4
From Figure 5 it can be seen that RLA would be higher than
RLB. Enterprise A recovered by a larger amount from its
disrupted state while enterprise B did not recover by the same
amount. Hence it could be concluded that enterprise A was
more resilient. Enterprise B suffered a lesser disruption that
enterprise A, but it recovered to a level closer to its original
level than enterprise A did. On the other hand, enterprise A
suffered a larger disruption and recovered by a larger amount
from the disrupted state that enterprise B. So, which enterprise
is more resilient? In order to use the level of recovery as an
indicator for resilience, it must be calculated involving all the
levels of performances of the original state, disrupted state and
the recovered state. Some define resilience to be the capacity to
tolerate disturbances – this means that there is no disrupted
state [36]. This indicates that in addition to the three levels
indicated above the potential disrupted state that was avoided
should also be included in indicating the level of resilience. But
further study is required to make these indications. In some
other cases, the initial state itself could be a disrupted state or
one full of challenges and difficulties – e.g. a startup or a small
enterprise that is competing with larger and well established
enterprises.
C. Level of Vulnerability to Potential Disruptions
The level of vulnerability to potential disruptions can also
be an indicator of enterprise resilience. It is important to note
that an enterprise that is resilient to one kind of disruption may
not be as resilient to another type of disruption. This leads us to
a possible definition of a measure of overall resilience of an
enterprise. This could be a function of the individual resilience
of the enterprise to various disruptions. As we begin to
consider the integration of these correlated metrics of
resilience, it would be very useful to examine concrete cases. In
this way we could come to an understanding of both resilience
and its opposite, that is, vulnerability to disruptions of various
kinds:
• Immediately after the terrorist attacks of September 11,
2001, there was a sudden increase in US flags, lapel pins
and other patriotic items. Wal-Mart noticed this and by the
evening of 9/11/2001, it had ordered all US flags in its
entire supply chain. Other retailers like Kmart and Target
were too late to respond to the new demand, and their
stocks emptied very quickly. So for some time, shoppers
could find US flags only in Wal-Mart [39].
• In 2002, a 10-day labor lockout shut down 29 ports on the
US West Coast. This affected hundreds of cargo ships that
had reached the pacific shores, and hundreds on the way
and about to leave from ports in Asia. This disrupted
supply chains of all companies that depended on the
supply of goods via the west coast ports. Like most of
these affected companies, Dell too was expected to be
severely affected by the lack of parts, more so with its just-
in-time manufacturing model. But the same model kept
Dell constantly in touch with its suppliers and the port
situation and predicted the port lockout. Dell then
chartered 18 Boeing 747s to ship all the parts it needed
from Asia, and survived the lockout without delaying
delivery to even a single customer [40]. Other companies
like Apple could not satisfy customer demands and took
much longer to recover from the supply chain disruption
due to the west coast port lockout.
• An earthquake that hit Taiwan on September 21, 1999,
affected the supply of semiconductors to many computer
manufacturers. Apple faced a shortage of chips and other
critical components for its latest laptop and desktop
models. Thousands of orders had already been placed, and
it was too late to modify the configurations with other
earlier versions of components. Apple was left with no
choice but to refund payments, and this severely affected
its sales figures. Dell on the other hand, faced a similar
shortage of supply, but in its model there was much lesser
time from order to supply and hence it did not suffer
backlogs of thousand of orders. It offered only models that
it had parts for, and hence sold only them – this enabled
Dell to survive the disruption and also increase its sales in
that quarter [29].
• Lighting struck a Philips semiconductor plant in New
Mexico on March 17, 2000 and caused a small fire that
polluted the clean room facility and destroyed thousands of
chips. These chips were being supplied to many companies
including two cell phone giants Nokia and Ericsson. Nokia
was about to rollout a new model, and it depended on
chips from the affected Philips lab. Nokia immediately
realized the potential damage due to this disruption, and
went on fast track to collaborate with Philips and source
the chips from other Philips plants located in other parts of
the world. As a result Nokia avoided disrupting supply to
any of its customers. Ericsson on the other hand, waited
for the situation to be handled by Philips, and lost millions
of dollars. It eventually was bought over by Sony [41].
The four scenarios briefly describe how some companies
were able to avoid or reduce the impact while others were
majorly affected by the same disruption. It is possible to
compare the responses of the two companies based on a set of
criteria, and this could lead to a qualitative measure of the
resilience of the companies. The level of vulnerability to
certain types of disruptions can a meaningful metric if it can be
integrated with the metrics that we have proposed which are
level of and time to recovery.
V. DISCUSSION AND CONCLUSIONS
This paper reviewed some of the existing resilience
measurement methodologies from which to develop a
comprehensive methodology for measuring enterprise
resilience. Based on the literature review we emphasized the
need to have well-defined quantitative and qualitative metrics
in order to evaluate enterprise resilience. We proposed some
quantitative metrics that took into account the dynamic, organic
features of complex systems, and thus our methodology took
into account the crucial parameters of recovery time, level of
recovery, initial vulnerability, and potential loss averted.
Further study will expand on the crucial area of correlation
between the various aspects of enterprise resilience. In
addition, other categories of measurement will be necessary in
order to more accurately assess reactions to particular
disruptive events. It is easier to measure resilience when it
seems to have been manifested over time, in commercial
success stories. However, the field is open for new and
comprehensive modes of measurement.
REFERENCES
[1] G. Bonanno, "Loss, trauma, and human resilience: have we
underestimated the human capacity to thrive after extremely aversive
events? ," American Psychologist vol. 59, pp. 20-28, 2004.
[2] W. Adger, "Social and ecological resilience: Are they related? ,"
Progress in Human Geography vol. 24, pp. 347–364, 2000.
[3] L. Mallak, "Toward a Theory of Organizational Resilience," in
Portland International Conference on Technology and Innovation
Management. PICMET. vol. 1: IEEE, 1999, p. 223.
[4] D. Callaway, M. Newman, S. Strogatz, and D. Watts, "Network
robustness and fragility: percolation on random graphs," Physical
Review Letters vol. 85 pp. 5468–5471, 2000.
[5] W. Arthur, "Complexity and the economy," Science, vol. 284, pp.
107-109, 1999.
[6] C. Folke, S. Carpenter, T. Elmquist, L. Gunderson, C. Holling, and B.
Walker, "Resilience and sustainable development: building adaptive
capacity in a world of transformations," Ambio, vol. 31, pp. 437–440,
2002.
[7] R. Starr, J. Newfrock, and M. Delurey, "Enterprise Resilience:
Managing Risk in the Networked Economy," in Strategy+Business.
vol. Spring 2003: Booz & Company, 2003.
[8] S. Carpenter, B. Walker, J. M. Anderies, and N. Abel, "From
Metaphor to Measurement: Resilience of What to What?,"
Ecosystems, vol. 4, pp. 765–781, 2001.
[9] C. S. Holling, "Resilience and stability of ecological systems,"
Annual Review of Ecology and Systematics, vol. 4, pp. 1-23, 1973.
[10] J. Fiksel, "A framework for sustainable materials management,"
Journal of Materials, vol. 58, pp. 15-22, 2006.
[11] O. Erol, B. Sauser, and M. Mansouri, "A Framework for Investigation
into Extended Enterprise Resilience," Enterprise Information
Systems, vol. In review, 2009.
[12] Y. Y. Haimes, K. Crowther, and B. M. Horowitz, "Homeland
Security Preparedness: Balancing Protection with Resilience in
Emergent Systems," Systems Engineering, vol. 11, pp. 287-308,
2008.
[13] E. Hollnagel, D. D. Woods, and N. Levesson, "Resilience
engineering: Concepts and precepts," Hampshire: Ashgate, 2006.
[14] A. Rose and S. Liao, "Modeling regional economic resilience to
disasters: A computable general equilibrium analysis of water service
disruptions," Journal of Regional Science, vol. 45, pp. 75-112, 2005.
[15] R. Westrum, "A typology of resilience situations," in Resilience
Engineering: Concepts and Precepts, E. Hollnagel, D. D. Woods, and
N. Leveson, Eds. Aldershot, UK, : Ashgate Press, , 2006, pp. 49-60.
[16] E. P. Dalziell and S. T. McManus, "Resilience, Vulnerability,
Adaptive Capacity: Implications for System Performance," in
International Forum for Engineering Decision Making (IFED) Stoos,
Switzerland, 2004.
[17] ResilienceAlliance, "Resilience Alliance Key Concepts," in
(www.resalliance.org), 2008.
[18] D. van Opstal, "The Resilient Economy: Integrating Competitiveness
and Security," Council on Competitiveness 2007.
[19] M. T. Gibbs, "Resilience: What is it and what does it mean for marine
policymakers?," Marine Policy, vol. 33, pp. 322–331, 2009.
[20] J. Fiksel, "Sustainability and Resilience: Toward a Systems
Approach," Sustainability: Science, Practice, & Policy, vol. 2, pp. 14-
21, 2006.
[21] D. Arsenault and A. Sood, "Resilience: A Systems Design
Imperative," Department of Computer Science, George Mason
University, Fairfax, VA 2007.
[22] T. J. Vogus and K. M. Sutcliffe, "Organizational Resilience: Towards
a Theory and Research Agenda," in Systems, Man and Cybernetics,
2007 . ISIC. IE EE International Conference on, 2007.
[23] K. Sutcliffe and V. T., "Organizing for Resilience " in Positive
Organizational Scholarship, K. S. Cameron, I. E. Dutton, and R. E.
Quinn, Eds. San Francisco: Berrett-Koehler,, 2003, pp. 94 - 110.
[24] L. Mallak, "Putting Organizational Resilience to Work," Industrial
Management, vol. 40, pp. 8-13, 1998.
[25] M. Scott, G. Sorcinelli, P. Gutierrez, C. Moffatt, and P. DesAutels,
"CONFERENCEXP: An Enabling Technology for Organizational
Resilience," in The Transfer and Diffusion of Information
Technology for Organizational Resilience. vol. 206, B. Donnellan,
Larsen T., Levine L., and D. J., Eds.: Boston: Springer, 2006, pp.
219-227.
[26] E. S. Patterson, D. D. Woods, R. I. Cook, and M. L. Render,
"Collaborative Cross-Checking to Enhance Resilience," Cogn Tech
Work, pp. 155-162, 2007.
[27] G. C. Gallopin, "Linkages between vulnerability, resilience, and
adaptive capacity," Global Environmental Change, vol. 16, pp. 293–
303, 2006.
[28] G. Goble, H. Fields, and R. Cocchiara, "Resilient infrastructure," IBM
Global Services September 2002.
[29] Y. Sheffi and J. B. Rice Jr., "A Supply Chain View of the Resilient
Enterprise," MIT Sloan Management Review vol. 47.1, pp. 41-48,
2005.
[30] M. Christopher and H. Peck, "Building Resilient Supply Chain,"
International Journal of Logistics Management, vol. 15, pp. 1-13
2004.
[31] D. D. Woods and E. Hollnagel, "Prologue: Resilience Engineering
Concepts," in Resilience Engineering: Concepts and Precepts, E.
Hollnagel, D. D. Woods, and N. Leveson, Eds. Aldershot, UK, :
Ashgate Press, , 2006, pp. 49-60.
[32] J. Paries, "Complexity, Emergence, Resilience..." in Resilience
Engineering: Concepts and Precepts, E. Hollnagel, D. D. Woods, and
N. Leveson, Eds. Aldershot, UK, : Ashgate Press, , 2006, pp. 43-53.
[33] J. Wreathall, "Developing Models for Measuring Resilience," John
Wreathall & Co., Inc., Dublin, Ohio 2006.
[34] B. Walker and J. A. Meyers, "Thresholds in ecological and social–
ecological systems: a developing database," Ecology and Society, vol.
9, 2004.
[35] E. Hoffman, "Building a Resilient Business," in Raptor Networks
Technology Inc., 2007.
[36] J. Fiksel, "Designing Resilient, Sustainable Systems," Environmental
Science and Technology vol. 37, pp. 5330-5339, 2003.
[37] D. J. Rosenkrantz, S. Goel, et al. , "Structure-Based Resilience
Metrics for Service-Oriented Networks," in 5th European Dependable
Computing Conference Budapest, Hungar, 2005.
[38] W. a. J.-L. G. Najjar, "Network Resilience: A Measure of Network
Fault Tolerance," IEEE Transactions on Computers, vol. 39, pp. 174-
181, 1990.
[39] N. a. M. L. Pal, "The Agile Enterprise: Reinventing Your
Organization for Success in an On-Demand World," in Emergence of
the Agile Enterprise, N. P. a. D. C. Pantaleo, Ed. New York:
Springer-Verlag, 2005.
[40] B. Breen, "Living in Dell Time," in Fast Company. vol. 88, 2004.
[41] Y. Sheffi, The Resilient Enterprise. Overcoming Vulnerability for
Competitive Advantage. Cambridge, Massachusetts: MIT Press,
2005.
