Model-based Strategic Knowledge Elicitation
J. Pedro Mendes
Centre for Marine Technology and Ocean Engineering (CENTEC),
Instituto Superior Tecnico, Universidade de Lisboa, Lisboa, Portugal
Keywords:
Knowledge Elicitation, Mental Models, Strategic Problems, Problem Symptoms, Problem Dynamics, Strategy
Implementation.
Abstract:
Strategic problems are difficult. They often only exist in the mental models of some top managers. They are
typically too vague to be given a precise meaning, yet at the same time concrete enough to cause discomfort.
They cannot be discarded but they cannot be tackled either because the means for diagnostic and solution
are lacking. The body of knowledge of strategy offers little help, in the sense that a set of tools for strategic
problem solving does not exist, in practice. In science and engineering, problem solving is model-based. In
the past, the social sciences and management have discarded model building due to its inherent difficulties.
Today, the means are available to elicit knowledge about the symptoms of strategic problems, create a model
to obtain a solution, and produce an implementation plan.
1 INTRODUCTION
In service or manufacturing organizations, informa-
tion about delivery delays or quality issues is fre-
quently handled down the ranks, according to well-
understood business policies. Problems are to be
solved right away, and managers at all levels are al-
legedly there for that purpose. Unsolved problems are
a threat, and those managers will do their best to hide
their existence. Eventually, top management may get
rumors of cost overruns, but there will be no reason
to change policies if some explanation can quickly
be rationalized. Top managers may only acknowl-
edge there is a problem when outright failures can no
longer be disguised. Then, as they start to feel un-
comfortable, something needs to be done.
Problems are noticeable through their symptoms,
but symptoms cannot be confused with problems.
Sometimes symptoms must be dealt with immedi-
ately, fully realizing there must be some underlying
reason for them but also knowing that searching for
that reason is a luxury that cannot be afforded. Con-
sider the familiar example of the kid with fever. The
fever must be brought down, but once the symptom is
gone its causes may be difficult to trace. If there are
recurring symptoms, the physician will ask questions
to elicit knowledge about them. Interviewing is the
most common knowledge elicitation technique, in this
case used to translate the patient’s mental model into a
diagnostic. A diagnostic is yet another model, which
will be validated through further tests and checks.
Keeping a log of the symptoms, even if in the
mind, is key to establishing reference modes for prob-
lem behavior. Without at least an implicit description
of how symptoms evolved over time, no problem is
ever acknowledged. We all believe we can solve prob-
lems, but only the problems whose nature we under-
stand. Early in our lives, the problems were given to
us. We now consider them simple, but their answers
only came if we applied the right procedures. In fact,
the chosen procedure coerces the nature of the prob-
lem. Academia keeps reinforcing this pattern: we
have financial problems, marketing problems, and so
forth. Most problems in organizations are dealt with
by executing procedures established by policy.
Problems that defy policy are difficult to acknowl-
edge (Argyris and Sch
¨
on, 1978). Devising policies to
cope with emerging problems is the role of strategy.
A strategy begins to unfold when top managers start
feeling uncomfortable about some problem. At that
point, the diagnostic consists of rightly capturing the
problem symptoms. The solution consists of identi-
fying what procedures need to be stopped or created
from scratch, where the sequence of steps for doing
so is called implementation. Therefore, the challenge
for knowledge management research is eliciting prob-
lems early, at forming stage, rather than waiting for
them to creep up and become serious.
228
Mendes, J.
Model-based Strategic Knowledge Elicitation.
DOI: 10.5220/0006082102280234
In Proceedings of the 8th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management (IC3K 2016) - Volume 3: KMIS, pages 228-234
ISBN: 978-989-758-203-5
Copyright
c
2016 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
2 BACKGROUND
2.1 A Recurring Issue
Strategic management topics carry a certain glamor
and are quite popular among business students. The
classroom problems covered are challenging and gen-
erate ample discussion. Nevertheless, business strat-
egy implementation remains problematic, perhaps
since the inception of the concept. Because this fact
remains mostly unacknowledged, we keep producing
strategies, and we keep believing we follow them, but
we actually rarely describe how to implement them.
One might think that strategic management prac-
titioners would know better. The promise of strate-
gic management is to identify what is important to
do now in view of intended future performance. And
yet, most practicing strategists don’t seem to follow
through the implementation of their strategies. With-
out proper implementation, no claims can be made
about the quality of a strategy. Unfortunately, at the
end of the day, companies keep struggling with how
to make their intentions show up in the bottom line.
The worst case scenario is when strategy theory
becomes akin to ”the emperor’s new clothes”. When a
strategy fails, executives and managers dare not blame
the failure on the approach they use, seeing as others
are seemingly following the same approach and al-
legedly doing well. Successive companies spend for-
tunes to buy or devise strategies that never produce the
intended results, sometimes even the opposite. Re-
peatedly failing to produce the intended results re-
veals a bankrupt approach.
2.2 Approach
Blaming failed strategies on difficult times, bad man-
agement or bad implementation is tempting, but the
argument does not hold water. According to Warren
(2012), both the theory and the practice of strategy
are seriously flawed. Adcroft and Willis (2008) ex-
plained that the reason is because knowledge trans-
fer from academia to practice is flawed. Thomas
et al. (2013) concurred, stating that the dominant
concern of academia appears to be a need for self-
justification rather than practical application. There-
fore, left mostly on their own, practitioners built a less
than flattering track record (e.g. Craig, 2005; Kihn,
2009; O’Shea and Madigan, 1998; Pinault, 2001).
Isaac Newton is credited for having said “If I have
seen further it is by standing on the shoulders of gi-
ants. The strategic management body of knowledge
contributes little to understand why people rely on
strategies whose implementation keeps failing. To
avoid this syndrome, a mathematical metaphor from
Kurt G
¨
odel’s incompleteness theorems (about inher-
ent limitations of formal systems) suggests that one
should search outside the current body of knowledge
to deal with the limitations of strategic management.
This possibility is supported by Johansson (2006),
who argued that intersectional innovations, combin-
ing different knowledge sources, create opportunities
for research and teaching. This paper uses key histor-
ical references to provide a contribution from the sys-
tems sciences to strategic knowledge. Some funda-
mental principles, that were available since the birth
of the field, have apparently been overlooked or for-
gotten by strategy makers.
3 THE SYSTEMS SCIENCE VIEW
OF STRATEGY
3.1 Origins of Strategy in Systems
Science
Engineering design and construction relies on find-
ing solutions to mathematical models that incorporate
knowledge about relevant laws of nature and desired
system properties. For a set of initial conditions and
environmental parameters, those solutions depict the
characteristic behaviors of the system as functions of
time, and show how the system can evolve in response
to different inputs. A model with good predictive
power cannot be derived when knowledge about sys-
tem properties is scarce. Then again, without such
knowledge, one strategy is as good as the next.
Strategy practice doesn’t use mathematical mod-
els because management theory seldom uses mathe-
matics to build knowledge about system properties.
The common belief in the social sciences that useful
models are descriptive, without predictive capabili-
ties, was acknowledged early in an article published
by the ”Society for General Systems Research” Ar-
row (1956). Adding to the proof that systems knowl-
edge was part of a manager’s background, the influ-
ential theorist Chester Barnard was a member of the
Society.
In those early days, general systems and cybernet-
ics authors were seeing common feedback dynamics
properties across disparate knowledge domains, quite
different from engineering or physics. (The term ”cy-
bernetics” was probably intended to interest an in-
terdisciplinary audience in the mathematical model-
ing background of control theory.) Among those au-
thors was von Bertalanffy (1968), from the Psychi-
atric & Psychosomatic Research Institute at Univer-
Model-based Strategic Knowledge Elicitation
229
sity of Southern California and one of the founders of
the Society.
Another early relevant author was Ashby (1956),
trained as a Clinical Psychiatrist but landing with a
double appointment at the Departments of Biophysics
and Electrical Engineering of the University of Illi-
nois. Ashby proposed the Law of Requisite Variety,
saying that a controller must have at least the same
number of states as its target system. This was a fun-
damental result for engineering. In face of an arbi-
trarily large variety of disturbance inputs (Figure 1),
a system’s output only behaves according to its ref-
erence input if its controller follows good enough a
model to produce at least an equally large variety of
counteractions.
This law is fundamental for strategic thinking. As
referred at the outset of section 2.2, blaming failure
on difficult times or something else is always easy, in
hindsight. Since the mid 1950s, it was known that a
strategist attempting to oppose the organizational en-
vironment without at least an equally vast repertoire
of actions was doomed to fail. Yet this principle is
ignored by current strategists, which leads to the fol-
lowing:
Proposition 1: strategy makers confined
to data analyses and to outguessing their envi-
ronment miss the chance to create procedures
that can face disturbances.
Figure 1: Basic control loop as a strategic management
metaphor.
3.2 Models and Systems
Another reason why strategy theory doesn’t use math-
ematical models to increase knowledge about system
properties lies in the common misperception from
social sciences that the complexity of issues is an
impediment to finding useful solutions. Engineer-
ing copes with complexity by solving increasingly
sophisticated mathematical models. For instance,
work never ceased after Alexander Graham Bell’s
1876 patent for the electric telephone. Bell him-
self and Thomas Edison were earlier pioneers. As
interest grew, problems brought by increasing tele-
phonic traffic became more complex. The solution
of newer mathematical models successively led engi-
neers from direct connection to manual switchboards
and to worldwide switching.
Complex challenges require sophisticated solu-
tions. System Dynamics authors distinguish two
kinds of complexity. Combinatorial or detail com-
plexity occurs when many variables can take many
values. Dynamic complexity occurs when effects
dont directly follow causes in time and space. Be-
sides these, there is a third kind, equally important for
strategy implementation: communication complexity
occurs when uncertainty in task execution calls for
intensive information exchange among participants.
The mathematical relationship between uncertainty
and information was established back then by Shan-
non (1948).
Regardless of how complex a challenge may be,
strategists believe their solutions are enough for a
strategy to succeed. In the absence of a powerful sys-
tems description language, faulty hierarchical com-
munication is often blamed for strategy implementa-
tion failures. Furthermore, most modern strategy text-
books also concur that the management of strategy
implementation is more art than science. This near
euphemism at least acknowledges the limitations of
the science principles in use in today’s practice, which
leads to the following:
Proposition 2: strategy makers who over-
look science and proper description tools must
live every day with the risk that resulting out-
comes may be detrimental.
In the early 1940s, at Bell Telephone Laborato-
ries, the discipline of coping with complexity be-
came known as Systems Engineering. The challenges
to communications engineering had grown past the
models of physical systems to include social con-
cerns such as user behavior. Integrating knowledge
from different technical domains was already difficult
within engineering alone, because different special-
ties used different vocabulary and graphical notations.
In other fast growing fields, pioneers faced with iden-
tical problems felt compelled to come up with differ-
ent representations (Olle et al., 1982).
Over time, different representations evolved into
the Systems Modeling Language (SysML), which be-
came a standard in 2007 (Friedenthal et al., 2015).
SysML addresses many known modeling difficulties.
Besides technological sophistication, systems models
can now represent the richness of business and man-
agerial interactions by taking into account the charac-
teristics and behaviors of human users and operators.
SysML models can describe personnel, facilities,
process, and performance requirements for a wide
range of systems. They also contribute to better com-
munication between people of different backgrounds.
The implication for strategy implementation is that
procedures can be richly communicated in SysML us-
ing terminology germane to each area of expertise.
KMIS 2016 - 8th International Conference on Knowledge Management and Information Sharing
230
4 THE DESIGN FRAMEWORK
VIEW OF STRATEGY
4.1 A Theory of Designed Strategy
Engineering is the design and construction of systems
to meet performance goals. The Accreditation Board
for Engineering and Technology defines ”design” as
the process of devising a system, component or pro-
cess to meet desired needs. Engineering design is iter-
atively revised to remove deviations from the specifi-
cation. Then, using constructive methods, implemen-
tation is made according to design.
Relying on quantitatively verifiable intentions is
the fundamental difference that separates strategy de-
sign from the ad-hoc adoption of whatever tools are
at hand. An example of a handy tool is the widely
used Balanced Scorecard (Kaplan and Norton, 1992).
Created for performance monitoring and control, it
is often misused for strategy specification or design.
However, even in engineering, the best construction
and operation practices (means) will not yield results
in the absence of design goals (ends). This leads to
the following:
Proposition 3: strategy makers who con-
fuse ends with means increase the risk of im-
plementing the wrong procedures.
In turn, ends are stated as specifications, which are
formal quantitative descriptions of required function-
ality and behavior. Specifications describe what a sys-
tem does when responding to inputs, but not how the
system does it and much less how those inputs orig-
inate. Organizational inputs are internal or external
disturbances. Seeing an organizational strategy from
the perspective of a system that can be designed leads
to the following:
Concept 1: the specification of a strategy
quantifies performance metrics for desired be-
havior in response to inputs.
All systems take inputs and deliver outputs. The
output of a strategy is the set of policies (procedures)
that cover expectable system behaviors, often includ-
ing rules for handling exceptions in regular transac-
tions. Like all complex systems, organizations are
sensitive to external events, often of unknown ori-
gin and nature. A common practice to cope with un-
certainty is to brainstorm around hypothetical events
and build scenarios accordingly. These scenarios may
have the merit of making people think about issues,
but they help solve no real problems whatsoever. To
actually solve those problems, competitive challenges
and external events need to be thought of as just alter-
native system inputs. What matters is not the nature
of those inputs but their impact on the system, and
the consequences thereof. This thinking leads to the
following:
Concept 2: behavior-based scenarios de-
scribe a system’s response to changes trig-
gered by unspecified arbitrary events.
Over the years, performance standards and other
engineering principles have been extended from phys-
ical to software systems, and then to management
systems. Kurstedt et al. (1988) defined a manage-
ment system as the set of responsibilities of a manager
bounded as a system. They generalized the definition
of manager to anyone who uses information to make
decisions that result in changes in the managed oper-
ations performance. Faced with external events, man-
agers make decisions and issue policies and directives
to control outputs. This leads to the following:
Concept 3: the output of a management
system is the net change in the operations per-
formance that results from implemented poli-
cies.
A common dictionary description of ”design” is
”to plan the form and structure of”, whereas ”to plan”
refers to ”any method of thinking out acts and pur-
poses beforehand”. Therefore, to design is to pre-
dict what a system will do and what behavior it will
display when properly built and operated in its envi-
ronment. The term ”properly” implies the adoption
of engineering principles throughout the system life-
cycle. Too many strategies are put together not on a
predictive basis, but rather on a wishful thinking ba-
sis. Designed strategies produce systems that respond
to external events (disturbances, inputs) with the in-
tended behaviors (performance). In turn, this leads to
the following:
Concept 4: to design a management sys-
tem is to think out beforehand the actions re-
quired to respond to external events and meet
specifications.
4.2 Strategy Performance
According to Wernerfelt (1984), the result of strategy
implementation is the series of business transactions
to acquire and dispose of organizational resources.
Good implementation covers the whole strategy life-
cycle for risks of overspending money and time while
performing those transactions. In turn, Fooks (1993)
described an approach for improving life-cycles by
reducing implementation cost and time. Similar to
the concept of product life-cycle, a strategy life-cycle
starts when a challenge is acknowledged and finishes
Model-based Strategic Knowledge Elicitation
231
when the challenge is considered overcome. Strategy
performance is the effectiveness and efficiency with
which the management system handles disturbances.
Renowned military strategist Col. John Boyd
showed how mental patterns or concepts of mean-
ing must be questioned and reshaped to cope with
a changing environment (Osinga, 2005). Boyd ex-
plained his well-researched theory in a two-day in-
tensive briefing, which he may have delivered some
1500 times but never published. He held that contin-
ually shortening a four-step observe-orient-decide-act
cycle (OODA loop) is critically important to outper-
forming opponents. The OODA loop is inspirational
to manage a strategy life-cycle and has many similari-
ties with the plan-do-study-act cycle (Deming, 1952).
The capability to continually evolve mental pat-
terns by detecting and correcting errors is called
learning. Argyris and Sch
¨
on (1978) showed how at-
titudes and beliefs influence the way people learn to
identify and cope with problems. They distinguish
single-loop learning, which is repeatedly solving sim-
ilar problems within prescribed goals, from double
loop learning, which is modifying a goal given previ-
ous experience. Single-loop learning is effective for
operational control (Figure 2). Double-loop learning
is a distinct capability that relies on open communica-
tion up the ladder, which gives managers the feedback
required to adapt the business model to changes in the
internal and external environment (Figure 2).
Figure 2: Acknowledged challenges impact single- and
double-loop learning.
The function of hierarchical control is to enforce
current policy, maintaining desired behavior in face of
successive challenges. However, there is a point when
current policy can no longer tackle perceived chal-
lenges, and a strategy must be devised to deal with
them. The timely decision to replace current policy
requires the capability to detect emerging problems,
which may lie beyond the scope of the strategy as
originally conceived. The problem detection process
represents the outer double-loop view of how the in-
ner single-loop management control system handles
inputs.
4.3 Argument
The likely reason why business strategy implementa-
tion still remains an issue is the lack of a sound body
of knowledge that supports going from inception to
practice. Strategy is all about living tomorrow with
the consequences of today’s decisions. Faced with
the need to state, or specify, the outcomes of deci-
sions, strategists know only how to be vague, hide be-
hind uncertain or turbulent times, and blame whatever
mishaps may arise on bad implementation. They can
do little else given the tools they have.
Weick (1996) warned against holding so dearly
to the tools of the trade that one may lose perspec-
tive of whether they still serve their purpose. Strat-
egy is preparing today for what the future may bring.
For strategists to be held accountable for the outcome
of their decisions, they not only need new tools but
also a new paradigm for using them. The paradigm
shift is focusing on the consequences of events, in-
stead of on the events themselves (proposition 1). In
turn, the consequences of events can be identified us-
ing system-based description tools (proposition 2).
The systems view looks at behaviors over time
rather than at point figures (concept 1). Considering
that behaviors are a systems response to inputs, the
paradigm shift consists of building future scenarios
around a necessarily finite number of consequence be-
haviors rather than around guesswork over the count-
less events that may originate them (concept 2). Then,
responding to each scenario calls for adopting policies
that drive the desired behaviors (concept 3).
The systems view of strategy also fosters im-
proved preparedness by supporting the creation of
policies that take into account the consequences of
unforeseen events (concept 4). This approach is quite
foreign to current teaching and strategy frameworks.
But strategy is about the means to reach desired ends,
not to be confused with the means to reach the strat-
egy itself (proposition 3).
5 KNOWLEDGE ELICITATION
5.1 Implications
A business strategy is often backed up by a busi-
ness plan. Ideally, this plan describes how the op-
erational and financial objectives of the business will
be achieved. In practice, the business plan consists
mainly of a financial forecast driven by market eval-
uation. The typical business plan may include sec-
tions describing physical operational needs, such as
location, facilities and equipment, as well as required
people skills and schedules. However, the traditional
business plan is seldom a true plan, in the sense that
its purpose is to secure approval and financing rather
than to provide an execution blueprint.
KMIS 2016 - 8th International Conference on Knowledge Management and Information Sharing
232
Blueprints are the outcome of engineering design
and a necessary means to share conceptual views
when building physical systems. Business strategies
deserve no less care. Under this metaphor, a strat-
egy blueprint consists of a comprehensive sequence
of activities supported by proven modeling tools. A
strategy blueprint contains the functional and behav-
ioral description of the actions that lead to the desired
results. The use of this blueprint as a communication
and sharing tool for the whole organization becomes
a benefit that is far from trivial.
Pfeffer (1993), who received the Academy of
Management Review’s Best Article Award that year,
said of organizational science that the field was losing
identity and drifting toward an ”anything goes atti-
tude, which seems more characteristic of the present
state” (p. 616). Strategic management may be falling
into the same trap. Warren (2012) already painted
a dark future for strategy, unless methods change.
A framework for methods change was suggested by
Mendes (2011), although at a price: new methods
need to be taught at school before they enter the main-
stream, which may be difficult if they go against the
current paradigm.
The main obstacle to overcome is teaching for-
mal modeling methods in business schools. Those are
mandatory for a proper diagnostic of problems. Not
too long ago there was a popular TV show titled Dr.
House. The main characters were a team of diagnostic
doctors that, in spite of sophisticated analysis meth-
ods, would always make two errors in each episode
before finally saving the patient. But even as they
were making mistakes, they were learning and cor-
recting their mental models. Currently, most strategic
problem solving methods rely not on models of the
problem, but on causal brainstorming and root-cause
analysis, which are ill-suited for the purpose (Mendes
et al., 2016).
5.2 Value
System Dynamics is probably the best strategy mod-
eling and diagnostic tool available. Its creator, Jay
Forrester, understood the effect of time delays and de-
cision amplification on the dynamic behavior of man-
agement systems. Forrester (1958) showed that the
pattern of information-feedback relationships among
management system components was far more impor-
tant to explain behavior over time than the compo-
nents themselves or any external events. This arti-
cle summarized his 1961 book, which is still in print
without revision.
Forrester created both a graphical notation and a
digital simulation language to represent dynamic re-
lationships. In his terminology, structure is the pat-
tern of those relationships, not the hierarchical or
functional chain of command. All management sys-
tems have a structure, whether designed or emer-
gent, known or implied. Changes to the structure can
be tested with a simulation model tuned to correctly
replicate a behavior that is specified upfront.
This capability to design and test a structure that
behaves as specified is a relevant addition to the strat-
egy designers toolkit. Simulation also uncovers unin-
tended consequences and suggests ways to cope with
them. However, powerful as it is, System Dynam-
ics never left the academic arena and remains in the
realm of specialists. Even for them, knowledge elici-
tation has been an issue (e.g. Ford and Sterman, 1998;
Vennix et al., 1992).
Ultimately, System Dynamics was never mas-
sively adopted because the solutions it produces must
be interpreted, and the bridge to implementation is not
straightforward. In the kid with fever example, the
family doctor or general practitioner was the trans-
lator between the parents and the medical specialist.
System Dynamics needs its translators. Those make it
easy to create a System Dynamics model and later, af-
ter a diagnostic and solution are available, help trans-
late back to managerial terms what needs to be done.
Mendes et al. (2016) described in detail the first step
in this strategy formulation process, the use of SysML
for problem knowledge elicitation.
ACKNOWLEDGEMENTS
The research leading to this paper was partly funded
by the Portuguese Foundation for Science and Tech-
nology (FCT - Fundac¸
˜
ao para a Ci
ˆ
encia e Tecnolo-
gia) under its annual funding to the Centre for Marine
Technology and Ocean Engineering (CENTEC). The
author thanks Marco DiMaio for suggesting the topic.
REFERENCES
Adcroft, A. and Willis, R. (2008). A snapshot of strategy
research 20022006. Journal of Management History,
14(4):313–333.
Argyris, C. and Sch
¨
on, D. A. (1978). Organizational Learn-
ing. Addison-Wesley, Reading, MA.
Arrow, K. J. (1956). Mathematical models in the social sci-
ences. In von Bertalanffy, L. and Rapoport, A., edi-
tors, General Systems: First Yearbook, pages 29–34.
Society for General Systems Research.
Ashby, W. R. (1956). An Introduction to Cybernetics. Chap-
man & Hall, London, UK.
Model-based Strategic Knowledge Elicitation
233
Craig, D. (2005). Rip-off!: The scandalous inside story of
the management consulting money machine. Original
Book Company, London, UK.
Deming, W. E. (1952). Elementary principles of the statis-
tical control of quality: a series of lectures. Nippon
Kagaku Gijutsu Remmei, Tokyo, JP.
Fooks, J. H. (1993). Profiles for performance: Total qual-
ity methods for reducing cycle time. Addison-Wesley
Longman, Reading, MA.
Ford, D. N. and Sterman, J. D. (1998). Expert knowledge
elicitation to improve formal and mental models. Sys-
tem Dynamics Review, 14(4):309–340.
Forrester, J. W. (1958). Industrial dynamicsa major break-
through for decision makers. Harvard Business Re-
view, 36(4):37–66.
Forrester, J. W. (1961). Industrial Dynamics. MIT Press,
Cambridge, MA.
Friedenthal, S., Moore, A., and Steiner, R. (2015). A Prac-
tical Guide to SysML: The Systems Modeling Lan-
guage. Elsevier, Waltham, MA.
Johansson, F. (2006). The Medici effect: What elephants
& epidemics can teach us about innovation. Harvard
University Press, Boston, MA.
Kaplan, R. S. and Norton, D. P. (1992). The balanced score-
card - measures that drive performance. Harvard Busi-
ness Review, 70(1):71–79.
Kihn, M. (2009). House of lies: How management con-
sultants steal your watch and then tell you the time.
Business Plus, New York, NY.
Kurstedt, H. A., Mendes, P. M., and Lee, K. S. (1988). Engi-
neering analogs in management. In Proceedings of the
Ninth Annual Conference of the American Society for
Engineering Management, pages 197–202. ASEM.
Mendes, J., Carreira, A., Aleluia, M., and Mendes, J. P.
(2016). Formulating strategic problems with Systems
Modeling Language. Journal of Enterprise Transfor-
mation, 6(1):23–38.
Mendes, J. P. (2011). Moving forward in organizational
science. http://www.nsf.gov/sbe/sbe 2020/2020 pdfs/
Mendes Pedro 190.pdf. [Accessed: Sept, 2016].
Olle, T. W., Sol, H. G., and Verrijn-Stuart, A. A. (1982). In-
formation Systems Design Methodologies: A Compar-
ative Review. Proceedings of the IFIP WG 8.1 Work-
ing Conference on Comparative Review of Informa-
tion Systems Design Methodologies. North-Holland,
Amsterdam, NL.
O’Shea, J. and Madigan, C. M. (1998). Dangerous Com-
pany: Management Consultants and the Businesses
They Save and Ruin. Penguin Books, London, UK.
Osinga, F. P. B. (2005). Science, Strategy and War: The
Strategic Theory of John Boyd. Eburon Academic
Publishers, Delft, NL.
Pfeffer, J. L. (1993). Barriers to the advance of orga-
nizational science: Paradigm development as a de-
pendent variable. Academy of Management Review,
18(4):599–620.
Pinault, L. (2001). Consulting Demons: Inside the Un-
scrupulous World of Global Corporate Consulting.
Harper Paperbacks, New York, NY.
Shannon, C. E. (1948). A mathematical theory of commu-
nication. The Bell System Technical Journal, 27(July,
October):379–423,623–656.
Thomas, P., Wilson, J., and Leeds, O. (2013). Construct-
ing the history of strategic management: A critical
analysis of the academic discourse. Business History,
55(7):1119–1142.
Vennix, J. A., Andersen, D. F., Richardson, G. P., and
Rohrbaugh, J. (1992). Model-building for group de-
cision support: issues and alternatives in knowledge
elicitation. European Journal of Operational Re-
search, 59(1):28–41.
von Bertalanffy, L. (1968). General Systems Theory.
George Brazziler, New York, NY.
Warren, K. (2012). The trouble with strategy: The brutal re-
ality of why business strategy doesn’t work and what
to do about it. Strategy Dynamics Ltd, Princes Ris-
borough, UK.
Weick, K. E. (1996). Drop your tools: An allegory for orga-
nizational studies. Administrative Science Quarterly,
41(2):301–313.
Wernerfelt, B. (1984). A resource-based view of the firm.
Strategic Management Journal, 5(2):171–180.
KMIS 2016 - 8th International Conference on Knowledge Management and Information Sharing
234