ble of Contents
Making system dynamics accessible to SMEs through user-
parameterised generic models
Daniel J. W. Arthur and Graham W. Winch
University of Plymouth Business School, Plymouth, PL4 8AA, U.K.
graham.winch@ pbs.plym.ac.uk
Abstract
System dynamics is a powerful strategic analysis approach, but it can also be a costly
one, if all models must be constructed largely from scratch. This appears to preclude
their widespread use in small-medium enterprise (SME) applications. Field research
has shown that generic parameterisable simulations can challenge mental models
and enhance confidence as a firm faces major change. We discuss whether such
simulators could be the only viable option for SMEs to benefit from system dynamics
modelling and scenario planning.
Introduction
In Industrial Dynamics, Forrester (1961, p365) argued that dynamic modelling was
applicable to smaller firms as much as to larger ones, and maybe even more so, particularly
firms facing the challenges of major growth:
“.,aggressive, rapidly growing, medium- and small-size organisations may be the places where the
methods discussed in this book will have their first important impact ... the smaller organisations may
be more fluid ... the costs of management systems research are low enough so that they present no
great difficulty...”
Historically, this does not seem to have been the case and the impact of SD within the SME
sector has not been high. With this statement in mind, a research project into how SMEs
prepare for fundamental business change led to the development of a ‘change visioning’ tool
built around a system dynamics flight simulator. One key feature of the simulator was a core
generic model that could be parameterised to a specific firm by non-specialists through a
self-managed interrogatory interface. Given the potential cost savings in this process, this
short paper discusses the potential for such tools in providing low-budget entry into system
dynamics modelling for SMEs
The Notion of User-parameterised Models and the Challenge for SMEs
The core model in the simulator was ‘generic’ and intended to reflect a wide range of firms
and change situations. The concept of using generic models which are tailored to specific
circumstances is not new, even in practical studies as opposed to learning situations (e.g. by
Alfeld, 1995; Lyneis et al., 2001). It was constructed using high level and aggregated
tepresentations of a business for the purpose of capturing the longer-term drivers of the
business (Winch et al., 1999). The idea here was that modelling the needs, values and
expectations from a top-level perspective was more important than a more concrete
operational view that ties the view of a firm to a particular business architecture. There are
clear parallels in this with the resource-based view of the firm (Bamey, 1991), and notions of
‘natural work groups’ as discussed by Teerlink and Ozley (2000). The model was tailored to
individual firms participating in the research through provision of an automated front end
through which the directors who were driving the business change could parameterise the
core model themselves. The high level approach also reduced the number of parameters
required and hence the time required for the parameterisation routine, although the
parameters then tend to be more abstract. The process of parameterisation of the model and
then its subsequent use by company managers to ‘pre-experience’ possible post-change
futures for their firm is summarised in Figure 1.
Generic Model Flight Simulator
Change Model |B | Parameterised to |— p> (Envisioned
Firm Future)
a a
Input of data * 4
2 = is Experimentation to
[ Interrogatory Interface | Simulator Interface ‘pre-experience’
S envisioned post-
change future
reflecting envisioned
post-change firm
é la
Change Post-change
Initiator Manager
Figure 1 - Schema showing the modus operandi of the simulator
The details of this approach are discussed in Winch and Arthur (2002) where it is argued that
such user-parameterised generic models may be the only way of making the benefits of SD
modelling available to SMEs. The argument is that SD models, custom-built from scratch to
a particular firm or issue and rigorous in their structure and detail, are very expensive in time,
expertise and cost, and often built with the assistance of consultants and through group
participation within the firm. The general belief that a major benefit of modelling can come
in the model-building process is not disputed here, but if there is an insistence that all “good”
models have to be constructed in this way, then unfortunately the costs involved would
appear effectively to make system dynamics predominantly, if not exclusively, a large
organisation approach.
Some managers (a growing number but still small) may have had previous exposure to
systems thinking and modelling, or have access to subsidies for the cost of such work,
although usually not as one-off company-specific problem solving or situation analysis
activities. Generally, however, time and cost constraints seem to preclude the use of SD
modelling by SMEs, which are often led by entrepreneurs who lack time (and frankly often
the inclination) to engage in systems thinking and modelling as a separate activity.
Generic models or archetypes can be used to categorise business situations and their use in a
teaching context to make simplified representations of general business cases is quite
widespread. Our research concluded that such models could provide pre-packaged tools to
make available some of the benefits of model-based scenario planning in the face of major
future business challenges. The research certainly confirmed that facilitated workshops with
managers using the simulator that had been parameterised by the directors were likely to be
more fruitful than unfacilitated ones. Nonetheless there could still be benefit in the latter case
in terms of raising awareness of issues surrounding a planned change, emphasising the
probable need for new thinking on the managers’ part, and in building confidence to face
new challenges. Of course, the structure of the model itself was not visible to the participants
and therefore retains the drawbacks of ‘black-box’ models. Trial users of the system
nevertheless reported perceived clear benefits, even in the unfacilitated sessions.
Conclusions
Given the obstacles to introducing modelling into the SME, it is argued here that user-
parameterised generic models could be a way of overcoming some of the barriers. Other
solutions could include:
= a longer term educational approach to raise awareness among SME entrepreneurs, in
particular, of strategy and systems thinking skills, with increasing use of modelling
through ‘diffusion’;
= persuading managers of the benefits of modelling as part of cost- and time-efficient
structured approaches to business planning;
= making capital funding contingent upon evidence of structured and holistic systemic
thinking in business plans, since this is an inherent part of a vision for a growing
business.
However, these other ‘solutions’ are costly and probably longer term, and have not yet
completely penetrated the large firm context. It is worth returning, therefore, to the original
premise of this article - namely the assumption that SD modelling is as relevant to SMEs as
larger firms. From business development activity in liaison with small firms (<50
employees), the first author has noted a priority amongst managers to focus on the essential
elements of keeping the business running: a marketing programme, ensuring product quality,
effective management of workloads, dealing with staffing problems, dealing with dissatisfied
customers etc. Structured systemic thinking often seems to be both a luxury and somewhat
esoteric for many managers. Getting managers to use basic strategy tools such as a SWOT
analysis is a challenge in itself. Experience from the second author (Winch, 2000, and other
consulting firms) shows that SD has probably the greatest impact to make in very high stakes
decisions faced by large firms. Perhaps the best positioning for user-parameterised generic
models is in medium size firms, also facing strategic decisions, but with the simulator
playing a sensitising role to raise awareness of possible future dynamics. It then serves as a
precursor (or “opening wedge” as Forrester, 1961, p 360) for possible later extension and
bespoke development.
References
Alfeld LE, 1995, Urban Dynamics - the first fifty years. System Dynamics Review, 11:3,
199-218.
Barney J, 1991, Firm Resources and Sustained Competitive Advantage. Journal of
Management 17(1): 99-120.
Forrester JW, 1961, Industrial Dynamics. Republished by Pegasus Communications.
Lyneis JM, 1999, System dynamics for business strategy: a phased approach. System
Dynamics Review 15(1): 37-70.
Teerlink R and L Ozley, 2000, More than a Motorcycle, Cambridge: Harvard University
Press
Winch GW, DJW Arthur and C Grey, 1999, Building and Conceptualising Generic Model
Structures to address Fundamental Business Change. Proceedings of the 18", System
Dynamics Conference, Wellington, New Zealand, SD Society .
Winch GW, 2000, System Dynamics: from Theory to Practice. Keynote Address:
International Conference on Systems Thinking in Management, Deakin University,
Geelong, Australia, November.
Winch GW and DJW Arthur, 2002, User-parameterised generic models: a solution to the
problem of modelling access for SMEs. Article submitted to System Dynamics Review
special issue on Small-Medium Enterprises.
Back to the Top
