-63-
Gaining Acceptance for a Systems’ Dynamics Model.
Andrew S. Breiter
A.B.C.-Tecniche Avanzate di Gestione
ABSTRACT.
The paper discusses the reasons why Systems' Dynamics models frequently
encounter considerable difficulties in gaining acceptance and suggests
several ways for overcoming this obstacle. Resistance to models within
organizations is usually generated by one or several of the following
causes: insufficient credibility of model's proponents, inability to grasp
model's usefulness, cultural background, fear of losing power and negative
previous experience with models. In the special case of models addressing
issues of wide public interest suggestions are presented on how to plan a
communications strategy designed to generate support for the model or for
the conclusions derived with its help.
INTRODUCTION.
Systems' Dynamics has made relatively little ground in gaining diffusion
over the last 20 years. The doubtless publicity from the exceptional
diffusion of a book like "Limits to Growth" does not appear to have
accelerated significantly its wide acceptance.
In spite of the broad spectrum of potential applications and of a relative
ease with which it is possible to communicate the contents of Systems!
Dynamics models, as compared for example with linear programming models, the
technique is slow in being widely accepted. It is used only in a few
countries and then it is confined to isolated applications in spite of the
fact that insights gained through study of behaviour of the modeled systems
doubtlessly help to make better decisions on a wide variety of issues.
This author believes that two causes contribute to delay the diffusion of
Systens' Dynamics.
Insufficient development of the Systems' Dynamics theory makes the success
of its teaching highly dependent on the ability and on the perseverance of
individual students to supplement this defficiency. Consequently Systems!
Dynamics practical use has remained limited to those few who have mastered
it through learning by trying, erring, sweating and correcting. The others
have either abandoned the efforts to use the technique or have produced poor
results that made in fact the diffusion of Systems' Dynamics more difficult.
The second major cause slowing down the diffusion of Systens' Dynamics is
seen in the frequent rejection either of the intent to build a Systems!
Dynamics model or in the rejection of its conclusions.
-64-
This paper explores this second problem, examines its causes and proposes a
methodological approach aimed at obtaining support for the models and for
conclusions derived from their use, provided that minimal conditions making
this feasible exist in a given situation.
The rejection of Systems' Dynamics models usually appears to be rooted in one
or more of the following reasons:
lack of credibility of the people who propose to build the model,
- lack of confidence that the technique can be useful,
- background which makes it difficult to understand this kind of
models,
~- fear of losing power by those who should contribute knowledge of the
system if their personal insights arrived at through effort and time
were to be explicitly stated and perhaps more widely diffused, thus
enabling others to take up a greater role in the decision making
process,
- unfortunate experience with models in general or with Systems!
Dynamics models in particular.
Several approaches are suggested to deal with the various situations
encountered within organizations.
A particular case is examined when Systems' Dynamics models address issues
of wide public interest and there is need to gain acceptance for the model
from wide segnents of public opinion.
MOST FREQUENT OBSTACLES TO APPLICATION OF SYSTEMS! DYNAMICS MODELS.
The causes and effects that most frequently interact and influence an
individual within an organization to accept or to reject a Systems' Dynamics
model are shown in Fig. 1.- The objections to such a wodel usually are to
be sought among the ones described below.
Lack of clearly established credibility of the person or group that will
build the Systems’ Dynamics model may ground any effort to build it or to
use it. Such credibility should not be taken for granted, in fact it is very
frequent that there is a lack of confidence in the people who propose to
build a Systems' Dynamics model. This lack of confidence typically leads to
doubts that such people know the system to be modelled or that they know how
to apply the technique.
“65a
Fig. 1.- An individual may accept or reject a Systems' Dynamics model as a
consequence of a number of different, partly interrelated
reasons. -
General background
and experience
i
Expectations Understanding Education
regarding usefulness «— of Systems' ~«<@#—— in Systems!
of Systems! Dynamics Dynamics models Dynamics
models
Experience with
Systems' Dynamics \ Credibility
models Cooperation in of the model
building, builders
validation and/or
use of Systems' N, Considerations
Dynamics models regarding own power
¢ if Systems! Dynamics
Skills of Acceptance of model will be built
model builders Systems' Dynamics
and adequacy of models
the technique ao oT ~}o
aa ae
| 7
ae
aT Influence on Influence by
other members other members \
of the of the
organization organization
Second comes the lack of appreciation for Systems' Dynamics models’
usefulness. It is hard for many people without the specific experience in
this field to understand how these models can help to master a _ better
insight into the behaviour of systems they represent or how this insight can
be used in practice. It is also very difficult for them to understand how
certain data can be obtained and they doubt that initial approximate values
or estimates of data even if fed into the appropriate logic will be of any
use at all.
Third to be examined is the background and culture of the users, supporters
or contributors to the models. Even when the appreciation of possible
usefulness of the Systems' Dynamics models exists problems are frequently
encountered in achieving understanding of the key steps that lead to insight
~66-
into the behaviour of the system being modeled. For many people it is
difficult to understand how conceptualization is accomplished and what
criteria are used to determine which variables are to be included in the
model while others are left out. Sone often are forced to exercise
considertable effort to understand the chains of causes and effects enbedded
in a model. It is particularly difficult for some individuals to see why
certain interactions were taken into consideration and others were left out
of the model. The behaviour caused by the interaction of the various
feedback loops is certainly not easy to grasp for people without training in
systems’ analysis. It becomes even more difficult for them to associate the
observed behaviour with some causes that have dominant effects but are
located either far away in the cause-effect chain or in some more distant
past. Finally the use of data which represents aggregate, ean or most
significant values is often suspect.
If any of the so far described objections exist to any extent, the reaction
tends to take the shape of unwillingness to devote any time whatsoever to
the models' building or use.
Fourth is the frequently encountered fear that the position of power or
respect enjoyed by a person through his or her insights or knowledge of the
system will suffer as this knowledge will become shared by others after it
has been contributed to a model and perhaps enhanced and farther developed
beyond the point at which it is at the moment. This situation makes for an
outright intent to reject any efforts for the building or utilization of a
model.
Fifth is the case when there is previous precedent of unsuccessful attempts
to build a computer model or a Systems' Dynamics model in particular. The
past frustrating experience frequently makes people take the attitude "we
know this already, it won't work so why bother". This attitude can best be
described as a "negative apathy" and it is rather difficult to overcome.
WAYS OF OVERCOMING OPPOSITION TO SYSTEMS' DYNAMICS MODELS.
First it is necessary to determine the causes of opposition to Systems!
Dynamics models in each case by the different people whose support is
essential, only then is it possible to determine whether and in what way can
such opposition be transformed into some measure of support.
If rejection of the model is caused by lack of credibility of the person or
group which is to build or introduce the model it is essential that such
credibility be established by demonstrating the ability to understand the
system to be modelled and the capability to build and/or guide effectively
the users in the utilization of the model.
67%
If the resistance to the building or to the use of a model is caused mainly
by lack of appreciation for its usefulness, it will generally be necessary
to build perhaps a simple model corresponding to the situation the
prospective user is interested in and to illustrate with its help how
insight can be gained into the systems' behaviour or how the model can be
used for improving decisions. Only with very open minded people is it
immediately fruitful to illustrate the usefulness of Systems' Dynamics
models on examples which are not directly pertinent to the area of their
experience and interest.
Frequently in order to achieve understanding for the usefulness of the model
it may be sufficient first to conceptualize the various elements that need
to be considered and then to draw the cause-effect diagram which becomes a
guide to reasoning about the behaviour of the system. On relatively few
occasions does it become necessary to build a complete computer model and to
simulate its behaviour for demonstration purposes.
Once the simple model has been built one can proceed to explain how it can
be made more perfect for its scope and what sort of effort is needed to
achieve the results.
When interest in the model has been established it becomes necessary to
obtain understanding for the way in which it is built. The best way to
achieve this is to build the model together with the users. This, however,
is not always possible. When one is faced with the need to obtain acceptance
for a model that already exists it becomes indispensable to plan very
carefully its explanation. Thought should be given as to whom the model
should be explained first and who should be left for the end. It is usually
best to start with the people with greatest power or prestige within the
organization in which the model is to be used, provided that their
background and time availability are such as to make it likely that they
will understand the explanation. The explanation should be planned in a way
that will reduce the time requirements to a minimum. It is advisable to give
at least a complete overview of the model and of its behaviour in less then
an hour even if at expense of detail. The interest should be stimulated,
however, within the first few minutes not to loose the audience from the
very beginning. Technical details should in most cases be left out of the
main presentations to be taken up only during sessions with specialists.
Only when a band-wagon situation of support for the model has been achieved
is it convenient to aim at the most recalcitrant members of the organization
in the hope that their interests and attitudes may have been influenced by
others in the meantime.
One useful way of selecting a sequence with which to approach the different
members of an organization can be derived from the Fig. 2.- naturally if it
is at all possible in a given case to choose the order with which to
proceed.
-68-
Fig. 2.- It is convenient to get involved first the people near the top
provided their background is adequate to understand the usefulness
of the model.
Understanding
for the model's
usefulness
High Target for Ideal target
initial area for initial
involvement involvement
only if no
other
possibility
exists
Essential
to get involved
To be involved but perhaps after
last if at all some initial work
and support have
Low been achieved
Position of power
= or prestige within
Low High the organization
If the model building effort or the conclusions of a model are rejected
because there is fear of loosing power on the part of some individuals it is
necessary to find out whether the person refusing to cooperate may be
immediately instructed to change his or her behaviour. If this can be
achieved the cooperation might be forthcoming although grudgingly. To
improve the situation or in the case when it is not feasible to have the
person in question instructed to cooperate, it becomes essential to
understand thoroughly the motives a person has to fear losing power. On
closer examination it may turn out that his position might actually become
reinfoirced through the use of the model and through a more thorough
understanding of the situation. If this were the case it may take time and
effort to persuade a person about the need to cooperate but small steps at a
time and persistence usually render results. This approach, however may
prove too costly and time consuming to be applicable under all
circumstances.
-69-
The last, but not least important, is opposition to Systems' Dynamics models
stemming from past bad experience with models. In this case one should,
while avoiding to condemn openly the preceding experience, show how and when
a correct application of a model can be developed and what the differences
are between the proposed model and the new approach to build it as compared
with the model built in the past. The explanation, if correctly handled,
leads to autodiagnostics as to why the previous model proved a failure and
some measure of support for the new effort is a frequent result.
GAINING APPROVAL OF VAST SEGMENTS OF PUBLIC OPINION FOR MODELS RELATED TO
ISSUES OF PUBLIC INTEREST.
When a model addresses issues that require involvement of different and
numerous social groups with widely varying backgrounds and interests it
usually becomes necessary to obtain consensus regarding the model itself or
its conclusions.
Several communications strategies can be designed to achieve the above
objective. In all cases a good strategy will be the result of a careful
planning of actions on the following issues:
- What kind of response is expected from the members of the various
groups as a result of their approval of the model or of its
conclusions.
- At what stage and how will the endorsement of the opinion leaders be
obtained.
- When and in what manner will the major power holders be approached
and what kind of involvement will be sought on their part.
- Which forums will be used and in what sequence to ensure the models
discussion and diffusion bearing in mind that the prestige and
credibility of a forum may transfer prestige and credibility to the
model and to its conclusions.
- What will be the contents of the communications and how will these
contents be presented to best reach each audience and to take
advantage of the particular media through which communication will be
delivered. For diffusion to a vast public through "ass media for
exaniple, the model's basic relationships must be presented in as
simple a way as possible.
Frequently it is also necessary to establish a two way communication with
some exponents of the various groups with interests in the model or in its
conclusions. This may require considerable intellectual and logistical
-70-
resources, particularly if it is likely that a deeper analysis of the system
and subsequent modifications or refinements of the model will be needed.
An inadequate handling of two way communications is risky as serious
objections to the validity of the model or to its conclusions, if not taken
eare off with zeal, could seriously hamper the model's usefulness. This
might be the consequence either of objective errors in the model itself, or
of erroneous interpretation of the model or of its conclusions. The first
requires a correction of the model, the second must be dispelled with
clarity as soon as possible to avoid a wide circulation of incorrect points
of view which might make impact on the audience.
CONCLUSIONS.
Model builders and users have ample opportunity to improve the degree of
diffusion of Systems' Dynamics models. An organized approach to obtain
acceptance for models and for the conclusions obtained with their help may
succeed in many cases. There is, however, no unique formula that could
assure success in all situations. Each must be studied individually and an
adequate strategy for model's and conclusions' acceptance must be developed.
It is probably correct to state that in some cases at least as much care and
work is needed to obtain acceptance of the model as is needed to build it,
validate it and use it. Without the model beimg accepted, however, the rest
of the effort is usually lost for all practical purposes.
CC BY-NC-SA 4.0