[Construction of Dynamic System Models Using Interactive Computer Graphics, Walter Hudetz, p.266-299]
p.266 The construction of system models and their simulation on digital computers has been
an important part of systems analysis in the physical sciences for several years now. More recently this method is
also being used in the social and life sciences.
p.285 The development of models for the simulation of dynamic systems is a complex and time-consuming process.
[Matching Man and Model: The 'Guide' System for Interactive Model Handling, Michael G. Strobel
and Hartmut Bossel, p.309-354]
p.311 The models to date [JLJ - the author is speaking of an attempt to model the Earth
and the future consequences of exponential population growth and limited resources] are exclusively controlled
by the model builder and programmer. Inevitably the models reflect his bias and perception
of reality.
p.314 Dynamic models are computer programs which represent, in the form of mathematical
equations, functional relationships between components as they are observed or perceived in the real world.
p.314 Systems in the real world tend to be complex and the model builder can reflect this complexity in
his model by stating all the relationships he can think of or which he considers relevant to his model of reality.
p.315 It is the possibility of letting a great many relationships influence each other
under precisely stated assumptions and of determining the consequences, which gives the computer model its enormous
power and advantage over conventional planning methods
p.315 The speed with which the great number of calculations are accomplished enables one to experiment
repeatedly with different assumptions for the future in different parts of the model, i.e. with different "scenarios".
The time series for the assumed future values of the parameters with the most likely outcome, or the outcome with the least
surprise, are then arbitrarily termed the "standard scenario".
p.315 Computer simulations have for many years been extensively used in the technical domain in order to
gain experience with the behavior of complex systems under various conditions... The use of the flight simulator
in pilot training and aircraft development allows the pilot or engineer
to completely familiarize himself with the various responses of a complex system without the threat of catastrophic failure.
Here the simulation is analogous to the socio-economic systems we are presently interested in: The complexity
of the various interactions is impossible to foresee with any accuracy by natural means, and the consequence of errors is
disaster.
p.315-316 The construction of valid computer simulation models of socio-economic systems is a costly and
time-consuming enerprise... Given a valid model, however, one then has a representation of reality with the
advantage that it is clearly stated, reduced in complexity, and managable. With this instrument it is then possible
to examine the likely behavior of the real system in response to various conditions, limitations, and interventions.
In principle, the user can experiment with the model... by... formulating and applying "policies" until he has a firm
grasp of the characteristics of the system and a thorough knowledge of its behavior potential.
[An Interactive Program for Energy Policy Assessment, Hartmut Bossel, Pieter van der Hijden, Walter
Hudetz, pp.355-391]
p.379 The program [a structural model of the energy supply system of a given region or country] has been
implemented both on the Siemens 4004/151 computer system and on the PDP 11/40 minicomputer with 28 K word usable memory...
Assessment of stored energy scenarios takes approximately five minutes per time-step or roughly an hour for the twelve time-steps
involved from 1970 to 2025.
p.380 the modeler's first responsibility is the creation of a reliable and valid model
[Man-Machine Interaction as an Answer to the Multiple-Goal Problem, Christophe Deissenberg, p.392-431]
p.396 The important issue, however, is not whether the proposed solution is optimal, but whether the method
permits an improvement in the solution significant enough to justify its use.
p.421 One cannot intuitively relate the optimal solution to the constrained problem. Hence, as a rule, an
objective function does not have a meaning intuitively. Decision-makers do not think in terms of overall objective functions,
but in terms of the desirability of a solution or of a change. The interactive procedure exactly interprets such
desirability information to help the decision-maker recognize the best feasible alternative.
