International Encyclopedia of Systems and Cybernetics

This interpretation of Systems Dynamics by E. LASZLO (after P. WEISS) is quite different from J. FORRESTER's models.

LASZLO writes: "In fact, the organic functions commonly referred to as "controls", "regulations", "compensations", and the like, come under suspicion of being instances of organic systems dynamics rather than mechanisms with predesigned and ready pathway channels. Under such dynamics the appropriate response channels form an ad hoc as manifestations of an integrated response under some such overall rule as "most economical maintenance of an ordered equilibrium state" or "attainment of a state of minimum free energy" (1974, p.118).

These views seem to relate more to the "why" of systems dynamics than to the "how" and could become a better guide to the construction of models, because they correspond to the basic nature of well connected systems.

On the other hand, this does not resolve the modelization problems of chaos in complex systems. In fact, most systems dynamics models basically postulate near-equilibrium stability, aim at maintaining it and do not have much to say when it is lost.

However, in a recent assessment of S.D., K.DE GREENE writes: "There are a number of important realworld behaviors of complex living systems that classical systems dynamics cannot completely explain. Systems dynamics should be integrated with dissipative-structure theory, synergetics, catastrophe theory, field theory and chaos theory in order better to explain and to predict evolution, the different kinds of stability and instability, structural change and structural constancy, the different kinds of equilibrium situations, bifurcations, the emergence of collective behavior and the qualitative meaning of information" (1994, p.3).

Consequently, DE GREENE states the following "questions of concern": "Does system dynamics theory as it now stands provide the necessary fit to the problems of a world that has become increasingly turbulent and unstable since the pioneering foundations were first laid? Can system dynamics be enriched with constructs from other systems theories? Can a larger metatheory, incorporating systems dynamics, be developed?" (p.5).

FORRESTER himself, in "Urban dynamics" (1969), proposed a number of auxiliary constructs that could lead to integrated system dynamics. They have been resumed as follows by DE GREENE (p.5):

"Complex systems have many levels and multiple, nonlinear feedback loops.

"System behavior is counterintuitive, that is, counter to intuition based on experience with simple feedback systems; cause and effect are not closely related in time and space.

"Complex systems are insensitive to changes of many parameters or constants.

"Complex systems resist most policy changes (changes in how information is used to determine action) because the system makes compensatory shifts.

"Complex systems can be controlled through influence points because the system is sensitive to a few parameters and to some changes in structure; a policy change may lead to the radiation of pressures throughout the system.

"Complex systems counteract externally imposed corrective programs or applied forces; corrective programs displace natural internal processes.

"Changes in a complex system can cause short-term responses opposite to long-term responses; conflict between the two can produce a worse-than-before condition; the short term can lead to long-term degeneration.

"Complex (social) systems tend to drift to a condition of low performance, because people adjust their norms in such a way that degraded performance is perceived as improvement so that they then apply more of the original "corrective" actions".

A good case-example is the generally applied policy to "improve" urban and suburban motor car traffic: more cars "need" more roads, while more roads are an incentive for more cars, in a repetitive positive feedback which would unavoidedly lead to global traffic jam in cities if some negative feedback (natural or planned) is not to introduce some limit.

DE GREENE concludes that a "larger systemic effort" is needed to incorporate in Systems Dynamics:

"1.The stochastic-deterministic, continuous/discrete/discontinuous evolution of real world systems.

"2. The family of equilibrium situations including nonequilibrium and far-from-equilibrium as well as equilibrium per se and restorable, perturbation-driven disequilibrium.

"3. The nature of critical thresholds and bifurcation points.

"4. The emergence of structural instability and true, qualitatively different structural change, as opposed to functional change mediated by change in the dominance of feedback loops.

"5. The qualitative nature of information and the evolution toward gain and loss of collective information.

"6. The overall properties and behavior of fields including order parameters" (Ibid).

This program would take in account (as a minimum) THOM's catastrophes, LIAPOUNOV's condition of stability, PRIGOGINE's thermodynamics of far-from-equilibrium systems and dissipative structuration, LORENZ, SMALE et al. chaos, and HAKEN's synergetics. This is evidently quite a tall order.