Méthode de modélisation-simulation
SILVER: Software in Support of the System Dynamics Modeling Process Nathaniel Osgood, Proceedings, The 27th International Conference of the System Dynamics Society, July 2009, Albuquerque
While the System Dynamics modeling process can yield invaluable high level insights, it gives rise to a tremendous amount of detail complexity. In the course of their work, modelers must track successive model versions, the motivation for and assumptions underlying particular “what if” scenarios, and the implicit relationships between scenarios, model versions and various external artifacts such as spreadsheets, symbolic mathematics calculations, and external documentation. Failure to adequately manage such complexity can reduce the transparency, reliability, and credibility of the modeling process. While adherence to good modeling practices can aid this process, it often falls prey to corner-cutting or human error. This paper describes software that helps manage such complexity, by permitting modelers to easily access and succinctly compare historic versions of a model, by making explicit linkages between scenarios, the model versions and assumptions underlying them, and the motivations for and external files associated with model artifacts.
Supporting System Dynamics Modeling
The System Dynamics modeling process offers great value in allowing stakeholders to rise above the welter of operational detail and perform long-term strategic planning. There is a certain irony in the fact that important higher-level insights can emerge from a modeling process that frequently generates and requires management of a massive amount of day-to-day detail complexity. In the course of their work, for example, modelers must keep track of successive refinements and variations on models, the sets of “what if” scenarios already explored, and the relationships between those scenarios and multiple other entities: The versions of models that were used to create them, the particular assumptions that underlay the scenarios, the intensions underlying those choice of assumptions, insights gained from those scenarios, and often implicit links to external artifacts that aided in determination of parameter values or that further analyze scenario results.