“Boundless” modelling

1In 2006 in Vietnam, a team worked on the simulation of a bird flu epidemic. The objective was to visualise the spread of the H5N1 virus through villages and understand the impact of political decisions on the progress of the epidemic. The team found that the observed spread of bird flu epidemics can only be simulated if the virus is capable of surviving in water, not just among wild bird populations. The veracity of this condition was confirmed by laboratory studies, demonstrating the relevance of an approach that the team later applied to the management of emergency relief within the context of natural disasters.

Modelling of the Red River flood in Hanoi.

2This change of paradigm and application encouraged researchers to find a form of generalisation shared by the different types of event studied. This is why they decided to create multi-purpose software capable of modelling a potentially infinite range of dynamic phenomena via the programming of IT objects referred to as “agents”, in areas as diverse as epidemiology, land use changes, the management of shared resources and urban mobility.

3In 2010, they brought GAMA on line (http://gama-plaform.org), a free software program for the modelling and simulation of spatialised phenomena. It uses a unique and intuitive language, GAML, specifically designed by the team to be easily operated by non-experts. Researchers intended to make modelling accessible to a wider audience and put this model-based approach within everyone’s reach, in particular by providing visualisation tools enabling the in-depth analysis of the phenomena simulated, in the form of animated maps, realistic 3D representations and serious games, where artificial agents are replaced with actual individuals capable of interacting with the simulation. GAMA is currently used by approximately 2,000 modellers worldwide.