Models‎ > ‎Simpop2 Generic Model‎ > ‎


The model environment is composed of information elements that are not part of the system's description, but which are necessary for adapting the simulation to a specific theoretical or geographical situation. It has two acceptations :

  • The geographical environment includes elements which are outside the geographical system of cities (information about the rest of the world, which may interact with the dynamics of the system)

  • The computational environment includes elements which are not part of the agents system (i.e. exogenous variables, global parameters and rules)

- Main components of the system's environment -

Initial situation : The "map" in our model defines the type of geographical space in which the city-agents interact. It shapes the interaction space of the agents. It includes their location, absolute and/or respective to territorial borders, their initial size and resources (functions). This initial situation may be a theoretical (regular or random) or an observed spatial distribution of cities. The observed maps include an implicit accumulated information since they contain the past dynamics of the urban system.

Type of urban system : A country is assigned to one of the four main types of dynamics, each one corresponding to a different version of the SIMPOP2 model. These main types are the old settlement systems in industrialised countries, recent settlement systems of industrialised countries, and old or new systems which were reorganised through colonisation and which belong to emergent or developing countries.

Exogenous parameters : They define the dynamic forces that complete the dynamics created by the interactions between the agents in the model (magnitude of general demographic and economic growth, qualitative typology of innovations in terms of urban functions, quantitative levels of productivity, supply and demand for each function). They also include parameters necessary to model the exchanges with external geographical entities (cities, countries, trading flows..) that are not directly part of the studied system but may affect its dynamics (for instance through international functions).

General knowledge about the dynamics of urban systems also may be considered as part of the environment of the model, since this multi-agents model is exceptionally rich in information injected from the large databases built by the Geographie-Cités team since several years.