Introducing GROWLab: A Toolkit for Layered Agent-based Modeling
In AGENT 2007 Conference, Chicago, IL,.
In this paper we introduce GROWLab (Geographic Research on War Laboratory) –a software toolbox to facilitate the modeling, simulation, analysis, and validation of complex social processes, with a special focus on geographic aspects. The paper aims to give a general, non‐technical introduction to agent‐based modeling with GROWLab. It focuses especially on the toolkit’s support for model structure, i.e. the creation of complex agent configurations and hierarchies. An important feature of GROWLab is that it makes possible the integration of real‐world empirical data collected with a GIS. More specifically, GROWLab can automatically create model structures from GIS datasets. This way, it is possible to run a model either on real or artificial geographies without changing the underlying data structures. We also introduce GROWLab’s GeoModel, a geopolitical template model which makes different geographic and non‐geographic datasets readily available to the modeler.
Weidmann, Nils, and Luc Girardin. 2007. “Introducing GROWLab: A Toolkit for Layered Agent-Based Modeling.” In AGENT 2007 Conference, Chicago, IL,.
@inproceedings{introducing-growlab-a-toolkit-for-layered-agent-based-modeling,
Title = {Introducing GROWLab: A Toolkit for Layered Agent-based Modeling},
Author = {Weidmann, Nils and Girardin, Luc},
booktitle = {AGENT 2007 Conference, Chicago, IL},
Year = {2007},
abstract = {In this paper we introduce GROWLab (Geographic Research on War Laboratory) \textendash a software toolbox to facilitate the modeling, simulation, analysis, and validation of complex social processes, with a special focus on geographic aspects. The paper aims to give a general, non-technical introduction to agent-based modeling with GROWLab. It focuses especially on the toolkit's support for model structure, i.e. the creation of complex agent configurations and hierarchies. An important feature of GROWLab is that it makes possible the integration of real-world empirical data collected with a GIS. More specifically, GROWLab can automatically create model structures from GIS datasets. This way, it is possible to run a model either on real or artificial geographies without changing the underlying data structures. We also introduce GROWLab's GeoModel, a geopolitical template model which makes different geographic and non-geographic datasets readily available to the modeler.},
pdf = {http://www.icr.ethz.ch/publications/weidmann2007introducing.pdf}
}