This presentation will explore how agent-based modeling coupled loosely with geographic information systems (GIS) is becoming a powerful technique to study cities. It allows us to grow social structures in an artificial world, and explore urban phenomena such as segregation or residential location. The models presented include the geometric aspects of cities and result in outcomes that differ from the more traditional regular representation of space (i.e. cells).
Problems and challenges with this approach and ABM in general will be identified. Following this I will argue that there is a need for fine scale and extensive datasets of the built and socio-economic environments to ground such models. To this context, I will introduce a detailed housing and built environment database for London which is currently underdevelopment, and explore not only its use for spatial analysis but how it can potentially be used as a set of building blocks for agent-based models focused on residential and urban issues which affect many cities.
Not only is there a need to make agent-based models operational but also we need new ways to visualize and communicate such models, especially to those whom we seek to influence and whose activities we believe can be informed by such modeling. This will be explored through using advances in computer technology, specifically Web 2.0 and Second Life. Such technologies provide outputs to which non-expert users can easily relate, thus allowing them to come under greater scrutiny than was possible in the past.
To conclude, I will highlight recent research interests such as crowdsourcing, which allows us to collect people's perceptions about specific events such as the current financial crisis and explore whether this type of information could be useful for agent-based models.
Further information about this research can be found at: http://gisagents.blogspot.com/
