Understanding cities with agents

Dear Jim,

Thanks for inviting me to write a guest entry on your blog about my research interests and background. It is actually quite strange writing on someone else’s blog rather than my own. Since starting at the Department of Computational Social Science in August, I have been overwhelmed with how friendly and supportive everyone has been.

Prior to arriving at GMU, I spent several years at the Centre for Advanced Spatial Analysis (CASA) at University College London, both as a PhD student and as a research fellow. My PhD was carried out under the supervision of Professors Mike Batty and Paul Longley and sponsored by the Greater London Authority (GLA) Economics unit. It focused on integrating geographical information systems (GIS) and agent-based models thus providing the ability to link agents to actual ‘real’ world places and explored general questions about residential location and spatial interaction. Specifically how agents locate and interact with their surrounding environment, along with how competition for land results in distinct spatial patterns emerging. This has subsequently led me to develop my research interests in agent-based modeling (ABM) of cities.

Cities are extremely important as they provide habitats for over half of the world’s population and this percentage is expected to increase further in future decades. This increase will cause many problems such as sprawl, congestion and segregation; along with environmental effects associated with land use and land cover change. However, understanding such systems is extremely complex because of the many different factors and activities that are seen within cities all of which operate at different temporal and spatial scales. As Professor Sir Alan Wilson writes, understanding cities represents “one of the major scientific challenges of our time.”

I believe that a greater understanding of cities can be gained through the use of agent-based models: from the split second decisions involving local movements such as people walking, to the development of land over months and years, the migration of peoples over decades, to the rise and fall of cultures and civilizations over eons. These processes all that have at their core people (in some shape or form), thus understanding the reasoning on which individual decisions are made may therefore help us better understand the effects of such growth. However, there are several challenges which need to be addressed ranging from validation of such models to the communication of models. More recently I have been exploring how one can take advantage of advances in digital data, 3D modeling environments and virtual worlds such as Second Life for the creation and outreach of agent-based models.


These broad research interests have been the foundation of the graduate course that I am currently teaching this semester entitled “Agent-based Modeling of Urban Systems” which explores many aspects of urban systems from the micro-movement of pedestrians to residential dynamics. Next semester I will be teaching two classes. The first is “Spatial Agent-Based Models of Human-Environment Interactions” which explores how one can link socio economic and environmental models, along with GIS to study topics in areas such as agriculture, forestry, human/non human populations. The second course “Land-Use Modeling Techniques and Applications” focuses on a variety of land-use change models including cellular automata and agent-based models, along with exploring the drivers of change.

I am finding it great to be among a group of faculty, staff and students who are interested and extremely knowledgeable about ABM from a variety of backgrounds and I am looking forward to developing further links with people at GMU in the future.


Andrew Crooks

One thought on “Understanding cities with agents

  1. Andrew….welcome to the Institute for Advanced Study and for an extremely interesting blog entry. We'll be sure to start reading your own research blog.

    Thanks again,

    Jim Olds


Comments are closed.