Name: Nick Malleson
Institution: University of Leeds
Research: Modelling criminal patterns using the NGS
Would you like to know the risk of your house being burgled? Would you like to know this information before you moved house? Well the NGS and Nick Malleson, a PhD student at the University of Leeds, may be able to help.
Nick is building an application that could be used by local authorities to predict the effects of new environmental developments or policies such as housing developments or improved transport networks. These types of developments affect communities as well as the physical environment and can increase or decrease individuals’ risk of becoming a victim of burglary. There is little work that goes into simulating the effect that these changes will have on crime rates, so a good model could pinpoint potential crime hotspots or other problem areas that are difficult to foresee otherwise.
Occurrences of crime are complex phenomena, driven by a vast number of interrelated elements that can include environmental factors as well as complex human behaviours. Traditionally crime occurrences have been modelled using statistical techniques and although such approaches are useful, they face difficulties with including large amounts of highly detailed individual-level data and the integration of behavioural information.
Nick is using agent-based modelling which is a new modelling paradigm to overcome these difficulties. An agent is an independent component of a system that interacts with other agents and its environment (such as a virtual person situated in a virtual city). Thus large systems of agents can be created to mimic real scenarios. Most importantly, the agents can incorporate behavioural information to determine how they should achieve their goals.
An accurate agent-based model that incorporates human behavioural factors and detailed environmental components could have a number of uses. Not only could it be used to analyse offender travel, but also provide ‘what if’ analyses, such as the effect that new environmental developments or crime reduction initiatives might have on a local area before their implementation. This could have a significant impact on local councils, town planners and the police.
The burglary simulation model is written in Java using the Repast Simphony agent-based modelling toolkit. Using MPJ Express message passing, multiple compute nodes are utilised to run separate models simultaneously. As with many agent-based models, large amounts of data are created which are stored in an NGS Oracle database for analysis during a model run and after it has finished.
The agent-based model was very large and therefore computationally expensive with each run taking days to complete on a desktop PC. This is where the NGS proved essential for the project to be feasible. Hundreds of identical simulations can run simultaneously on different nodes giving hundreds of results in only a few days.
“Without the use of NGS resources the project would not have had the computational power it required to generate reliable, robust results” explained Nick. “Adapting the model – which was originally designed for a desktop PC – to run simultaneously on hundreds of NGS compute nodes simultaneously was extremely easy, due largely to the ease of access and the excellent support from NGS staff.”
Image: GeoTime(R) software used courtesy of Oculus Info inc. All GeoTime rights reserved.
Project funding - National e-Infrastructure for Social Simulation (NeISS) (JISC grant no. 95551620).
PI - Prof. Mark Birkin
Download a summary slide of this case study
