Search site

School of Geography

Capturing and simulating criminal behaviour through advanced spatial analysis and agent-based modelling

Supervisors: Dr Nick Malleson and Dr Alison Heppenstall

Classifications of criminal behaviour are largely limited through one of the following: size of the group of individuals; poor consideration of spatio- temporal movement and often poor, or no linkage to - geodemographics or other indicators of socio economic status. This project will apply advanced spatial analysis methodologies to a database of tens of thousands of individual reported crime events over a 5- 10 year period of time to develop a more realistic and detailed classification of criminal behaviour based on spatial movement. The classification will be further enhanced through the inclusion of literature and other national data sets such as the Index of Multiple Deprivation and those provided by local government such as unemployment data, abandoned buildings, council administered benefits and health status information.

This new behavioural classification will be used to inform the construction of a customised individual - level synthetic population. This, in turn, will be incorporated into an existing individual -level computer model to test whether the patterns of criminal behaviour/social unrest are more effectively captured than present techniques. For example, if a new housing development is constructed, can we estimate the potential for attracting burglars?


The overarching aim of this PhD is the creation of a new classification of criminal behaviour that will be used for creation of a customised population for an existing agent - based model. The objectives can be broken down into:

  • A critique of existing classifications and an identification of their shortcomings in terms of grouping types of criminal behaviour
  • Detailed spatiotemporal analysis of the known movement of criminals;
  • Linkage of movements to geodemographics to create a new behaviour classification
  • Testing of the new classification through (i) creation of a customised population and (ii) benchmarking against an existing models

Applications are invited from candidates from either computational science or social science backgrounds; candidates with a good understanding of criminology or socioeconomics are welcome, and training in programming, GIS and agent -based modelling can be provided


For information on funding opportunities click here


For project related enquiries please contact the supervisors.
For application enquiries please contact Jacqui Manton