Crime is a seemingly ubiquitous feature of modern urban environments.  However, it is clear that crime is not distributed evenly throughout the urban landscape.  Rather, crime has long-been known to cluster tightly in time and space forming so-called crime hotspots.  Associate Professor of Anthropology, P. Jeffrey Brantingham, leads an ISR project on Mathematical and Computational Modeling of Crime that seeks to understand the fundamental mechanisms that drive the emergence, spread and dissipation of crime patterns and how various policing strategies may be used to disrupt crime pattern formation.
 
Contrary to popular perception of crime as exquisitely planned and executed by professional criminals, most crimes are crimes of opportunity.  Offenders search for cars to steal and houses to break into only incidentally as they go about their regular daily routines.  Gang shootings and bar fights that end in  a serious assault or even homicide are just as likely to result from chance encounters as from pre-planned activities.  Crime patterns are in large part driven by how offenders and victims move and mix against a structured and complex environmental backdrop.
 
With support from the National Science Foundation, Brantingham and colleagues Andrea Bertozzi and Lincoln Chayes (both from UCLA Mathematics) and George Tita (UC Irvine Criminology, Law and Society) teamed up with the Los Angeles Police Department to use contemporary mathematical methods to understand crime pattern formation from first principles.  The models developed have been able to show that very local spatial diffusion of risk, a well-documented empirical phenomenon for burglary, is sufficient to drive the emergence of crime hotspots.  Moreover, new results suggest that different types of hotspots respond in different ways to police suppression, sometimes being displaced to adjacent areas but at other times dissipating without reemerging.  These results are being used to help develop crime forecasting and geographic profiling tools under a new initiative on “predictive policing” being spearheaded by the LAPD.
 
More information is available at http://paleo.sscnet.ucla.edu