Is it possible, thanks to research and technology, to manage urban security in a more efficient and effective way and also acquire elements useful to predict crimes before they occur? This might seem a fantasy of some futuristic film; nevertheless, it is a possible scenario due to “eSecurity – ICT for knowledge-based and predictive urban security” (eSecurity), a European project coordinated by the eCrime research group of the Faculty of Law – University of Trento (Università degli Studi di Trento), in partnership with the ICT Centre of Fondazione Bruno Kessler (FBK), Questura di Trento (i.e. Trento Police Department), and the Municipality of Trento. The project was co-funded by the European Commission under the ISEC program (2011) “Prevention of and Fight against Crime” of the Directorate General for Migration and Home Affairs, and lasted 36 months: from November 2012 to November 2015. eSecurity, piloted in the city of Trento (Italy), is one of the first projects worldwide on predictive policing and the first-ever project on “knowledge-based and predictive urban security”.
Project eSecurity is based on principles pertaining to rational choice theories of crime and environmental criminology. It assumes that, in any urban environment, crime and deviance concentrate in “some areas” (streets, squares, etc.) and that past victimization predicts future victimization. This concentration of crime in time and space results from the concentration of opportunities and causes in time and space that need to be investigated in order to impact on crime within urban territories and to manage urban security issues extensively (Brantingham and Brantingham, 1991). As a consequence, it is knowledge of these “hot spots” and existing criminal opportunities within urban contexts that enables identification of the underlying criminological factors to be considered so that efficient and effective preventive and counter strategies can be deployed (Clarke, 1997; Wartell and Gallagher, 2012). Accordingly, eSecurity builds on pilot schemes for predictive policing carried out with reference to the United States and United Kingdom: for example, the projects conducted by IBM together with the University and the local police of Memphis (USA); by the University of California of Los Angeles and Irvine with the local police of Los Angeles (USA): by the Jill Dando Institute of Security and Crime Science (University College of London); and by the police of Trafford, Greater Manchester (UK). “Predictive policing refers to any policing strategy or tactic that develops and uses information and advanced analysis to inform forward-thinking crime prevention” (Uchida, 2009). More in detail, predictive policing involves the analysis of police data on past crimes, their spatio-temporal locations (geo-referenced reported offences), and recurrences in the behavioural patterns of offenders. The purpose is to forecast places and areas of future concentration of crime within urban territories, with the ultimate goal of allocating police resources and efforts optimally (RAND, 2013).
In order to develop a new urban security management model with which to predict and prevent future concentrations of crime and deviance, the eSecGIS geographic information system, with its related predictive algorithms, has not only analysed the past, geo-referenced and anonymised data on crime events collected in the SDI (“Sistema di Indagine” – Investigation System) database of the Italian Ministry of Interior and stored in eSecDB. It has also integrated other geo-referenced socio-demographic and environmental data from the smart city: for example, neighbourhood street lighting, weather conditions, the distribution of local businesses.