• agent-based modelling
• crime prevention
• decision support system
• environmental approach to security
• social behaviour
• space syntax

This thesis delves into the intricacies of urban security from an environmental perspective, examining how environmental conditions, crime opportunities, and the fear of crime interact to define spaces as more or less secure, or perceived as such. Structured in three parts, the research tackles the topic from theoretical, model-based, and practical viewpoints.

The first section, entitled 'Space and Crime: an Impossible Literature Review', offers an updated literature review on the environmental approach to security. It traces the origins of the urban security concept, following the evolution of the approach from its founders and the developments in environmental criminology to the third generation of Crime Prevention Through Environmental Design (CPTED). This review outlines the theoretical, methodological, and practical challenges encountered in both academic and public institutions, which undermine the effectiveness of the environmental approach to security, thus presenting a framework from which to address the issue of urban security.

The second part, 'Modelling Human Behaviour: Spatial Crime Impedance and Crime Opportunities', focuses on modelling human behaviour in urban environments and interpreting behavioural metrics through a criminological lens to investigate crime dynamics. It introduces 'spatial crime impedance' as a configurational measure of urban environments that identifies their intrinsic ability to deter specific crimes. Employing both top-down (Space Syntax) and bottom-up (Agent-based Modelling) methodologies, it proposes new analytical tools for exploring crime dynamics. This section also engages with the crime dataset used, addressing the issues of data collection, cleansing, categorization, spatialization, and the suitable digital-cartographic visualization, presenting four alternative spatializations of data that highlight the potential and limitations of various systems of collection, usage, and sharing, which severely impact the implementation of environmental approaches to security policies.

The third part, 'Pisa Unveiled: Bridging Theory, Modelling and Practice', applies the tools developed in Part II to the case study of Pisa and integrates them into a comprehensive study representing a potential multi-scalar, multidisciplinary, collaborative, and participatory decision support tool. This tool should be available to local administrations to scientifically and consciously address ii
the issue of urban security. This final section demonstrates how the environmental approach to security can reveal criminal dynamics and inform a conscious and collaborative urban planning among all stakeholders involved, enabling the formulation of policies aimed at crime prevention and, more broadly, the creation of safer, more sustainable, and liveable cities.