LAUSR

Deliberate fires are a very common problem affecting all countries around the world. They create a high sense of insecurity within communities, consuming and straining many resources (human and financial). Yet, despite various attempts, significantly tackling and reducing deliberate fires has remained largely ineffective, mainly due to the case-by-case approach implemented in responding to these incidents. Drawing on the repetitive nature of some types of deliberate fires, it was shown that adopting an intelligence-based approach is promising in tackling and reducing repetitive deliberate fires. This paper presents a two-fold procedure developed to produce intelligence on a dataset of fire events that were either deliberate or unknown in origin. Firstly, through the creation of a relevant dataset (which is a peculiar problem due to the specificities of the event of fire) and secondly through the application of specific analyses. This procedure was implemented on a dataset of fire events collated from a nine-year period in the State of Geneva, Switzerland. Results show that rudimentary data and simple processing can already generate valuable intelligence, often unsuspected until then. These results provide responding agencies with a clearer understanding of the problem, which can also support their decision-making process. This study proposes the basis for the development of an integrated real-time intelligence process. Such a process would allow the systematic and real-time monitoring of fire events in general and deliberate fires in particular by providing an immediate view of the problem, detecting recurrent events and revealing linkages between cases indicating repetitions. In terms of policies and governance, such a study should encourage institutions that deal with fires to collectively reshape their objectives, share data and analyses, and coordinate their actions to reduce harm.