Location matters

A look at how we’ve used innovative techniques to tackle persistent hotspots of violence across the force. This story is one of a series that make up our Strategic Needs Assessment. To read the rest, visit tvvru.co.uk/strategic-needs-assessment.

Thames Valley Police is committed to reducing violent crime and this year has introduced an innovative and trail-blazing experiment attempting to reduce such incidents in areas identified as significantly impacted.

Academic and operational research has provided evidence-based conclusions indicating that crime is less likely to occur if an area has noticeable ‘high visibility’ police activity (this peaks with patrol times of around 15 minutes up to 4 days apart).

As a process hot spot policing is not new however, this trial is by far the most stringent and we believe accurate implemented within the force to date. The identification of these areas comprised a multi-step process starting with just under 145,000 hexagons being laid over the force area, then 59,000 (going back several years) incidents were mapped and split into 2 x 12 hour periods covering days and nights; these were then aggregated into the hexagons, which gave the initial ranking of potential areas of interest.

Each hot spot was then assessed and sanity checked for such things as location, was it centred on a prison etc. and as such could be removed. Could the hexagon be rotated or nudged slightly to incorporate more incidents and/or to negate an area impossible to patrol such as a dual carriageway or lake.

Finally, to ensure the hot spot was enduring various geo-spatial techniques were used to determine the final set of hot spots ranked for both day and night shifts.

The next stage of the experiment was where the force moved away from the hot spot policing ‘norm’ as we wanted to be able to compare, in detail, effects of patrols, allow officers to self-task throughout their shift and be able to compare activity within the hotspots on days they weren’t patrolled, we had to innovate!

The first part meant we had to introduce a randomisation process where hot spots were scheduled for patrols (treatment) or not (control) this did not mean however that on days when randomised as ‘control’ hot spots officers stayed away from the designated areas, activity was classed as ‘business as usual’.

Next came the tasking mechanism, an in house application was developed which offered officers a list of hot spots that had been randomised as ‘treatment’ during their shift. When officers started a patrol location details were recorded automatically, at the end of the patrol officers were able to record 3 metrics detailing what interactions with the public they had undertaken during the activity.

A screenshot of our hotspots policing application, showing dummy data. A map is shown which identifies the hotspot, along with text telling an officer where to go and what to do.
A screenshot of our hotspot app (showing dummy data).

The next part of the puzzle was being able to measure activity within the hot spots when randomised as ‘control’ to do this we used the existing police radio data to analyse business as usual activity within these areas.

The final part needing to be addressed was monitoring, this was achieved pulling all the recorded data together into Operational Dashboards that could be used by staff to monitor all aspects of the experiment.

At the time of writing, the experiment is still underway so no results are available.