On the 15 December 2009 West Midlands Police (WMP) initiated project Birmingham Safe And Sound in Birmingham city. In February this year a report was made public showing the details of this project and how it was helping to bring down gun crime in Birmingham, but recently the Birmingham Community Safety Partnership – which funded the system – decided to pull the plug because of issues over accuracy and cost.
To better understand the system, first lets look at some statistics involving gun crime in the Birmingham area.
- During 2007/08 there were 589 firearms incidents recorded for the Birmingham area.
- In 2009/10 there were 440 firearms incidents, 60 of which were assessed as involving the discharge of a firearm.
- During 2010/11 there were 468 firearms incidents, 73 of which were assessed as involving the discharge of a firearm.
- So far this performance year (2011/12), there have been 244 firearms incidents recorded, 45 of which have been assessed as involving the discharge of a firearm.
As you can see you can see there have been a fair few incidents in the Birmingham are involving gun crime, and 60-70 incidents a year where a firearm was discharged. This is where Shotspotter comes into play. Shotspotter is a system that detects gunshot noises and triangulates their position to within a 25 metre radius. This is useful for a city with a gun crime problem as it can alert the police the moment a gunshot is fired, giving them a location accurate to within 25m. It is a system that has been heavily used in 51 cities across America and has resulted in a drop in gun crime of almost 35%.
ShotSpotter uses either wired or wireless sensors deployed over large areas utilizing the principle of acoustic triangulation to locate gunfire. The computer system then uses basic triangulation calculations to give an accurate location of where the shot was fired from. The ShotSpotter website describes it as being “able to deliver instantaneous system reports to dispatchers within seconds of a weapon being fired“.
During the initial installation and calibration phase for the Birmingham area the statistics looked promising. The February report noted the following points;
- During the system acceptance testing on the 6th December 2010, the system heard 100% of all shots discharged during the test exercise and physically located 82% to within 25 metres with the furthest detection being 230 metres away from the actual discharge site.
- During the first additional live test fire, on Monday 18th April 2011, the system heard 100% of all shots discharged during the test exercise and physically located over 80% to within 25metres with the furthest detection being 200 metres away from the actual discharge site.
- During the second additional live test fire, on Monday 31st May 2011, the system guarantee was not met, as the system heard 100% of shots discharged during the test exercise incidents but only physically located 75% within 25 metres against, that of the 80% set system guarantee. In addition, some of the gunfire within the test exercise was misclassified as firecrackers rather than gunshots. It was against this specific backdrop that the decision to temporarily ‘pause’ the system was made to enable system enhancements to be made. The system remained active and alert during this period whilst additional system enhancements and fine tuning took place.
- On the 1st August 2011, a further live calibration testing event took place. This involved the use of standard and non standard weaponry. This was the first occasion within the UK that the testing of non standard weaponry had been undertaken. The system guarantee was met with 80.5% of standard weapon firing being detected. The detection rate for shots fired from non standard weapons was 93%. In respect of both standard and non standard, over 80% of the reported firing position was within 25 metres.
On the face of it it seems like quite an accurate system. So why has it been scrapped? Well the cost was one of the reasons given. It’s estimated to cost around £40,000 per year to run, which is probably equivalent to a single front line officer in terms of salary, equipment and training, maybe even less. Another reason given was the fact that its accuracy rate has dropped recently. It was found to have failed in detecting 4 confirmed shootings in the past 12 months bringing the accuracy down to below 80%.
Councillor Waseem Zaffar (Lab, Lozells and East Handsworth) said: “I was originally very supportive of this and I was keen to try and use technology to tackle this problem.
“But in January and again last month I voiced my concerns about the accuracy. It was clearly not working as well as in America.
“There is no point having a system that is nowhere near good enough. We were promised 85 per cent accuracy and we got nowhere near that.
“I think it was a wise decision to cut our losses and move on by using the resources to tackle the problem in other ways.”
A spokesman for the force said: “We’ve always said technology alone won’t stop gun crime and this technology is not necessarily reducing gun crime.
“It’s no secret there’s been technical difficulties so we’ve decided not to carry on.
“We’ve always said it’s a pilot and we’ve just decided not to carry on with it.”
During the 12 months first installation the February report indicates that there was a reduction in gun crime saying “The Birmingham area recorded 371 gun crimes (Crimsec 30) during the 12 month period December 2010 – November 2011 this is a reduction of 77 offences (-17%) compared to the 12 months pre installation.”
With those statistics in mind and given the relative low cost, it seems to me to maybe be a bit a bit early to be scrapping the scheme. Especially as the very nature of this technology means that things can only really improve, as has been the case in the US. One things for sure, it will be interesting to see the gun crime figures in the years to follow.
To read the full February report click here