Cause for alarm

The problem of false alarms within the UK security industry has been well documented, but to date there has been a worldwide lack of rigorous research on the causes of false alarms. In fact, policies on alarm management have been implemented in something of a knowledge vacuum, using data provided by the Association of Chief Police Officers (ACPO). This data has driven subsequent policies of police response, equipment manufacture and installation policy, underpinned by insurance risk management.

In an attempt to redress the balance, and to further reduce the direct and indirect costs on police forces, alarm users, alarm companies and society in general, the NSI has conducted the world's first academic study of the causes of false alarms, providing a detailed insight into false activations and therefore, the first rigorous baseline for improvement.

As its basis for study, the report focuses on three alarm companies, two of which provide a national coverage. It looks at all recorded false alerts, not just those reported to the police, and the researchers also looked at the interaction between alarm company and subscriber, including the ways in which false alerts are classified.

Policy drive
Around nine out of every 10 alarms notified to the police are false. Figures show that in the UK only one in 208 activations results in an arrest, and on average police attend 17 genuine activations for every intruder they arrest. The research aimed to understand the day-to-day dealings between subscribers and alarm companies, especially those aspects of the relationship that lead to false signals.

While increasingly strict police policies have resulted in the propensity of individual alarm systems causing false alarms to be reduced, the increasing popularity of intruder alarms has maintained the absolute number of false alarms at a level that severely tests police resources. Alarm companies, receiving centres, subscribers and their employees have also felt the impact of direct and indirect costs associated with these false alarms.

The UK's police policy to reduce alarms has centered on the threat of removal of police response, while in the US, users whose systems offend have been fined, and users deemed to have been negligent in their use of alarm systems have had to undergo mandatory training.

Subscriber error
When inspection engineers were asked about specific causes of false activations, there was almost universal agreement that errors had occurred on entry and exit from premises. Line faults were also viewed as fairly common, although these were much less likely to be attended by police.

The three case studies confirmed these findings. In all three, subscriber error was the largest cause of false alerts, accounting for around half the number of incidents. Most of the errors focused on entry to and exit from the premises within a peak period between 6.00-9.00 am.

The subscriber errors fell into four main types:

• use of wrong entry/exit route

• deviation from the correct route

• elapse of time before disarming the system, and

• incorrect entry of code at the keypad.

After user error, equipment malfunction and poor siting of detection devices presented the largest contribution to false alarms. In both A and B case studies, faulty equipment was blamed for a significant number of false alerts and policed false alarms. On the basis of fault attribution by Company A, faulty equipment was second only to user errors in contributing to the level of policed false alarms. Alerts due to faulty equipment are more likely to be policed because they often happen when the alarm is set and therefore when no human is likely to intervene and confirm that there is no intrusion. However, since the introduction of DD243, it is expected that false alarms caused by equipment failure are likely to show a significant decline.

The report also found that manufacturers of movement detectors have had to take more than their fair share of the blame for false alerts. Around 95 per cent of movement detectors returned to manufacturers as faulty were found to have no fault. Subscribers were also identified as requiring better training, in particular for refresher training when systems are updated.

The report shows that good practice, aimed at reducing false alarms, varies, but more can be done to tackle false alarms.

It puts forward a series of recommendations for debate and consideration within the industry:

• entry and exit procedures need to be improved – the introduction of DD243 is thought to have recognised this problem and addressed the issues involved with a series of measures, but further data is required on its effectiveness

• alarm companies should involve suppliers in the false alarm management process – companies need feedback about the extent to which they incorrectly blame faulty parts for false alerts

• policies for dealing with subscribers need to be reviewed – some companies have suggested introducing punitive measures against customers who negligently cause false alerts but this may be difficult in practice. More systematic training of subscribers is also advocated, sometimes by subscribers themselves.

• company policies need to be reviewed when systems are taken over – for example, companies are not always informed of changes of ownership. The criteria for sending out engineers also needs to be reviewed. There is also wide variation in practice for remote resets and the existing criteria needs to be more rigorously applied.