printer friendly version

Identifying the cause of operator detection failure

Organisations spend intensive amounts of capital expenditure on CCTV and there is a significant cost to paying operator personnel on a long-term basis. Having spent that kind of money, there is a major expectation from organisational management that the system has to deliver.

Where there is a failure to detect an incident by an operator that becomes widely known and contrary to management expectations, there is often a strong emotional reaction by management. It is also a blow to the pride and responsibility of all involved in the security department and surveillance function. I’ve seen examples of where top management is justified in their frustration, but also others where operators have become unfairly criticised for the failure to deliver in the detection of a particular incident.

I’ve listed some points to consider when reviewing whether an operator is at fault for the lack of detection of an incident that becomes highlighted. It is important to consider failure in the context of the whole system. Often, top management sees a video clip of the incident and has no awareness of the context of the control room and the scope of the operator’s responsibility. Something easy to detect when viewing a single evidence clip on one computer screen may prove far more difficult in the context of a multifunctional and busy control room where an operator is viewing multiple inputs.

How obvious is the incident?

• How visible is the incident? – this can be established easily by sitting in the operators chair and viewing the footage in question on the monitors in front of them.

• How quickly does it occur and could it be expected that if multiple monitors are being viewed, the incident could still be detected?

• Has the operator been trained to recognise the indicators of such an incident and can they be expected to see it? Aligned to this is does it require any specialised knowledge?

• Has the operator been trained to detect the early warning signs that something is likely to happen? If the actual occurrence of the incident takes place in a couple of seconds, this would allow them to recognise beforehand that something is likely to occur.

Is the camera displayed as part of the normal camera management strategy?

• Is the camera displayed on an ongoing basis as a risk area that should be under constant observation?

• If the camera view is displayed intermittently, is it displayed automatically as part of a camera management strategy that displays the most crucial cameras at the most important times?

• If the camera is displayed as a standard part of a display wall, how many other monitors are displayed at the same time and could the operator be expected to see the incident based on how obvious it is? If the operator has to look at 20 monitors and switch attention between these, how feasible is it to spot the incident? Is there a strong capacity to spot everything that happens within the camera views?

• Is the camera view part of dedicated viewing, or does the operator have to switch between viewing various other parts of the operation which includes only occasional views of the camera where the incident took place – what is the likelihood that the operator would have been viewing that particular location relative to others which had to be viewed as part of the operator's responsibility?

• Are there any visual analytics aids such as motion detection which would have notified the operator, and if so is the VMD distinctive or does it get lost in a large number of false alarms?

Is it possible to see the incident clearly?

• Is the size of the camera view displayed on the monitor sufficient to identify the relevant incident behaviour? A camera view displayed on a 10 cm x 15 cm monitor viewed from four metres may show a car in a bus lane, but is hardly going to allow somebody to identify discrete behaviour. This is one of the key issues in control room displays.

• Is the camera view in line of sight, or is it being shown in an area outside of the normal viewing lines?

• Is the resolution of the displayed camera view sufficient for purpose?

Failure to observe an incident could be due to an operator, technical or management issue. Also, quite simply, CCTV systems have limitations and the failure to observe something may be related to this. One would need to ask if the system is fit for purpose in the case of the specific incident, and whether the operator could have identified the incident adequately within the working environment and conditions.

What if there is an operator problem in respect of the lack of detection? If an operator lags significantly behind all the others in detection performance at any time, performance needs to be increased through training and coaching, and if this fails to lead to an improvement the person should be transferred to another area of security. I’ve seen situations where operators have not detected anything in 12 years, while their colleagues are spotting things on a regular basis. It is not fair to compare people based on a few months performance, or where there are situations where very few incidents occur, but if there is regular detection by colleagues it indicates a shortfall in abilities.

Some kind of competitive or comparative measure can also be good for performance stimulus, where competition between shifts can be used as a basis for comparison. Where you may be concerned about whether groups of operators are not performing, or there is little if no detection, you may want to consider getting an independent review of your video material offsite as part of a review process, or remotely which can be done live as part of a 'third tier' surveillance function.

If incidents are then picked up off-site, it can be used as a comparative measure. Incidents detected offsite can then be given to on-site operators as a testing process to see if they can also see the incident signs. This kind of approach can also be used for incidents detected internally by some operators where the footage can be given to other personnel to see if they can also pick it up. This is likely to increase the skills base of other operators and can be used as a refresher to those coming back from leave.

We need to balance the desire for a highly effective detection process against the fact that CCTV cannot cover everything, and operators are working in complex environments. However, inadequate operator performance strikes at the heart of the CCTV system and company viability, and we cannot abandon standards in this respect. One of the best ways is to ensure that you have the right kind of calibre of the people appointed to such positions. Performance then needs to be evaluated in context, and processes and practices to improve detection performance on a continual improvement basis need to be addressed by management.

Dr Craig Donald is a human factors specialist in security and CCTV. He is a director of Leaderware which provides instruments for the selection of CCTV operators, X-ray screeners and other security personnel in major operations around the world. He also runs CCTV Surveillance Skills and Body Language, and Advanced Surveillance Body Language courses for CCTV operators, supervisors and managers internationally, and consults on CCTV management. He can be contacted on +27 (0)11 787 7811 or craig.donald@leaderware.com

Share this article:

Further reading: