Measuring Occupancy for Safety Applications

The ability to measure and report Occupancy figures for various safety related projects is an often requested feature.

Applications such as maximum floor loading, with an automated system to calculate occupancy and alert administrators when the occupancy is approaching the maximum allowed level, would clearly be a use case.

Similarly, there are many public places that could use such a system to aid with managing for over crowded and to ensure that fire evacuation procedures can cope if the unfortunate did happen. Indeed, there have been many unfortunate incidents reported where over crowding has resulted in loss of life, and it is therefore becoming more and more common for places like night clubs to have some kind of occupancy reporting system in place before they will be granted a license.

At this point it is vital that the inherent issues with calculating occupancy are fully understood. See the issues already discussed here: Occupancy Issues.
As well as the known inherent issues with calculating occupancy, it should also be pointed out that in an emergency situation, count accuracy would be further affected by frightened people crowding and/or hurrying out through exits. This will mean that the calculated occupancy value will be out further.
Lastly, some parts of the counting system could become damaged or inoperable in certain types of emergencies, (i.e. fire).
It is also extremely important to point out that most of the authorities who are specifying occupancy reporting systems before they will issue licenses for buildings and events, often do not know the inherent issues related to occupancy calculating.
Because of these issues, and the overall inaccuracies of any counting system, and the other problems discussed elsewhere, measuring occupancy for use in safety critical applications is strongly discouraged.

 

Evacuation Counting

Evacuation counting allows incident managers to supply emergency services with fairly reliable information about how many people might still be inside a building; what exits might be blocked; etc. 

While evacuation counting can be considered a safety application, it is different in that it essentially starts from a known (usually fairly accurate) occupancy count in the event of an emergency, and therefore the majority of the issues relating to the inaccuracies in counting systems are reduced significantly.
The known occupancy count is usually derived from an alternative system such as a card operated turnstile or barrier, and because these systems work on a ‘one card - one person’ principle, they can provide a very accurate count of people entering and leaving the building and an occupancy figure derived from this is inherently very accurate.

Because barriers restrict people entering and exiting, these have to be fully opened during an emergency and so cannot be used to count people out during the emergency situation. Additionally, people will also leave via fire exits, so in an emergency, the task of counting is switched over to people counters installed at the fire exits and over the (now open) barriers instead. 

The evacuation system effectively counts down from the ‘known’ occupancy figure, at the time the emergency was declared. The principle being that when everyone has evacuated the building, the occupancy figure should be at, or around, zero, and any numbers significantly higher than this could indicate that people are trapped somewhere in the building.

But note that anyone who bypasses the access control system to enter or exit the building (visitors who are ‘buzzed’ in without a card, staff who tailgate other staff, etc.) will not be known to the access control system, and therefore will not be included in the 'known' occupancy count at the start of the emergency. This, combined with the other known issues with occupancy counting, mean that such a system should only ever be used to give a rough idea of the seriousness of a particular emergency, (e.g. there are 20 people not accounted for rather than 200).

An evacuation counting system should be considered a ‘better than nothing’ solution with known, inherent, issues and should be specified and operated as such. Potentially life-threatening decisions to enter a building should NOT be made solely on the reported occupancy value.
In all cases a proper roll call procedure should be incorporated and operated as soon as practically possible after the start of the emergency. 

As long as the problems associated with people leaving in a hurry and potential damage to the system during an emergency (as mentioned above), are fully understood, a workable evacuation counting solution may be achievable, but in all cases, end user expectations should be carefully managed.