Response to fire alarms

http://irc.nrc-cnrc.gc.ca

R e s p o n s e t o f i r e a l a r m s

N R C C - 4 9 2 7 2

P r o u l x , G .

A v e r s i o n o f t h i s d o c u m e n t i s p u b l i s h e d i n

/ U n e v e r s i o n d e c e d o c u m e n t s e t r o u v e

d a n s : F i r e P r o t e c t i o n E n g i n e e r i n g , n o . 3 3 ,

W i n t e r 2 0 0 7 , p p . 8 - 1 4

Response to Fire Alarms

The way people respond to a fire alarm depends on a number of factors. One of the key factors is the type of building where the alarm is activated. Frequently, a fire alarm activated at home will lead to a very fast response by the occupants. In comparison, the response in an assembly building could be expected to be slow and in some instances occupants may completely ignore the signal and pursue their activities [1].

In fact, it is not so much the type of building that makes a difference but the occupant’s perception of their role in the building. When the smoke alarm goes off in their own home, occupants feel that it is their responsibility to take action. Within seconds of hearing the alarm, the occupant will take action attempting to identify the problem that led to the activation of the alarm.

In an assembly building such as an airport, a cinema or a shopping mall, the activation of a fire alarm without any additional cue may not trigger any particular response from occupants who are visiting these premises. Again, it is not so much the type of building that explains this lack of response but the role and responsibility occupants perceive they have in such venues. As visitors,

occupants will possibly continue their activity, particularly if other visitors are also not paying attention to the alarm. Visitors act as visitors; they are waiting to be told and directed by staff if something is expected from them. On the other hand, occupants who are employees in assembly buildings are much more likely to take action after the activation of a fire alarm. Particularly if these employees feel they have a responsibility to take specific actions during an emergency and if they have been trained for such situations.

Consequently, the role people feel they have in a specific venue is an important indicator of a potential response to a fire alarm activation [2]. What this response might be though can vary widely depending on the person’s characteristics, which includes training and past experiences.

Our high expectations

Fire alarms were first introduced in buildings several decades ago from industrial environments to every apartment; sometimes in every room in residential

buildings. Fire alarms have been seen as the best way to fulfill four main objectives:

1. Warn occupants of a fire 2. Prompt immediate action 3. Initiate evacuation movement 4. Allow sufficient time to escape

In the first place the fire alarm is expected to warn occupants of a fire. In some conditions, the fire alarm may detect a fire prior to any person directly perceiving it with their senses. This is particularly true for interconnected detectors in large buildings. Assuming the alarm is audible and occupants can hear it, it should act as a warning that something has activated a detector. This is a few steps away from concluding that there is a fire. People rarely make the direct connection that the fire alarm went off so it means there is a fire unless there are additional cues [3]. When the fire alarm goes off, occupants are more likely to question ‘what’s going on’, ‘what’s wrong’ instead of concluding that there is a fire and they should be evacuating the building.

Another objective of the fire alarm is to prompt immediate action. As in our example above for the residential fire alarm activation; when occupants feel they have a role of responsibility toward the property or its occupants, immediate actions will be taken after perceiving a fire alarm. Most times the initial action would be to investigate the reason for the alarm activation, unless the situation is evident [4]. A variety of follow-up actions might take place from fighting the fire, warning others, retrieving belongings, or calling the fire department. These actions take time and may endanger occupants. Although such responses are not recommended behavior, they are nevertheless fairly common. In some situations, however, if occupants do not feel a role of responsibility, if their interpretation of the situation is that there is no threat or if others around are paying no attention to the alarm, they may decide to ignore the warning. This is a potential failure of the alarm signal: not to trigger an immediate response. The ultimate action expected from occupants when the fire alarm is activated is to initiate evacuation movement. Unfortunately, this is probably an objective that is rarely met. The activation of the fire alarm by itself is rarely sufficient to trigger the evacuation of a building. Additional cues such as a smell or sight of smoke or flames are needed to certify the credibility of the fire alarm. If such fire cues are not present, incentives such as instructions from fire wardens, messages through the voice communication system or on hearing the horns from the arriving fire trucks will give credibility to the fire alarm and will motivate most occupants to start their evacuation. On its own, however, the fire alarm rarely triggers an immediate evacuation of a building. In buildings where training takes place regularly it is possible to obtain a fairly fast start of the evacuation

movement upon the alarm activation during drills. When the fire alarm goes off in an elementary school for example, it is standard practice for all pupils to leave in ranks with their teachers and gather on the playground. Such evacuation drills are completed in record times but they are only practices and may not unfold in such a timely fashion during an actual incident.

Overall it is expected that the fire alarm will provide an early warning to

occupants allowing them sufficient time to escape. This is an objective that is getting more and more difficult to accomplish. Traditionally, it was assumed that upon smoke alarm activation, in a single family house, that occupants had an average of 17 minutes before untenable conditions would be reached [5]. Recent studies conducted by NIST have reduced this elapse time to escape to

about 3 minutes. In essence, it appears that new construction design and new materials introduced in residential construction in the last few decades are creating fires that burn more rapidly and which are more lethal. The smoke alarm, when present, in workable condition and heard by the occupants’, remains essential but even so would only provide a very short window of time to escape. Although fire alarms definitely save lives, their capacity to provide enough time to escape may be jeopardized by ultra-fast fires. Fire alarms in other type of facility, however, could provide varying elapse times to escape depending on the fire, the building design and the location, and the capabilities of the occupants.

Occupants Fail to Respond

There are ample anecdotal evidences that occupants often fail to respond to fire alarms particularly in facilities other then a private apartment or a house.

Occupants may ignore the fire alarm signal for different reasons. There are four reasons that are particularly interesting to explore:

Failure to recognize the signal as a fire alarm Unaware of the proper response

Loss of confidence in the system because of nuisance alarms Failure to hear the signal

Failure to Recognize the Fire Alarm Signal

One explanation for the lack of reaction may be the occupants’ failure to

recognize the signal for what it is. They may mistake the fire alarm for another signal, such as a burglar alarm, an elevator fault warning, or a security door alarm. A few years back in 1985, interviews by Tong and Canter [6] showed that over 45% of a small sample of building occupants were unable to distinguish fire alarms from other types of alarms. A more recent fire alarm study, by NRC, with over 300 participants demonstrated that only 14% identified the slow whoop and 38% the alarm bell as fire alarms [7].

The need to devise a unique, universally recognizable fire alarm signal was recognized many years ago. Since the 1970s, numerous discussions to develop a standard signal have taken place [8, 9]. Experts finally agreed not to limit the signal to any one sound (such as a bell, horn, chime or electronic sound) but instead to support the concept of a specific sound pattern. The Temporal-Three pattern, often referred to as the T-3, became a requirement for fire alarms in many jurisdictions including new US construction through the 1996 NFPA 72. Although the T-3 is described in ISO 8201 as the “evacuation signal” it is

questionable that this signal by itself could trigger evacuation of a building. It will probably take over 25 years to have the T-3 signal as the sound of most fire alarms in North American buildings. Meanwhile, occupants are getting exposed to this signal during different alarm activations so they become used to

associating the T-3 pattern to the fire alarm signal. While it is hoped that more and more countries will adopt this standard, by itself the pattern is unlikely to solve the problem of occupants ignoring the fire alarm. In the long run there

should be more and more recognition of the signal but not necessarily a specific response.

Unaware of the proper response

Discussions with evacuees of non-residential buildings show that in the initial moments of the alarm activation either occupants didn’t know what to do or were unaware that they were expected to do anything [10]. Training could counteract this inertia, however, it is not always possible to train occupants, particularly punctual visitors. It is difficult to devise how visitors of a shopping center could be trained but in some buildings there are some missed opportunities. For instance, in cinemas, just before the main feature, a short 2-minute segment could play the fire alarm signal and instruct the public as to what action is expected from them and the location of the exits. In buildings where training is difficult to accomplish, it becomes essential to have systems in place that will support the fire alarm such as trained staff and voice communication messages.

Loss of Confidence in System because of Nuisance Alarms

The large number of nuisance alarms, such as false alarms, test alarms and fire drills, is another reason why occupants do not take action when a real fire alarm is sounded. The problem with nuisance alarms is that, after a time, occupants tend to lose confidence in the system, including all similar systems. Many

occupants assume that whenever they hear a fire alarm, they can safely dismiss it as a nuisance alarm; denial that the alarm signal is an indication of a genuine problem is fairly common. During mid-rise residential evacuation studies, it was found that less than 25% of occupants interpreted the sound of the fire alarm as a potential indication of a real emergency [11].

The number of nuisance alarms and their deterring effects has to be studied over a period of time. In a given building, three nuisance alarms in one week will have a greater deterring effect than three nuisance alarms over the course of a year. The time of occurrence and the type of building might also play an important role. If a nuisance alarm occurs in the middle of the night in a high-rise residential building, it may have a more lasting negative effect on occupants than a nuisance alarm happening in an office building on a warm sunny day.

How many nuisance alarms in one year can be considered too many? Three? Five? Ten? How many will cause people to lose faith in the fire alarm system? No research data have been found to answer these questions. Specialists in the field tend to agree, however, that more than three nuisance alarms in one year can undermine the credibility of the system. One thing is certain: nuisance alarms tend to downplay the sense of danger or urgency that should be associated with a fire alarm signal. Confronted with many nuisance alarms, people are likely to ignore the signal or attempt to disconnect the system. The public often assumes that nuisance alarms are largely the work of

mischievous teenagers. Through the use of surveillance cameras it was found that pranksters could indeed be teenagers, but they were also younger children and adults, even senior citizens. Further, the assumption that nuisance alarms

are usually prank alarms is not true. In fact, most nuisance alarms are due to a system malfunction. In 1999, fire departments in the United States received over 2 million calls that turned out to be nuisance alarms. Of these, 44% were system malfunctions, 30% were well-intentioned calls that turned out not to be fires, 15% were mischievous false calls, and 11% were other types of false alarms, such as bomb scares [12].

It is essential to strive to reduce the number of nuisance alarms to a minimum. Communication is also important. Opportunities to reinforce occupants’

confidence in the alarm system are lost when managers do not inform occupants about the causes of nuisance alarms or the action they are taking to rectify the problem. Every time a fire alarm is activated for any reason a message should be issued to all occupants through the voice communication system or e-mail explaining the cause of the activation. Knowing that somebody is aware of these alarm activations and is doing something about it would help rebuild some

credibility.

Failure to Hear the Alarm

A fourth explanation for occupants’ lack of response to the fire alarm is the audibility of the signal itself. Studies in mid-rise and high-rise residential

buildings have shown that in some instances occupants could not hear the signal from inside their apartments [13, 14]. This audibility problem was typically

observed in apartment blocks where the alarm appliances were located in the common corridors. Even though the alarm signal was very loud in the corridors, the signal was not audible inside dwelling units, especially in rooms located furthest from the public corridor. Further, the ambient sounds of everyday life, emitted by televisions, audio units, air conditioning systems or human activities, can easily mask the sound of the alarm signal.

Recent studies have found that a number of occupants may not be awakened by the sound of a fire alarm meeting existing codes. Is has been found that young children, adults who are sleep deprived or under the influence of a relatively small amount of alcohol as well as elderly people, may not awake readily at the sound of a smoke alarm [15, 16, 17]. Research is underway to investigate what characteristics of the signal could be changed to ensure awakening. It should be kept in mind that awakening is only the first step, taking action after awakening will be a determinant on the outcome of a fire.

“Make it unbearably loud: they will leave”

A building manager suggested increasing the sound level of the fire alarm to the maximum permitted which, in his view, would drive occupants to leave. Well, although the sound level is too low in some buildings where for instance the alarm sounders are located in the public corridor and not in each suite,

increasing the sound of the alarm to its maximum is not a recommended solution as only 15 dBA above the average ambient sound is necessary to be perceived by people who are awake. A very loud alarm, that goes off for no emergency reason, would result in many complains and occupants tampering with the

system to avoid a future reoccurrence. More importantly though, it would prevent communication between occupants as they decide on what actions to take. Once occupants have perceived the fire alarm and possibly some additional cues they engage into the milling process, which is to look at what others do and discuss with others the situation and the best course of action. Milling is an essential step in the decision making process. Instead of trying to eliminate milling it should be facilitated with additional information, which will speed up confirmation of the situation and decision-making. In fact, in most facilities louder alarms are not needed; on the contrary, quieter alarms could be beneficial,

particularly in locations where all occupants are awake and active and where the background noise is low.

While the fire alarm signal should be low enough to allow verbal exchanges, it should not be so low that occupants might think that the fire alarm signal has been switched off. During a high-rise evacuation study, the alarm was turned off after five minutes to facilitate walkie-talkie communication between firefighters [11]. It was observed, through video recordings that many occupants who had started down the stairwell stopped and turned back when the alarm signal was switched off. Because the fire alarm signal was silenced, occupants assumed the emergency was over. This reaction explains why it is very important to maintain the alarm signal activated until the emergency is finished.

Conclusion

It should not be concluded that we might as well get rid of fire alarms since occupants tend not to respond to them as traditionally expected. In fact,

although response to fire alarms is not what most would expect, this signal is still an excellent means to provide a warning to the occupants. Such a signal gives people a cue that something is going on, to search or watch out for more.

Consequently, later on, during the incident, when people perceive additional cues such as smelling smoke or seeing staff running around, they will be more likely to conclude that there is indeed something serious happening and they should do something about it.

Studies have demonstrated that it is overly optimistic to expect that the fire alarm signal alone will warn all occupants, prompt immediate action, initiate evacuation movement and allow sufficient time for all to escape safely. In addition to using the T-3 evacuation signal, complementary cues including voice communication messages, staff instruction, training, drills and a well devised building emergency action plan will greatly increase the probability that the occupants of a building will respond appropriately and quickly to a fire alarm.

References

1- Bryan, L. J., 2002, “Behavioral Response to Fire and Smoke” Chapter 3-12,

SFPE Handbook of Fire Protection Engineering, 3rd Edition, pp. 3-315-3-341.

2- SFPE, 2003, “Human Behavior in Fire”, Engineering Guide, Society of Fire Protection Engineers, 46 p.

3- Bryan, L. J., 2001, “Psychological variables that may affect fire alarm design”,

Fire Protection Engineering, www.sfpe.org, pp. 42-48.

4- Canter, D., Breaux, J., Sime, J., 1990, “Domestic, multiple occupancy, and hospital fires”, Chapter 8, Fires and Human Behaviour, Edited by D. Canter, Second Edition, John Wiley & Sons Ltd, pp. 117-136.

5- Bukowski, R. W. Peacock R. D. Averill J. D. Cleary T. G. Bryner N. P. Walton W. D. Reneke P. A. & Kuligowski E. D. 2003, Performance of Home Smoke Alarms: Analysis of the Response of Several Available Technologies in Residential Fire Settings, National Institute of Standards and Technology, Washington, U.S.

6- Tong, D. and Canter, D., “The decision to evacuate”. Fire Safety Journal, Vol. 9,

No. 3, 1985, pp. 257-265.

7- Proulx, G.; Laroche, C.; Jaspers-Fayer, F.; Lavallée, R., “Fire Alarm Signal

Recognition”, Internal Report, Institute for Research in Construction, National

Research Council Canada, 828, pp. 32, 2001 (IRC-IR-828)

http://www.nrc.ca/irc/fulltext/ir828/

8- Mande, I., “A standard fire alarm signal temporal or ‘slow whoop’”, Fire Journal,

Vol. 69, No. 6, 1975, pp. 25-28.

9- CHABA., “A proposed standard fire alarm signal”, Fire Journal, Vol. 69, No. 4,

1975, pp. 24-27.

10- Proulx, G., 1999, “Occupant Response to Fire Alarm Signals”, National Fire Alarm Code Handbook, NFPA 72, Supplement 4, pp. 403-412.

11- Proulx, G., Latour, J.C. and MacLaurin, J.W. “Housing evacuation of mixed abilities occupants”, IRC-IR-661, Internal Report, Institute for Research in Construction, National Research Council of Canada, 1994.

12- Karter, J. M. “Fire loss in the United States during 199”,. National Fire Protection

Association, Internal report, 2000.

13- Proulx, G., Laroche, C. and Latour, J.C. “Audibility problems with fire alarms in apartment buildings”, Proceeding of the Human Factors and Ergonomics Society

39th Annual Meeting, Vol. 2, 1995, pp. 989-993.

14- Sultan, A. M. and Halliwell, R.E. Optimum location for fire alarms in apartment

buildings. Fire Technology, Vol. 26, No. 4, 1990, pp. 342-356.

15- Bruck, D. Reid, S., Kouzma, J., & Ball M., 2004, The Effectiveness of Different Alarms in Waking Sleeping Children, Third International Symposium on Human Behaviour in Fire, Belfast, September 2004, pp. 279-289.

16- Ball, M., & Bruck, D., 2004, The Effect of Alcohol upon Reponse to Fire Alarm Signals in Sleeping Young Adults, Third International Symposium on Human Behaviour in Fire, Belfast, September 2004, pp. 291-302.

17- Bruck, D., Thomas, I., Kritikos, A, 2006, Investigation of Auditory Arousal with Different Alarm Signals in Sleeping Older Adults, Report for the Fire Protection Research Foundation.