Human behavior and evacuation movement in smoke

(1)

Publisher’s version / Version de l'éditeur:

ASHRAE Transactions, 114, 2, pp. 159-165, 2008-02-01

READ THESE TERMS AND CONDITIONS CAREFULLY BEFORE USING THIS WEBSITE. https://nrc-publications.canada.ca/eng/copyright

Vous avez des questions? Nous pouvons vous aider. Pour communiquer directement avec un auteur, consultez la

première page de la revue dans laquelle son article a été publié afin de trouver ses coordonnées. Si vous n’arrivez pas à les repérer, communiquez avec nous à PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca.

Questions? Contact the NRC Publications Archive team at

PublicationsArchive-ArchivesPublications@nrc-cnrc.gc.ca. If you wish to email the authors directly, please see the first page of the publication for their contact information.

NRC Publications Archive

Archives des publications du CNRC

This publication could be one of several versions: author’s original, accepted manuscript or the publisher’s version. / La version de cette publication peut être l’une des suivantes : la version prépublication de l’auteur, la version acceptée du manuscrit ou la version de l’éditeur.

Access and use of this website and the material on it are subject to the Terms and Conditions set forth at

Human behavior and evacuation movement in smoke

Proulx, G.; Fahy, F. R.

https://publications-cnrc.canada.ca/fra/droits

L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.

NRC Publications Record / Notice d'Archives des publications de CNRC:

https://nrc-publications.canada.ca/eng/view/object/?id=6c183da2-8d0e-45c7-af38-6c9166568e7d https://publications-cnrc.canada.ca/fra/voir/objet/?id=6c183da2-8d0e-45c7-af38-6c9166568e7d

(2)

H um a n be ha vior a nd e va c ua t ion m ove m e nt in sm ok e

N R C C - 5 1 1 5 7

P r o u l x , G . ; F a h y , F . R .

F e b r u a r y 1 , 2 0 0 8

A version of this document is published in / Une version de ce document se trouve dans:

ASHRAE Transactions, v. 14, no. 2, pp. 159-165,

The material in this document is covered by the provisions of the Copyright Act, by Canadian laws, policies, regulations and international agreements. Such provisions serve to identify the information source and, in specific instances, to prohibit reproduction of materials without written permission. For more information visit http://laws.justice.gc.ca/en/showtdm/cs/C-42

Les renseignements dans ce document sont protégés par la Loi sur le droit d'auteur, par les lois, les politiques et les règlements du Canada et des accords internationaux. Ces dispositions permettent d'identifier la source de l'information et, dans certains cas, d'interdire la copie de documents sans permission écrite. Pour obtenir de plus amples renseignements : http://lois.justice.gc.ca/fr/showtdm/cs/C-42

(3)

ABSTRACT

Codes and standards are developed on the premise that occupants of a building that meets relevant requirements should not have to evacuate through smoke during a fire. Inves-tigations of actual fires demonstrate, however, that occupants often move through smoke during a fire evacuation. This paper will review common behaviors observed during fire emergen-cies, particularly those that have been observed in the pre-movement phase of the evacuation. Three case studies are presented to illustrate occupant movement through smoke in fires in office and apartment buildings. Two new technologies that can support occupants’ evacuation through smoke are described: directional alarm sounders and photoluminescent wayguidance systems.

INTRODUCTION

In order to protect people in buildings during emergen-cies, it is essential to understand the factors that may influ-ence their responses and behaviors. Research has shown that occupant response and behavior is influenced by factors such as their familiarity with the building, their previous experi-ences in emergencies, their commitment to any task or situ-ation in which they are engaged, whether they are alone or in a group, their ties to others in the group, their role, and train-ing, as well as individual characteristics such as age and gender (SFPE 2003).

Codes and standards tend to assume that adherence to code requirements will prevent occupants from having to evacuate through smoke during a fire. Investigations of actual fires demonstrate, however, that occupants often move through smoke during a fire evacuation. In fact, the cases studied when occupants did not move through smoke are relatively rare.

Stairwell pressurization systems to achieve smoke control can be designed with the assumption that only one door will be open to the interior of the building; however, if the building’s evacuation plan is such that multiple floors of occu-pants will be entering the stairwell, the smoke control system will fail and occupants are very likely to evacuate in a smoke-filled stairwell. This situation is particularly likely for occu-pants who do not evacuate in the early stages of the fire inci-dent or for occupants located above the fire floor. If the evacuation plan calls for the simultaneous evacuation of occu-pants from several floors, the smoke control system design must take that into consideration and must factor in whether the discharge door opens to the interior or exterior since the effect of opening a door to the outside is usually much greater than that of opening interior doors. In any case, the system is required to maintain a tenable atmosphere for occupants for as long as necessary during evacuation, although that atmosphere does not have to be free from smoke (NFPA 2006). Tenability is not specifically addressed by most standards.

Some jurisdictions have begun to consider that movement through smoke is a possibility during fire emergencies. For example, since 1989, California has required the installation of low-level exit signs next to exit stairwell doors in newly constructed apartment buildings, condominiums, and hotels (currently shown in the 1997 edition of the California Building Code, Section 1007.6.2.1). Presumably, if the high-mounted exit signs were obscured by smoke, the occupant could locate the exit door with the low-mounted exit sign. New York City now requires photoluminescent exit signs to be placed 18 in. (0.46 m) above the floor on the exit door or the wall next to it on every floor in new and existing commercial high-rise build-ings (NYC 2005).

Human Behavior and

Evacuation Movement in Smoke

Guylène Proulx, PhD

Rita F. Fahy, PhD

Guylène Proulx is a senior researcher in the National Research Council Canada’s Institute for Research in Construction, Ottawa, ON, Canada. Rita F. Fahy is a manager of fire databases and systems in the Fire Analysis and Research Division of the National Fire Protection Association,

Quincy, MA.

SL-08-016

© 2008, American Society of Heating, Refrigerating and Air-Conditioning Engineers, Inc. (www.ashrae.org). Published in ASHRAE Transactions Vol. 114, Part 2. For personal use only. Additional reproduction, distribution, or transmission

(4)

160 ASHRAE Transactions

HUMAN BEHAVIOR AND PRE-MOVEMENT TIME

The behavior of building occupants faced with a fire emergency has been studied through evacuation drills and evacuee interviews following actual fires. The responses of occupants are divided into sequences, which relate to the chro-nology of the unfolding event. In the initial moments, an absence of response due to ignorance or disbelief is often observed. This period is characterized by ambiguous cues being perceived in the environment. This could be the smell of burning objects, the sight of light smoke, flickering of lights, or a tremor in the building that essentially may not initiate any response in the first few moments. If occupants are committed to a specific activity, they might very well complete this task before turning their attention to an unusual ongoing event. Forget about panic—it happens too rarely even in the most severe incidents to really consider it a likely response (Sime 1980; Keating 1982). In fact, the initial moments of an emer-gency are often characterized by inertia or a perplexed slow response of the building occupants.

As these initial cues persist and accumulate with addi-tional cues, occupants are likely to start investigating. This investigation response can take several forms: some will try to find the source of the problem by moving around the floor, visiting rooms, and opening doors that could very well bring oxygen to the fire and allow the smoke to spread to other rooms. Other occupants might call the superintendent or a person responsible for the building to try to obtain informa-tion. This attempt at obtaining more information is necessary in order for the occupants to appraise the situation and make a decision on the best course of action. If the information about what is going on is not readily available, occupants will look for it. As part of this sequence, occupants will engage in the milling process—they will start talking and milling around with others in an effort to get a better understanding of the situation at hand. This period of time from the initial perception of an unusual cue until the person makes the deci-sion to start an evacuation is called pre-movement time. This pre-movement time may involve considerable movement that is not yet related to evacuation movement. Depending on the information and cues available, this pre-movement time can last from a few seconds to several minutes (Proulx 2002a).

Part of the pre-movement time also involves actions such as fighting the fire. Although it is fairly rare that occupants attempt to fight a fire in an office building, it is possible that some individuals will perceive their role as requiring that they try to fight the fire. Attempting to fight a fire is much more common in residential occupancies (Bryan 2002). Some occu-pants will be inclined to warn others of the emergency. The affiliation concept may come into play, as occupants tend to gather with family members or close friends to evacuate together. It has also been observed, in major emergencies, that occupants will contact family members by phone or e-mail to let them know that they are “all right” and are starting their evacuation (Proulx and Fahy 2004). These types of actions as well as retrieving belongings, personal and work-related, are

all part of getting prepared to leave and are part of the pre-movement time.

Pre-movement time has been assessed, for instance, for high-rise office occupants during evacuation drills and through interviews following fires (Fahy and Proulx 2001). During drills, occupants tended to invest 35 seconds to one minute in pre-movement time (Proulx and Fahy 1997; Proulx et al. 2000). In real emergency evacuations, however, a wide variety of times have been reported, from five minutes to over 25 minutes (Fahy and Proulx 1997; Averill et al. 2005; Proulx and Reid 2006). Although common knowledge would suggest that strong cues such as the presence of smoke should trigger a quicker response compared to evacuation drills, results of fire investigations do not confirm this assumption. The presence of smoke may strongly suggest to occupants that there is a real emergency, which in turn is interpreted as the necessity to engage in actions such as warn others and gather valuables before leaving. Consequently, the presence of smoke is not a guarantee of a short pre-move-ment time and quick evacuation response.

Certainly, pre-movement time is a critical time to consider, as the longer the pre-movement time the more likely occupants might be faced with smoke during their evacuation. Means to reduce the pre-movement time need to be devised. For example, training prior to the event and information deliv-ered to occupants during the event through a voice communi-cation system can contribute to shortened pre-movement time.

MISCONCEPTIONS ABOUT OCCUPANT BEHAVIOR IN FIRE EMERGENCIES

There are a couple of key misconceptions concerning occupant behavior in fire emergencies. The first is that occu-pants will panic and stampede when they become aware of a fire emergency. In fact, it often requires multiple cues, more than simply an alarm signal, to get people to begin to evacuate the building, and panic has rarely been observed in actual fires. In fact, during fires occupants are usually relatively cool and composed; the trauma comes after the event. During the emer-gency, they are rational and will try to make the best possible decision according to their understanding of the situation. Occupants will always try to save themselves and loved ones. Since people are not fire experts, however, they might engage in behavior that increases the risk to themselves and others. For example, during the evacuation of the World Trade Center towers after the bombing in 1993, some windows were broken by occupants in order to clear the smoke in the spaces where they sought refuge and to get fresh air. Yet an evacuee reported that conditions became worse in his area after a window was broken (NFPA 1993). Falling glass also endan-gered firefighters working below and evacuees escaping the building. Breaking the windows was not an irrational response to the smoke condition, but the outcome was not positive— many fire experts could have foreseen this, but members of the general public often lack such expertise and experience.

(5)

ASHRAE Transactions 161

Another misconception is that people are reluctant to move through smoke. Studies of major incidents have shown that the opposite is true: occupants have been found to be will-ing to move through smoke, even for extended periods of time, when they believe they are heading toward safety. It is impor-tant to realize that the population in general knows that smoke kills (NFPA 1999). What they do not realize, however, is that a fairly small quantity can kill and that in some cases a couple of breaths could be fatal. People who have escaped from a fire and moved through smoke did not do it fearlessly—they assumed that by holding their breath, breathing through a cloth and moving as fast as they could, they would reach safety. As survivors, they can testify that this worked.

EVACUATION MOVEMENT

Once occupants have made the decision that an evacua-tion is necessary, they have to select an egress route. Famil-iarity with the building will be a factor, as people tend to move in familiar directions rather than experiment with new routes during an emergency. Training and the presence of floor fire wardens may help in the selection of the closest route to egress. What others are doing and their direction of movement will also influence where individuals head. Presence of smoke may play a role in the decision. If occupants find a stairwell filled with smoke and they are aware of another stairwell, they are very likely to go to the other one. If the smoke doesn’t look too threatening or if they have been told by a credible source that they absolutely need to evacuate, they might brave the elements, holding their breath and going down the stairwell.

Behavior inside stairwells during a high-rise office build-ing evacuation is unlike what many may imagine; there will be no running or pandemonium. The movement is likely to be very slow due to the high density of occupants if the building is fully occupied. In a high-rise residential building, on the contrary, density is likely to be light and occupants will be free to move at the speed of the slowest member of their group. Demonstrations of altruistic behaviors are prevalent during fire evacuations, with people helping the injured and the disabled and encouraging highly anxious others with small talk and soothing phrases such as “we will make it,” “we are almost there,” etc.

In office buildings, the speed of movement will be rela-tively slow in the stairs as the density is likely to be high with people trying to enter the stairwell from different floors. Usually evacuees already in the stairwell will feel themselves to be the priority and will allow only one or a few persons at a time to enter into the crowded stairwell, which may keep doors open on several levels. Recent evacuation studies have demonstrated that in a 44 in. (1.2 m) stairwell, occupants will not squeeze more than one person every two steps in a scat-tered pattern (this is about 2.3 persons per m2 [0.21 persons per ft2]) (Proulx et al. 2007). At such density, the speed of move-ment is considerably reduced. If a slow person, such as some-one with a walking impairment (e.g., obesity or a knee injury), enters the stairwell, he or she is likely to have a major impact

on the movement speeds of the evacuees following behind. Counterflow can also create a slowdown of movement. Absence of lighting or signage, and water or debris will also have an impact on the evacuees’ speed of movement. The pres-ence of smoke may prompt evacuees to move as fast as they can, but the discomfort due to reduced visibility, their eyes and noses running, coughing, and difficulty in breathing could reduce the speed of movement.

OCCUPANT MOVEMENT THROUGH SMOKE

In actual fires for which a thorough human behavior study was conducted, there has always been a majority of the evac-uees who had to move through smoke. Fires that are studied are usually the most dramatic cases where large loss of life and property have taken place. In these fires, evacuees who did not have to move through smoke are limited to three types of situ-ations: in a high-rise building when the occupants were located below the fire floor, when the occupants were remote from the fire site in a large horizontal structure such as an airport or a shopping mall, or when the occupants started evac-uation movement very early in the event. In all other types of situations in these case studies, occupants moved through smoke to evacuate their building.

CASE STUDIES

In the past few decades, there has been an increasing number of studies on human behavior following major fires. These studies are part of the overall fire investigation and are aimed at better understanding how occupants became aware of the incident, what they did, and why they did it in order to better understand the fire development and the loss of life. In all these fires, it is never one system that went wrong but rather the multiplication of mishaps and failures that led to the tragic outcome. In all these cases, however, occupants were exposed to smoke: some survived to testify to it but some perished.

Office Building Fire

One such informative case study is the fire of the 36-story Cook County Administration Building on October 17, 2003, which resulted in six fatalities and a dozen people injured (Proulx and Reid 2006). It is extremely rare to have fire fatal-ities in a high-rise office building; this is why this fire was so important to study. At the time of the fire (on a Friday around 5:00 p.m.), only approximately 250 occupants (out of over 2,000 employees) were in the building; these were on their way out, were preparing to leave for the weekend, or were still working. During this fire, over 40% of the occupants who mentioned noticing smoke on their floor used an elevator to egress despite the “elevators should be avoided during a fire” warnings posted on elevator call buttons, in the public address messages issued, and in the fire safety training they received prior to the event. Occupants who did not use an elevator to evacuate used one of the two stairwells, which were both contaminated with smoke. It appears that two thirds of the evacuees saw smoke in the stairwell when they opened the

(6)

stairwell door, but they entered anyway and started their jour-ney down. Throughout the evacuation, public address messages were issued instructing occupants to evacuate using the stairwells. Firefighters attacking the fire from one of the stairwell enclosures contributed to the stairwell filling up with smoke: this is the stairwell where all the fatalities were found. Evacuees opening doors on several floors contributed to letting smoke travel to all floors above the fire floor. This fire event shows that occupants are prepared to move through smoke and some may disobey instructions by using elevators if they seem to work, as this is their familiar way out.

Residential Building Fires

Movement through smoke is particularly common during residential fires. In a single-family home, duplex, or triplex, familiarity with the space makes it relatively easy for an occu-pant to hold his breath and move as quickly as possible to the outside. More often than not, these movements through smoke are successful (the occupants make it out alive). The message learned is that moving through smoke is the right thing to do, since the alternative, staying in place, might be fatal. In high-rise residential buildings, the travel distance to the outside increases the risk; however, movement through smoke happens nonetheless.

High-Rise Senior Housing. In January 1997, a fire took

place at Ambleside, a 25-story condominium high-rise for seniors located in Ottawa (Proulx et al. 1998; Proulx 1999). The fire started around 4:30 p.m. on a Friday afternoon from an improperly cleaned pipe left in a jacket that was stored in a closet of an apartment on the sixth floor. The two occupants of the apartment of fire origin, aged 80 and 82 years old, could not escape since the fire was at the entrance. Although the fire alarm signal sounded and all occupants heard it, most of them decided to protect-in-place, which was the procedure in this apartment building. Upon their arrival, the fire department asked for an announcement to be made on the building voice communication system instructing residents to immediately evacuate the building. Following this message, most occu-pants attempted to evacuate. All the occuoccu-pants located above the fire floor mentioned encountering smoke conditions during their evacuation. It appears that the three stairwells became clogged with smoke very rapidly. According to one couple, the smoke was so dense they could not see each other when descending two steps apart. Just over half of the occu-pants who attempted to evacuate managed to escape. Occu-pants who could not keep going in the smoke returned to their apartment or took refuge in someone else’s apartment. Follow-ing the evacuation, two seniors had heart attacks; one of them died ten days later. Although initially occupants decided to protect-in-place, a majority attempted to comply with the fire chief’s evacuation order delivered through the voice commu-nication system. Occupants above the fire floor who stayed home during the whole event did not get smoke inside their apartments, but smoke migrated inside the apartments of occu-pants who opened their main entrance doors. When they

followed instructions from a credible source, occupants simul-taneously opened doors on several floors, which allowed the smoke to travel throughout the building.

High-Rise Apartment Building. The Forest Laneway

fire in Toronto, which killed six occupants in 1995, is another informative case study (Proulx et al. 1995). In this building, the fire started from a dropped cigarette on a couch of an apart-ment on the fifth floor of the 29-story building. The fire was discovered around 5:00 a.m., when most occupants were in bed in their apartments. The occupant escaping the apartment of fire origin left open the main door of the apartment, allow-ing smoke to contaminate the corridor. Smoke entered the stairwell located across the hall because the door to the stair-well was unlatched. Occupants became aware of the fire when they were warned by neighbors or heard the fire alarm, which stopped after one or two minutes. Some discovered smoke in their corridor or in the stairwell when they went to investigate. About two thirds of the occupants above the fire floor decided to stay in their apartments because they thought there was too much smoke to evacuate. Others attempted to evacuate despite the smoke, but only one third of them were successful—the others returned home or took refuge with neighbors. Among the people who decided or were forced to protect-in-place, 76% indicated that smoke entered their apartments from their main door or the ventilation system. None of the occupants who stayed in place during this fire were seriously injured, although 90% of the respondents in apartments located above the 15th floor reported considerable smoke. The six people who died were overcome by smoke at the top of both of the two stairwells. Investigators believe that they were attempting to seek refuge on the roof of the building, but as in most high-rise buildings, the door to the roof was locked. This case study shows again that occupants who believe that it is imperative to evacuate the building will try to move through a substantial amount of smoke to reach a safe area.

Fires in Public Assembly Occupancies

Through the years, a number of nightclub fires have shown that movement through smoke is also a recurrent response (Duval 2006). Tragic fires such as the Cocoanut Grove, Rhythm Club, Beverly Hills Supper Club, and The Station nightclub fires all share this commonality—most evac-uees had to move through smoke, and some survived but several perished. Some other factors were instrumental in the outcome of these fires, such as overcrowding and the limited number of exits, the delay in warning the public, and the substantial amount of combustible material present. But over-all, the smoke built up at an incredible speed, which surprised all present, who simply did not have enough time to reach an exit before being overcome by smoke.

ALTERNATIVES TO FULL-BUILDING EVACUATION

Sometimes, evacuation is not the optimal response in a fire emergency. This has been shown to be true in several recent high-rise building fires where occupants were killed

(7)

while attempting to escape (Proulx 2001). Today’s high-rise buildings are not designed for full-building evacuations. In fire safety planning, phased evacuations are the norm—the fire-involved floor and those immediately above and below are evacuated while occupants in the rest of the building remain in place.

Non-evacuation in fire-resistive buildings has been advo-cated for over two decades (MacDonald 1985). More recently, Proulx (2001, 2002b) has outlined the conditions necessary for an effective protect-in-place strategy in high-rise build-ings: non-combustible construction, self-closers on all entry doors, a central alarm system, and a voice communication system with good intelligibility in every compartment. Although not specified at the time the report was published, sprinklers would also be advantageous in providing an addi-tional level of safety to the occupants.

The advantages of a phased evacuation or a protect-in-place strategy are clear. Building occupants are protected from the developing fire and do not have to move in corridors and stair-wells that might be contaminated by smoke. Further, they do not impede firefighters who may be conducting firefighting opera-tions from the stairwells and as a result may be clogging stairwells and corridors with their equipment. In the aftermath of the collapse of the towers at the World Trade Center in 2001, however, there are real questions as to whether occupants would be willing to remain in place over an extended period of time, especially if they see fire and smoke in their building. It is essen-tial to learn more about the general public’s perception of their level of safety in high-rise buildings. A Fire Protection Research Foundation project, which was completed in the summer of 2007, may begin to answer some of those questions (Zmud 2007).

INNOVATIVE TECHNOLOGIES

TO ENHANCE OCCUPANT EVACUATIONS

Although evacuation in a smoke-free environment should remain a design objective, the potential for evacuation under some smoke conditions must also be considered. The preva-lence of occupants’ movement in smoke during actual fires demonstrates that such a scenario should be anticipated.

There are two innovative technologies that could support occupant movement in smoke: directional alarms and photo-luminescent wayguidance systems. Both technologies have the advantage of also supporting potentially faster and more efficient evacuation under normal conditions, power failure, or smoke conditions.

The directional alarm is a new technology that has sound-ers emitting short bursts of broadband sound over a large range of frequencies, which uses people’s ability to localize sources of sound to attract them toward the exit (O’Connor 2005). This new approach to a fire alarm is different from the traditional design of placing several sounders throughout the space to achieve a certain audibility level that allows the sound to be heard by the building occupants. The directional sounders are instead placed along the evacuation route and in particular above the exit doors to assist occupants in locating the closest

means of egress. Experimental studies have shown that the psychoacoustic response to directional alarms allows occu-pants to quickly locate exits in total darkness or theatrical smoke (Boer and Withington 2004; Catchpole et al. 2004; Winthington 1999).

Photoluminescent wayguidance systems are made of inor-ganic chemical compounds, referred to as photoluminescent

pigments, encased in a flexible or rigid material such as vinyl or

polyvinylchloride (PVC) or dispersed in a liquid such as paint. The pigments are characterized by their ability to store energy when exposed to a light source. When all ambient lights are cut off, the material will emit light over a period of time. The lumi-nescence of the photoluminescent material is very strong initially and diminishes with time. It can then be recharged by re-exposing the material to light for a short period of time. Photoluminescent material has many applications. In fire safety, the most promising uses are for safety markings such as exit signs, directional signage, door markings, path markings, obstruction identification, and other components that compose a safety wayguidance system. Under normally lit conditions, it can be used as a wayguidance sign system. In blackout situa-tions resulting from power failures or fires, photoluminescent material will glow and therefore will aid evacuation by guiding and directing people along evacuation routes.

Following the bombing at the World Trade Center in 1993, where several thousands of office workers had to evac-uate in total darkness, a photoluminescent wayguidance system was installed in all the stairwells of the complex. This installation proved invaluable during the September 11, 2001, attacks. Although the lighting remained operative in most areas of the towers, 33% of Tower 1 and 17% of Tower 2 survi-vors stated being helped by the photoluminescent marking in their movement to safety (Averill et al. 2005). Renovations to the Pentagon following the September 11, 2001, attack included the addition of a photoluminescent wayguidance system in corridors and stairs, while in the United Nations’ headquarters building, such a system was installed in 2003.

As a result of the events of September 11, 2001, and the blackout of August 2003, New York City recognized that a photoluminescent wayguidance system is an essential compo-nent to ensure occupant safety in high-rise buildings due to the potential full evacuation of high-rise buildings under emer-gency power failure or the presence of smoke. Consequently, on May 31, 2005, New York City passed Local Law 26, Photo-luminescent low-level exit path markings (NYC 2005), which required all existing or new high-rise office buildings in New York City to install a photoluminescent wayguidance system by July 2006. Other building codes are also beginning to include the use of photoluminescent safety markings as a means of better defining escape routes and to support occupant evacuation under blackout or smoke conditions.

CONCLUSION

Research has shown that occupant responses to fire emer-gencies follow some fairly consistent patterns. Designers need

(8)

to plan spaces in ways that complement expected occupant behavior. With the knowledge we have about expected behav-iors, we can facilitate the occupants’ evacuation, relocation, or sheltering in place. This can be done either by accommodating their actions (e.g., providing additional exiting capacity at the main entrances), by adding building features that would support protect-in-place (such as self-closers on doors, fire doors in corridors and elevator lobbies, two-way communica-tion systems, control over the sound of the alarm, balconies, windows that can be opened), or by directing their actions through the use of voice messages, staff interaction, direc-tional alarm sounders, and photoluminescent marking.

Overall, people in a fire are rational and make decisions based on the information available at the time of the event. They will attempt to make the best decision to preserve them-selves and loved ones. Evacuees are often helpful, courageous, and cooperative. They will move through smoke if they believe it is the right thing to do and if they feel the route leads to safety. They will follow instructions from a person in authority, if the instructions match their own judgment of the situation. Sometimes they’ll make it, but sometimes they’ll perish. High-rise building occupants are not emergency experts; they lack experience with such events. This is why, for them, education and training are so important.

Smoke control systems can provide essential protection to evacuating occupants; however, their effectiveness can be defeated by some of the behaviors observed in actual fire evac-uations, particularly when occupants move simultaneously off too many floors into the stairwell. Occupants have also propped stairwell doors open during their evacuation to ensure re-entry onto floors, if necessary, or to provide light into dark-ened stairwells. For a smoke control system to work properly, it should be designed to maintain the delicate balance between the mechanical system’s operation and the reality of human behavior in fires.

REFERENCES

Averill, J.D., D.S. Mileti, R.D. Peacock, E.D. Kuligowski, N. Groner, G. Proulx, P.A. Reneke, and H.E. Nelson. 2005. Federal Building and Fire Safety Investigation of the World Trade Center Disaster: Occupant Behavior, Egress, and Emergency Communications. NIST NCSTAR 1-7, NRCC-48362. National Institute of Stan-dards and Technology, Gaithersburg, MD. http:// wtc.nist.gov/oct05NCSTAR1-7index.htm.

Boer, L.C., and D.J. Withington. 2004. Auditory guidance in a smoke-filled tunnel. Ergonomics 47:1131–40.

Bryan, J.L. 2002. SFPE Handbook, 3d Ed., Chapter 3-12, Behavioral Response to Fire and Smoke, pp. 3-315–3-341. Quincy, MA: National Fire Protection Association. Catchpole, K., J.D. McKeown, and D.J. Withington. 2004.

Localizable auditory warning pulses. Ergonomics 47:748–71.

Duval, R.F. 2006. NFPA Case Study: Nightclub Fires. Quincy, MA: National Fire Protection Association.

Fahy, R.F., and G. Proulx. 1997. Human Behavior in the World Trade Center Evacuation. Proceeding of the

Fifth International Symposium on Fire Safety Science,

pp. 713–24.

Fahy, R.F., and G. Proulx. 2001. Toward creating a database on delay times to start evacuation and walking speeds for use in evacuation modeling. Proceedings of the

Sec-ond International Symposium on Human Behaviour in Fire, pp. 175–83 (NRCC-44758). http://irc.nrc-cnrc.

gc.ca/pubs/fulltext/nrcc44758.

Keating, J.P. 1982. The myth of panic. Fire Journal 76(3):57–61.

MacDonald, J. 1985. Non-evacuation in compartmental fire-resistive buildings can save lives and makes sense.

Pro-ceedings of the International Conference on Building Use and Safety Technology, Los Angeles, pp. 169–74.

NFPA. 1993. Survey of evacuees of the World Trade Center following the bombing in February, 1993. Unpublished. National Fire Protection Association, Quincy, MA. NFPA. 1999. 1999 NFPA National Fire Escape Survey.

Quincy, MA: National Fire Protection Association. NFPA. 2006. NFPA 92A, Smoke-Control Systems Utilizing

Barriers and Pressure Differences. Quincy, MA:

National Fire Protection Association.

NYC. 2005. Local Law 26, Reference Standard 6-1, Photo-luminescent exit path marking. New York City Building

Code, 27-383(b). www.nyc.gov/html/dob/downloads/

pdf/rs_6-1.pdf.

O’Connor, D. 2005. Directional sound. NFPA Journal 99(3):50–56.

Proulx, G. 1999. Occupant response during a residential highrise fire. Fire and Materials 23(6):317–23. (NRCC-43149) http://irc.nrc-cnrc.gc.ca/pubs/fulltext/nrcc43149. Proulx, G. 2001. High-rise evacuation: A questionable con-cept. Proceedings of the Second International

Sympo-sium on Human Behaviour in Fire, pp. 221–30.

(NRCC-44675) http://irc.nrc-cnrc.gc.ca/pubs/fulltext/nrcc44675. Proulx, G. 2002a. SFPE Handbook, 3d Ed., Chapter 3-13, Movement of People: The Evacuation Timing, pp. 3-342– 3-366.

Proulx, G. 2002b. The pros and cons of protect-in-place. Fire Australia. (NRCC-45668) http://irc.nrc-cnrc.gc.ca/pubs/ fulltext/nrcc45668/.

Proulx, G., N. Bénichou, J.K. Hum, and K.N. Restivo. 2007. Evaluation of the effectiveness of different photolumi-nescent stairwell wayguidance installations for the evac-uation of office building occupants. Research Report 232, Institute for Research in Construction, National Research Council Canada, pp. 77, July 6, 2007 (IRC-RR-232). http://irc.nrc-cnrc.gc.ca/pubs/rr/rr232/. Proulx, G., and R.F. Fahy. 1997. The time delay to start

evac-uation: Review of five case studies. Proceedings of the

Fifth International Symposium on Fire Safety Science,

(9)

Proulx, G., and R.F. Fahy. 2004. Account analysis of WTC survivors. Proceedings of the 3rd International

Sympo-sium on Human Behaviour in Fire, pp. 203–14.

(NRCC-47309) http://irc.nrc-cnrc.gc.ca/pubs/fulltext/nrcc47309. Proulx, G., B. Kyle, and J. Creak. 2000. Effectiveness of a photoluminescent wayguidance system. Fire Technol-ogy 36(4):236–48. www.nrc.ca/irc/fulltext/nrcc44216/. Proulx, G., M.J. Ouellette, P. Leroux, and K.R. Bailey. 1998.

Study of the Occupants’ Behavior during the Ambleside Fire in Ottawa on January 31, 1997. Internal Report (IRC-IR-771), Institute for Research in Construction, National Research Council Canada, Ottawa. http:// irc.nrc-cnrc.gc.ca/pubs/ir/ir771/.

Proulx, G., J. Pineau, J.C. Latour, and L. Stewart. 1995. Study of the Occupants’ Behaviour During the 2 Forest Laneway Fire in North York, Ontario, January 6, 1995. Internal Report (IRC-IR-705), Institute for Research in Construction, National Research Council Canada, Ottawa. http://irc.nrc-cnrc.gc.ca/pubs/ir/ir705/.

Proulx, G., and I. Reid. 2006. Occupant behavior and evacua-tion during the Chicago Cook County Administraevacua-tion Building fire. Journal of Fire Protection Engineering 16(4):283–309. http://irc.nrc-cnrc.gc.ca/pubs/fulltext/nrcc 48146/ or http://dx.doi.org/10.1177/1042391506065951. Sime, J.D. 1980. The concept of panic. Fires and Human

Behavior. D. Canter, ed. Chichester, UK: John Wiley

and Sons.

SFPE. 2003. Engineering Guide: Human Behavior in Fire. Bethesda, MD: Society of Fire Protection Engineers. Withington, D.J. 1999. Localisable alarms. Human Factors

in Auditory Warnings. N.A. Stanton and J. Edworthy,

eds. Aldershot, UK: Ashgate Publishing Ltd.

Zamud, M. 2007. Public Perceptions of High-Rise Building

Safety and Emergency Evacuation Procedures Research Project. Quincy, MA: The Fire Protection Research

Foundation.