Publisher’s version / Version de l'éditeur:
Fire Prevention, 291, pp. 24-27, 1996-08-01
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
Critical factors in high-rise evacuations
Proulx, G.
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=ef957388-34e1-4394-8950-41a53ca3858d
https://publications-cnrc.canada.ca/fra/voir/objet/?id=ef957388-34e1-4394-8950-41a53ca3858d
http://irc.nrc-cnrc.gc.ca
Crit ic a l fa c t ors in high-rise e va c ua t ions
N R C C - 4 0 3 3 3
P r o u l x , G .
A u g u s t 1 9 9 6
A version of this document is published in / Une version de ce document se trouve dans:
Fire Prevention, (291), pp. 24-27
'i'
24
-
Fire Prevention 291 July/August 1996
1
The tenant exited the apartment, leaving both the patio door ro the balcony and the
apartment entry doortothe corridor open.
This allowed smoke ro flow from the apartment into the corridors and stairwells. The local Fire Department arrived at 0514 hours. They discovered the apartment engulfed in fire. The six fatalities were found in the rwo staircases, between the 27th and 30th floors. Many other occupants suffered smoke inhalation and had to be treated in hospital.
The sole occupant of Unit 509 reported he had been working on his computer for most of the night in one of the rwo bedrooms, when he saw smoke coming from the living-room couch. He rapidly tried to extinguish the fire with water and opened the patio
doortothe balconytovent the smoke out
of the room. However, the situation got out of control when the couch suddenly burst into flames.
The building
The building consists of13 apartments on
each floor, organised around a 'D' shaped
corridor. The corridor surrounds the four elevators and the two staircases on three sides. The two staircases are accessible on every floor
and, even though they are located nexttoeach
other, they are enclosed in entirely separate shafts (see Figure 1).
The design of the two staircases is complex as they are organised in a scissor-stairs configuration (see Figure 2). The staircase design appears ro have confused a number ofoccupants as well as firefighters and rescue officers during the fire.
Confusion can easily arise because, ,when travelling one flight ofstairs, a person moves
from one side of the buildingto the other.
For example, a person entering the fifth floor on the east side will exit on the fourth floor on the west side. Furthermore, there is no 13th floor in the building and therefore stairs will not always lead ro the same side on even-numbered floors or always to the other side on odd-numbered floors. The two staircases will be referred ro as the Fire Staircase and the Other Staircase (see
Figure 1). The doortothe Fire Staircase on
the fifth floor is located diagonally across the corridor from Unit 509, where the fire started. Due ro this proximity, the smoke was much denser in the Fire Staircase than in the Other Staircase.
Because of the complexity of the staircases, many occupants had difficulty determining the floor level they had reached during their evacuation. It also complicated the work of the firefighters and police officers who were 03 05 07 01 11 East
.J
Other Stairs
Council of Canada (NRCC) ro conduct a joint study of the occupants' behaviour during the fire. It was hoped that the findings of the study would be used ro develop recommendations to improve fire safety in high-rise apartment buildings. The results are also being used ro verify the NRCC's computer model FiRECAM [2], which is used ro assess the risk ofoccupants from fire.
The fire started at around 0500 hours in Unit 509, on the north-east corner of the fifth floor of the 30-storey building, which contained approximately 550 occupants.
02 14 04 08
06
12 10..
. .
Westｾ
Fire Stairs
Figure 1. Typical floor plan of the building.
The fire began in apartment 509.
Main Entrance
ｾ
ｎ 6 JANUARY 1995, acigarerre left. unattended. ignited a living-room
couch and caused a fite that killed six people in a high-tise apartment building in North York, Ontario.
The fire had an unusually high death roll: fire statistics for the province of Ontario show that a rotal of42 people (or an average of eight per year) died in high-rise residential buildings in Ontario berween 1989 and 1993. During this period none of the fires recorded resulted in mote than rwo fatalities. This led the Office of the Fire Marshall and the National Research
Dr Guylene Proulx presents the results ofa study into occupant behaviour
during afatal high-rise apartmentfire
Critical factors in
Fire Prevention 291 July/August 1996
25
-Figure 2. Simple representation of the scissor-stairs found in the Forest Laneway Building.
o
They will provide valuable informationon the behaviour of different occupants during the event and the rationale behind their decisions
o
Findings will enhance understanding ofthe conditions encountered in the building during the fire.
This knowledge can result in improved
codes, regulations and practicestoprevent
future similar occurrences. Although tragic, this fire provided a unique opportunity to study the occupants' behaviour under actual
fire conditionsl
.
The study used a questionnaire mailed to
occupants ofevery unit in the building. The questions were aimed at identifYing the way and the time at which occupants became
aWare that something unusualwas happening
and, once they became aware of the fact that it was a fire, on their first few actions. There were questions on the time at which occupants left their units, the fire alarm's audibility, the smoke and lighting conditions encountered during evacuation, and the actions occupants undertook to ensure their safety. The questionnaire also determined
occupants' previous exposuretofire safety
information. Variables such as gender, age and limitations were identified, since these can play an important role during a building evacuation. Completed questionnaires were received from 54% of the units from which responses could be expected (the units occupied by the fatalities' families were excluded).
-RESULTS"
Initial response
Over one third of respondents did not think that the situation was at all serious, while only 18% believed that it was extremely serious. Women and respondents aged over
65 were more likelytoperceive the situation
as serious; the location within the building of the respondent made no difference.
Most occupants' initial response was to
investigate, ie check the corridor, feel the door or look out of the window (62%) or seek information, ie call a neighbour or the building management (9%). Another 15% simply waited, while 6% alerted someone else. None of the residents immediately began to evacuate.
Occupants who had received fire safety
information were more likelytoinvestigate,
whereas those who had not tendedtoseek
information or wait.
Most occupants (64%) only became aware that there was an emergency on actually seeing or smelling smoke, while another
East
Floor 6
Floor 4
,
Floor 5
The study
Conducting a study of the occupants' behaviour during a fire incident is one of
the best ways for researcherstolearn about
the impact of human factors on the circumstances and outcome of a fire. This is because:
o
The survivors. of a fire are primewitnesses.
o
They can easily describe their perceptionof the event, their interpretation and their reactions during the fire.
staircase that was closesttotheir apartment.
This could mean that they initially triedto
use the closest staircase bur, because of
excessive smoke in that staircase, decidedto
try the other one. Two of the victims were found above the 30th floor, where the
staircase leadstothe roo£
Either they were lost in the stairs trying to
return to their apartments, or they were
tryingtoescapetothe roof, but would have
been incapable ofdoing so because there was
no accesstothe roof.
/ / / / / / / / / /1'-/ /1'-/ I ... /' I ... / I ...
Other Stairs
/ / /
I ...'1 • c, / / Iｾ / / I / / 11 ...'y/ I / / 11--/ I I<
I I I ... I I I ... I I ... I I ... " " / I I IｾＮＮＮＮＮＮ I II ... IIFloor 6
Floor 5
ｾ
Fire Stairs
Floor 4
West
not familiar with the building and who had
toorient themselves in the smoke. During
the fire, access doorstoboth staircases were
repeatedly opened and closed by occupants and rescue personnel on every floor, which
contributed significantlyto the rapid and
extensive spread of smoke throughout the building.
Fatalities
The six fatalities, two men and four women, were all aged between 16 and 35 and all lived on or above the 16th floor (above the neutral plane). They all travelled in the smoke before they collapsed. One victim, a woman from the 16th floor, was found in the stairs on the 27th floor. Some of the victims were removed from the staircases by other occupants or rescuers who attempted to resuscitate them.
Dueto the complexiry of the staircases, it
was difficult laterto retrace exactly each
victim's evacuation movement. Two of the victims were found in the Fire Staircase, and the other four were found in the Other Staircase. Three of them were not in the
26
A fire began in me boarded-up apartment
15% were told by others and 9% saw fire appliances outside the building. Ninery six people (46%) watched the television or listened to the radio to glean news of the situation. They found this frustrating and frightening as the news reported that there was a fire, which they already knew, and that people had died without giving any advice to people still left in the building.
Most of the occupants (55%) who did not attempt to evacuate decided to stay in their unit because there was too much smoke or heat to evacuate; another 20% stayed because emergency personnel told them to, while 13% believed that it was safer inside. All occupants who stayed in their apartmems throughout the incidem were safe, especially if they sealed their from door and vents. THE EVACUATION
Most of those who evacuated without assistance (17% of respondems) did so between 0500 and 0600 hours, ie within an hour of the beginning of the fire. Another 6% evacuated between 0600 hours and 1000 hours. All others were evacuated by rescue personnel once it was safe, starting at around 1000 hours.
EVACUATION VIA THE LIFTS Five occupants who lived on the fifth floor, including the occupam of the flat in which the fire began. evacuated successfully early on using a life. .
Three other respondents reported attempting to use an elevator to evacuate but found that it was not in operation or were forced back by smoke. The emergency services later used the lifts to evacuate many of the building occupams.
EVACUATION VIA THE STAIRS 105 respondents (or 48%) attempted to evacuate using the stairs. However 45% of them turned back, generally because of smoke, and returned to their units or sheltered in other units until rescue personnel arrived. 14 people (or 13%) of stair users had to change staircases due to smoke.
The persontoescape from the highest floor
using the Fire Staircase, a woman from the 20th floor, explained her successful evacuation through a dream. 'I had a dream on the night of the fire,' she said, 'that I and my neighbours had to evacuate the building by the stairwell because ofan earthquake, not a fire. When the alarm sounded, I was memally prepared.' Above the fire floor, only nine people managed to escape successfully by the stairs without help before 0600 hours; four by the Fire Staircase and five by the Other Staircase. They all started evacuating before 0520 hours.
Five respondents reported difficulry opening the door to the staircase for various reasons. Two people reported a locked door on the 6th floor: this door is in fact not fitted with a lock and another respondant reported using the door to evacuate a few hours later without any problem.
INJURIES
Eleven people (or 5% of respondents) reported that they were injured during the incident. Seven of them suffered smoke inhalation, rwo exhaustion, one exposure to the cold (while waiting on a balcony) and one did not specify the nature of the injury.
Significant factors
Four major factors which affected the behaviour of occupants and their likelihood of escape were idemified during the study. First ofall, the characteristics ofthe occupants of the building proved to be significant.
Age
The most significant differences were related to the ages of the respondems, since seniors aged 65 and ovet had a tendency to exhibit differem behaviour than younger occupants. D Seniors were significantly less likely to hear the building fire alarm and therefore were alerted by neighbours who told them about the fire.
Fire Prevention 291 July/August 1996
D Since seniors were usually warned by
others, they learned about the situation at a later time than younger residents who perceived smoke or heard the fire alarm.
D Once aware of the problem, older
occupants appear to have been quite realistic about their physical limitations.
Thoseｬｩｶｩｾｧ above the fifth floor did not
attempt an evacuation by themselves, preferring to wait for the assistance of rescue officers.
D Prior to that day, as many as 24% of
seniors had nor had any fire safety ,?
information, compared with only 15% of younger respondents.
Location
The second important factor was the physical location of the respondem in the building.
D The situation was much more serious
on the fifth floor, where the fire started, and was accurately perceived so by its occupants. The location of the occupants had a direct effect on their choice ofactions and on their evacuation possibilities.
D The four floors situated below the fire
floor appear to have been only slightly affected by the fire and most occupams of these floors evacuated unaided, using the stairs.
D Occupants on the upper floors were more
likely to have to turn back and seek refuge
when attemptingtoevacuate via the stairs.
When the situation was comrolled, rescue officers took them down in the lift.
D A greater proportion of upper-floor
residents thought that they could go to the roofto take refuge, even though there is no access to the rooE
D Alarm bells were located in the corridor on
each floor. Residents ofapartments situated in the corners of the building were significantly less likely to hear the fire alarm
duetodistance and to sound attenuation.
Smoke conditions
The third factor, smoke conditions, cannot be considered separately from the location of the respondems since smoke conditions varied greatly from one area of the building
toanother.
D All respondems on the fifth floor saw
smoke in their units, in the corridor and in the stairs, when they used them.
D With the exception of the fifth and sixth
floors, the floors below the 15th had significantly less smoke than the floors above it. This confirms the neutral plane phenomenon of smoke movemem in a high-rise building.
ｾ..---_.- ---_._--_.-..-...._.__.
__
..-..----_..._•...｟ｾＭＭＭ｟Ｎ｟ＭＭ｟Ｎ｟ＭＭ｟Ｎ｟ＭＭＭＭＭＭＭＮＭＭｾＭＭＭＭＭＭＭＭＭＭＭＭＭＭFire Prevention 291
July/August
1996
27
I - -
Aware of fire - Leave unit - Exit buildingSummary statistics of evacuation over time
References
1. Proulx, G., Pineau,
J.,
Latour.J.e.
andStewart,L., 1995. Study of the occupants'
behaviour during the2Forest Laneway fire in North York. Ontario,january6,1995, Imernal Report, IRC-IR-705, Institute for Research in Construction, National Research Council of Canada, Orcawa, Ontario.
When providing information on fire safety in high-rise buildings, a number of points
need to come across very clearly. Most
occupants seem to know that they should not use the lift during a fire but many do not fully understand the dangers of smoke and the importance of sealing their doors and vents and closing windows and doors. Once fire safety information is provided, it
would be useful to assess the occupants'
understanding of what they are expected to do during a fire by carrying out fire drills. Holding regular fire drills is a good way to educate occupants and to increase public
awareness of fire safety0
Dr Guylene Proulx is a Research Officer at the National Fire Laboratory, National Research Council ofCanada. Ottawa. Ontario. KIA OR6.
100 90 80 70 <: ｾ 60 u c., ｾ 50 0g "3
e
::l 40 U 30 20 10 0 3 4 6 7 8 9 10 11 12 13 14 15 16 Tune(hours after midnight)be perceived in some apartments, particularly in the corner units. Older occupants in any location were less likely to hear the fire alarm. This could be due to the location of the person, to the high frequency ofthe alarm or to the non-modulation of the signal. This fire demonstrated that communication to the occupants on immediate measures to be taken during a fire needs to be improved.
The use of aPAsystem could allow rescue
officers and building management to
communicate with occupants during an emergency.
Analyses showed that many occupants were prepared to travel through smoke, even though it could endanger their lives. This illustrates die need for public education on the dangers of smoke.
A large number of occupants had not received appropriate fire safety information. Some specific groups of people lacked it, particularly senior citizens, the unemployed
and those with different cultural
backgrounds. It seems important to use multiple means to communicate fire safety information to make sure that the message reaches building occupants and that everyone has access to the information.
The graph shows the times at which the respondents became aware of the fire, left ;heir units. and exited the building. The times are all as estimated by the respondents and may not be enmely accurate. By 0600 approximately half had left their units. but not all left the｢ｵｩｬ､ｩｾｧ ｳｴｲ｡ｩｧｨｾ away. They seemtohave attempted an evacuation and then either been forced backtotheir own Units or had to seek shelter elsewhere. usually due to smoke.
o
Even between the seventh and 14thfloors, more than 40% of respondents indicated that they saw smoke in their units, which denotes the possibility of leakage (ie smoke entering through vents and around the doors).
o
Residents located in the north-eastquadrant, which was the fire quadrant, were more likely to see smoke in their apartments than others.
o
Since 65% of the respondents who sawsmoke reported that the smoke entered their units around the door, it is likely that the smoke was quite dense in the corridor and that it was partly coming from the staircases. Both staircases appear to have been clogged with smoke only minutes after the fire alarm sounded.
DISCUSSION
Time to start evacuating
The time to start evacuating is the fourth factor and it is closely linked to smoke conditions. The presence of smoke was the major factor in determining the evacuation potential of occupants. For people living above the fifth floor, only those leaving their units at a very early time had a chance of reaching ground level safely.
As time progressed, the propagation of smoke and heat in both staircases made it impossible for people to get past the fifth floor and many had to return to their apartments or seek refuge in another unit until the situation was under control. They were evacuated later by rescue personnel.
Three important aspects of occupant fire safety should be raised.
o
An early attempt to evacuatecanprovidea better chance of exiting the building safely.
o
Communication to occupants during afire is essential.
o
Fire safety education and goodknowledge of the building evacuation procedure are paramount for the safety of occupants.
It is clear that the considerable propagation of the smoke impeded the evacuation of many people and increased the risk to their lives. It is impossible, however, to determine precisely the location and time at which the situation became intolerable and prevented
any further evacuation.An early exit attempt
gave the best chance for a safe evacuation of the building through the stairs.
The sound level of the alarm was judged overpowering in the corridors and toO low to
Fire
Prevention
bue 291J
ulylAugust 19%Editor
Liz Catchpok
CONTENTS
Detector developments, p.9-24
Yorkshire FCDA was unsuccessfully sued for negligence follOWing this school fire, in one of a recent rash of cases, p.8
Front cover:A blaze at an MFI store in Gloucester adds weight to calls for sprinklers to be installed in large single-storey buildings, pA Negligem firefighting - are brigades liable?
Malcolm Dewis examines some recent cases.
A balanced view of fire
Steve Brown examines the benefits of multi-criteria and multi-sensor detectors.
Ionisation detectors in the balance
Two views on what the future holds.
Firex review
The latest developments in detection technology.
16 14
28
Statistical focusStorage area fires during 1994.
3 I All that flickers
Recent developments in flame imaging technology.
Regular pages
Letters 13 Publications 18 Parish pump 33 LPCB News 34 Statistical information 36 Fire reports 38Products and services 42
From the Chief's desk 44
Calendar 45
LPC/FPA Training 46
Hazard sheec 47
24 Critical factors in high-rise evacuations
Dr GuylEme Proulx presents a study into occupant behaviour during a fatal high-rise apartment fire.
22 Addressable detection in healthcare premises
Analogue addressable detectors play an important role in NHS fire safety regulations, as David Charters explains.
20 Developmems in analogue addressable systems
Dr Donald Macfarlane highlights some product and industry trends.
News
.
3 Workplace Regulations update' Spate of supermarket fires' Fire threatens gas holders' 'Nightmare' £ 18m bill for Hampshire • New extinguisher standard • Local authorities unite over fire service funding crisis • Company fined £8000 for means of escape offences
9 8
Special features
Assistant editor
BukyPaym
The Fire Protection Association is the national fire: safety organization
ofme United Kingdom, oneoftwenty similar bodies existing worldwide for me promotion of greater fire safety. The FPA is one of five component parts ofthe Loss Ptevention Council, me UK-based organizarionsupported
by the:AssociationofBritishInsurers
and Lloyd's involved in all aspects of loss prevemion and eomrol. The FPA
journalFire PrnJrnnondisseminates
information about LPC activices as a
whole in addition to its particular coverage of fire safety matters. The
views expressed in articlesbyoutside contributors arc: rnosc: of the authors
and nor necessarily mose of the FPA. Claims made for products and services
in news items or advertisements do
not imply endorsemem by the FPA. No responsibility is accepted for any
such views or claims.
Publications secretary
PatridaKIijn
The Fire Protection Association
Melrouａｶｾｮｵ･Ｎ Borehamwood Hmfrmishir< WD62BJ T<I:0181 2072345 Fax: 0181 2076305 Advertising enquiries Micha./Gak
The Fire ProttC'tion &socitzrion
T.i: 0181 2072345 Fax: 0181 2076305
Advertising｣ｸ･｣ｵｴｩｶｾ
Pauk't< Elgi.
T<I:01366501176
Design and production editor
John Hay"
Design and layout
R.b.kah Grry
Consultant editor
Arthur Boy'
Publications and edirorial manager
Micha" Gak
© Fire Protection Association 1996
rlONAL
FP
lpartment,
House,
