Zeszyty Naukowe SGSP 2025, No.

93(1) 59

REQUIRED AND SAFE EVACUATION TIMES – THEORY

VERSUS PRACTICE IN A MEDICAL FACILITY SETTING
Marek Długosz1*, Marek Marzec2, Wojciech Kowalski2, Andrzej Krauze2, Klaudia Madej-Węgier2,
Robert Piec2, Barbara Szykuła-Piec2, Paweł Wolny 3
1
Independent author
2
Fire University, Warsaw, Poland
3
Lodz University of Technology, Lodz, Poland
*Correspondence: [email protected]

Abstract
Evacuation safety in medical facilities, such as hospitals, is a key element of emergency management,
especially in the context of people with limited mobility. The article discusses the specifics of
evacuation in hospitals, which are facilities with complex infrastructure and require adjustments to
accommodate people with disabilities. Research conducted at a specialized hospital showed that
actual evacuation times differ significantly from the theoretical model used in the British standard.
The article provides an analysis of sources of such discrepancies, discusses assumptions of PD 7974-6
standard and its applicability.

Keywords: emergency evacuation; indoor wayfinding; people with disabilities: evacuation drill;
computer simulation; accessibility

1. Introduction

The safety of patients, medical staff and other people staying in hospitals is
a priority task in the management of healthcare institutions. Critical incidents
such as fires, leaks of hazardous substances or natural disasters may require an
immediate evacuation of the facility. The specificity of a hospital as a medical
facility is associated with additional challenges in those situations. These result
from, among others, the presence of patients with limited mobility and the need to
coordinate multi-entity activities (McGlown, 2001).

According to the applicable regulations in Poland, each facility intended for the
presence of people must provide conditions enabling quick and safe egress from
the danger zone. Hospitals are classified to ZL II category of human hazard, which

DOI: 10.5604/01.3001.0055.0606
Received: 28.11.2024 Accepted: 17.02.2025 Published: 25.03.2025
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
60 Zeszyty Naukowe SGSP 2025, No. 93(1)

is reserved for facilities where at least part of the occupants has impaired mobility.
From this vantage it is necessary to adequately adapt the infrastructure of ZL II
buildings. Technical requirements include, for example, the appropriate number of
evacuation exits of a specific width with evacuation routes and access points with
rigorous parameters. In the light of Polish regulations, particular attention should
be paid to smoke management, alarm systems and emergency lighting, which are
designed to enable safe egress in the event of a fire Polish Journal of Laws/Dz.U.
2010 No. 109 item 719).
The adequacy of safety measures adopted in the facility may be proven on
a prescriptive basis – strictly following the code and referenced standards. However,
in many buildings, especially the existing facilities or critical infrastructure (i.e.
hospitals) it is not feasible to meet restrictive requirements of the code. In such
a situation it can be demonstrated that egress conditions are not worse than they
would have been when design adheres to applicable regulations. This approach is
called performance-based and is grounded on the concept of Available Safe Egress
Time (ASET) and Required Safe Egress Time (RSET). There are multiple methods
for assessing those measures, however, not all of them are well-suited for ZL II
category facilities (i.e. hospitals) without specific adjustments.
The aim of the research project outlined in the paper was to assess the applicability
of one of the most common methods of RSET calculation provided in the British
Standards (British Standards Institution, 2019) for fire safety assessment in hospitals.
Analytical results were confronted with experimental data obtained during
evacuation drill of a specialized hospital. Moreover, to refer those values to the
ASET, a computer analysis of fire development in analysed facility was performed.
Results of the research have been discussed with consideration to the Polish
regulations and challenges faced by medical institutions in the case of a fire. The
article contains with brief conclusions on RSET calculation methods, validation of
those methods with evacuation drill and suggestion of further research in the field.

1.1. Hospitals in the health care system

The Act of 15 April 2011 on medical activity (Polish Journal of Laws/Dz.U. 2011
No. 112 item 654) , which entered into force on 1 July 2011, replaced the concept
of a health care facility with that of a medical entity. Previously, pursuant to Article
1 of the Act of 30 August 1991 on health care facilities (Polish Journal of Laws/
Dz.U. 1991 No. 91 item 408), a health care facility was an organizationally separate
group of people and assets established and maintained for the purpose of providing
health services and promoting health.

The current law specifies:

• principles of performing medical activities;
• rules for non-business healthcare providers;
Zeszyty Naukowe SGSP 2025, No. 93(1) 61

• p rinciples for keeping a register of healthcare providers;

• working time standards for employees of healthcare entities;
• principles for the supervision of the exercise of therapeutic activity and of
providers of therapeutic activity.

Health care may be as follows:

• entrepreneurs;
• independent public health care facilities, budgetary units, including state budgetary
units;
• research institutes;
• foundations and associations;
• legal persons and organizational units acting under the provisions on the
relationship between the State and religious associations;
• military units.

A hospital is a medical facility that provides medical activities such as hospital

services. Hospitals operate in a 24-hour regime providing health services consisting
of diagnosis, treatment, care, and rehabilitation that cannot be performed as part
of other stationary health services (Polish Journal of Laws/Dz.U. 2011 No. 112
item 654).

1.2. Problems with hospital evacuation

One of the biggest challenges during hospital evacuation (or any other facility
providing medical care) is the evacuation of people with severe mobility
impairments. Regardless of its cause or severity, at the moment of an incident they
are unable to evacuate from the danger zone on their own. Hence, for the purpose
of evacuation management, all those patients need to be treated as persons with
disabilities (PwD). However, the evacuation of PwD requires special attention
given their specific needs, limitations of the infrastructure and evacuation
procedures. As indicated in the article “Emergency evacuation of people with
disabilities: A survey of drills, simulations, and accessibility” (Hashemi, 2018),
PwD are faced with numerous obstacles, such as the lack of infrastructure
availability, difficulties in emergency communication or the failure to adapt
standard evacuation procedures to their needs. PwD often encounter physical
barriers such as narrow passages, stairs or lack of access to ramps, which
significantly extends the evacuation time. In addition, alarm systems based on
sound signals are insufficient for the deaf, while blind people may find navigating
through buildings difficult due to illegible signage and the lack of adapted
evacuation routes (Hashemi, 2018).
62 Zeszyty Naukowe SGSP 2025, No. 93(1)

On the other hand, Tuvalu’s research has shown that the lack of awareness and
adequate preparation among both PwD and their families significantly increases
the risk in crisis situations. Low involvement of people with disabilities in the
evacuation planning process results in overlooking of their needs and as a result
leads to more injuries and deaths during critical incidents. To adapt the building
for evacuation of PwD multiple physical changes are required, such as wider
doorways, ramps, and appropriate signage and alarm systems adapted to different
types of disabilities. Accessible evacuation routes for people using wheelchairs
have been proven to significantly reduce evacuation times. Both simulations and
actual evacuation drills should include people with disabilities to better understand
their needs and identify potential problems in procedures. Such exercises also
help raise awareness among staff and other participants (families, visitors etc.).
PwD need adapted communication, such as visual alarm signals, Braille
instructions and educational materials in plain language. Educational programmes
should involve PwD along with their families and caretakers to enhance crisis
preparedness. People with disabilities should be involved in creating evacuation
procedures. Their experiences and perspectives can significantly expand
evacuation plans and increase their effectiveness (Elisala, 2020).

1.3. Evacuation strategies

Hospital evacuation may be carried out in various ways, depending on the location
of the threat, the number and condition of patients. Four main strategies may be
highlighted (Kuligowski and Omori, 2014):
• total evacuation,
• phased evacuation,
• delayed evacuation,
• stay-put strategy.

Total evacuation assumes that all occupants of the endangered building leave it
simultaneously. This strategy is often used during evacuation drills and can also
serve as a response to an actual emergency. However, this strategy is not always
feasible due to the safety of the building occupants – i.e. in multi-storey buildings.
During evacuation, occupants from the top floors must travel a considerable
distance to reach the final exit. Another aspect is the congestion on the evacuation
routes, especially staircases. When total evacuation is applied, congestions on the
evacuation routes can lead to extended evacuation time. Implementing total
evacuation strategy can pose a risk to the occupants’ safety and prove to be
ineffective (Kuligowski et al., 2010).

Phased evacuation splits the building and its occupants into phases according to
evacuation priority. The highest priority (first phase of evacuation) is assigned to
Zeszyty Naukowe SGSP 2025, No. 93(1) 63

people directly in endangered areas. Then, people in the vicinity of the critical
areas egress to the adjacent safe zone or outside the building. If the situation
requires it, additional phases are initiated. This strategy helps to avoid congestions
on evacuation routes and exits and reduces people flow density. To effectively
implement this strategy, the building needs to be divided into at least several fire
compartments. The effectiveness of evacuation will be improved by the availability
of fire protection systems in the building, training and preparation of staff, and the
provision of appropriate means of emergency communication in the facility. Main
disadvantages of the phase evacuation include potential panic among those who
must remain on site and wait for help from the emergency services. These people
may feel threatened and may try to act on their own, potentially disrupting the
rescue operation (Kuligowski and Omori, 2014).

The idea of ​​delayed evacuation is to move evacuated occupants into a building to

designated safety zones or adjacent fire compartments, where they can wait for
further instructions from those in charge of the rescue operation. In this context,
delayed evacuation refers to the deliberate postponement of evacuation as part of
an adopted strategy, rather than a situation forced by specific conditions. This type
of strategy can be useful in evacuating people with limited mobility—both
permanently and temporarily, such as people injured in a fire, people with injuries
or those with immobilized lower limbs. Their independent evacuation is
challenging and often even impossible. In such cases, the help of rescuers or
caregivers is needed to reach the safe zone, especially when the only possible
escape route is vertical, such as stairs. Therefore, this strategy seems to be functional
and especially effective in high-rise buildings with many PwD or more generally
buildings of ZL II category (intended primarily for use by persons with reduced
mobility, such as hospitals, nurseries, kindergartens, homes for the elderly). This
solution has been used successfully around the world, and implemented into
building codes in some countries, such as Sweden or Hong Kong, mandating the
provision of refuge areas in specific buildings to implement this evacuation
strategy. The effectiveness of this strategy will be influenced by the repeatable
layout of rooms on each floor. This helps in reducing the time needed to familiarize
potential evacuees with the building, which in turn improves the evacuation
process and enhances daily use. When escaping, occupants typically have two
routes to choose from. This strategy will make it easier to choose an evacuation
route in a crisis, as people prefer to use the routes and passageways they routinely
use (Wolny, 2021).

The concept of stay-put strategy is that occupants remain in their current location
during an emergency. People should seal the door to the room they are in and wait
for rescuers. Additionally, it is recommended to inform rescuers of your location
i.e. by hanging a white flag through the window or notifying the emergency centre
by phone. This concept is especially important for evacuating PwD. People with
64 Zeszyty Naukowe SGSP 2025, No. 93(1)

limited mobility, cognitive or visual abilities may have difficulty moving around
the building. They may also panic due to their inability to recognize the threat,
making stay-put strategy the most optimal solution. If possible, this strategy should
be discussed with PwD using the facility so that they know how to proceed in an
emergency and get their feedback. Stay-put concept was proven to work best
during emergencies in high-rise residential buildings, such as student housing,
hotels or apartment complexes (Proulx, 2002).

As regards hospitals, which are characterized by a complex infrastructure and the

presence of people with limited mobility, it is especially important to adapt
evacuation plans to the individual needs of patients. Providing appropriate
infrastructure changes, such as wider doors, ramps or alarm systems adapted to
different types of disabilities, can significantly improve the efficiency of evacuations.
In addition, regular evacuation exercises that take into consideration the diversity
of users are essential to identify potential problems and increase awareness among
staff. Collaboration with people with disabilities in the evacuation planning process
can provide valuable information and contribute to the creation of more effective
procedures (Wolny, 2021).

2. Methodology

The research consists of three main parts. The first and second are based on
calculations according to the British Standard PD 7974 (British Standards
Institution, 2019) and aim to assess ASET and RSET with common engineering
tools. The standard defines evacuation safety using two terms: Required Safe Egress
Time (RSET) and Available Safe Egress Time (ASET).

The basic concept behind the ASET and RSET analysis is to ensure ASET higher
than RSET. For many applications the appropriate safety margin (the difference
between the two) is required. Hence, to prove evacuation safety those two
parameters need to be calculated.

2.1. Characteristics of the facility

The facility in which the trial evacuation was carried out for the purposes of this
study was a specialized hospital administered by provincial government. The
hospital was built in the mid-1960s. There are total of 11 buildings in the facility.
The main building is divided into pavilions A to F. The entire building is a single
fire compartment and is classified to the ZL II category of hazard for humans. The
building has 4 floors, including 1 below the ground level. The ward that was the
subject of evacuation drill was on the 2nd (top) floor.
Zeszyty Naukowe SGSP 2025, No. 93(1) 65

Table 1. Technical data of the hospital building

Data of facility Parameter
Building area 6977.9 m2
Usable area 15824.3 m2
Total area 22802.2 m2
Building height 11.85 m
Total volume 79,724 m3
Number of storeys 4
a) underground 1
b) above the ground 3

2.2. Required Safe Egress Time

RSET is the time from the outbreak of the fire to the moment when all users have
safely left the building. It consists of the times of: detection, alarming, pre-
evacuation (recognition) and travel (passage through the evacuation route).

Formula 1 – Required Safe Egress Time

Where:
– required safe egress time
– time to detection
– time to a general alarm or warning
– pre-travel time
– travel time

If the condition occurs, it is assumed that evacuation is safe

(British Standards Institution, 2019).

In the literature, including (Chołuj, 2012) we can find the accepted detection and
alarm times depending on fire alarm systems.

2.3. Available Safe Egress Time

ASET, on the other hand, is defined as the time from fire initiation until conditions
become unsuitable for evacuation, plus a safety margin (British Standards Institution,
2019).
66 Zeszyty Naukowe SGSP 2025, No. 93(1)

Critical conditions were defined for visibility and temperature based on the British
standard PD 7974-6. A visibility of 10 m was assumed as the limit for effective
evacuation due to the possibility of locating exits or evacuation signs and the
willingness to pass through smoke of corresponding density. In these conditions,
the concentration of toxic substances in most fires is also defined as tolerable
during a 30-minute exposure. The temperature criterion was set at 60°C. This
threshold is recommended as tolerable for a 30-minute exposure as well (British
Standards Institution, 2019, Purser and McAllister, 2016).

Conditions on the evacuation routes of the building during a fire were estimated
using a computational fluid dynamics (CFD) model. The geometry, initial and
boundary conditions were introduced into the Fire Dynamics Simulator (FDS)
software version 6.9.1, which is developed by the National Institute of Standards
and Technology (NIST). The most important parameters used in the analysis are
summarized in Table 2 and the geometric model of the analysed storey is presented
in Fig. 1.

Table 2. Physicochemical parameters of the fire model used in the numerical simulation

Parameter Value Unit Source

Fire area 37.56 m2 (Thomas, 1981)

Heat release rate
250 kW/m2 (Hopkin et al., 2019)
per unit area
Fire growth rate 0.01172 kW/s2 (National Fire Protection Association, 2024)
Soot yield 0.113 g/g (Hurley et al., 2016)
Carbon monoxide yield 0.024 g/g (Hurley et al., 2016)
Cell side size
0.2 m (McGrattan et al., 2024)
(regular cubic mesh)

Figure 1. Geometric model used in numerical simulation. The axes in the floor plane [m]
and the division into areas for ASET analysis are shown
Zeszyty Naukowe SGSP 2025, No. 93(1) 67

The fire scenario assumed the initiation of a fire in the dirt room on the second
floor of the building in pavilion C1. Considering the materials stored in the room
(mattresses, bedclothes etc.) polyurethane was assumed as fuel (Hurley et al.,
2016). Heat release rate (HRR) was limited by ventilation conditions to 8.17 MW
according to (Thomas, 1981). The maximum HRR obtained in the simulation
oscillated around 5.6 MW. Heat release rate per unit area was assumed as for hotel
rooms, similarly equipped to healthcare rooms (Hopkin et al., 2019). Fire growth
rate was assumed to be average (National Fire Protection Association, 2024). The
room of fire origin is separated from the adjacent ones by lightweight partition
walls. Therefore, the computer simulation assumed the possibility of fire spread
beyond the initial compartment. Each part of the building (pavilions) is equipped
with a separate staircase, which provides the possibility of evacuation for its users.
Hence, the analysis area was limited to the pavilion in which the fire was initiated
and the adjacent areas. These boundaries included staircases used by evacuees and
whose availability was a subject of the analysis. The duration of the simulation was
limited by exceeding critical conditions in all analysed parts of the building or by
the intervention of the fire brigade after approx. 1100 s (Kuziora, 2024).

Moreover, two possible door configurations were distinguished–a pavilion

separated by smoke-proof doors and a non-separated (open) pavilion.

2.4. Empirical evacuation time measurements

The third and final element of the study was travelling time measurement during
the evacuation drill, which took place in a ward of specialized hospital with group
of 33 poseurs simulating patients. Medical staff and local fire brigade were informed
of the drill and briefed in advance; details, however, (i.e. precise test fire location)
were concealed from them.

Time measurements were performed by manned checkpoints laid out along evacuation
routes. This solution allowed tracking the evacuation progress of each evacuee
separately with simultaneous limiting to necessary the amount of recorded data.

The training fire scenario was as similar as possible to the scenario simulated in
ASET analysis described in Sec. 2.1. Thus, a fire has started during the day shift in the
dirt room at the end of the pavilion (see Fig. 1). Time of the drill has been optimized
so that the number of medical staff at the ward was maximal for this facility.

Patients were played by students, average age of 24, from one of the universities.
They were equipped with the following incapacity simulators:
• back pain simulators,
• aging simulator,
68 Zeszyty Naukowe SGSP 2025, No. 93(1)

• v ision defect simulator,

• pregnancy simulator,
• protective headphones that simulate hearing loss.

The patients – the poseurs – were not familiar with the building and not aware of
evacuation exits locations. Although the situation was created artificially such
a measure assured a natural behaviour of evacuees. They relied solely on their
impaired senses and help from staff or fire brigade members.

3. Results

3.1. Required Safe Egress Time

RSET analysis, as described in Sec. 2.2, was performed according to PD 7974-6
standard. First of all, the total distance (horizontal and vertical) of theoretical
evacuation route was calculated along with characteristics of occupants and
evacuation conditions. Those parameters are compiled in Table 3 and then used to
calculate travelling time.

Table 3. Evacuation route data

Evacuation route parameters Value Unit
Number of evacuees 33 people
Evacuation routes total area 150.39 m2
people
Density of people on evacuation routes 0.22
m2
Horizontal length of the evacuation route
16.1 m
(from the furthest access to the staircase on the second floor)
Horizontal length of the evacuation route
20.87 m
(from the staircase to the ground floor emergency exit)
Length of the staircase path (horizontal – landings) 12.09 m
Length of the staircase path (vertical – stairs) 16.8 m
Length of the thread 15 cm
Riser length 35 cm
Local narrowing width 0.8 m
Total horizontal distance 62.3 m
Total vertical distance 16.80 m
Zeszyty Naukowe SGSP 2025, No. 93(1) 69

Detection time depends on the complexity and accuracy level of the system
provided in the building. The level of the system can be determined i.e. by the
application of automatic and autonomous detection devices and coincidental
operation of such devices. Manual detection time was assumed to be 35 s according
to numerical simulation as the time when smoke propagates outside the room of
fire origin.

Alarming time, on the other hand, depends on the way in which the notification of
such alarm is propagated among the occupants. It can be reduced with the use of
automatic warning systems with optical and acoustic signals. The standard
describes three categories of alarm systems (Table 4). For the analysed hospital the
former level A3 should be applied, since no automatic alarm system is present
there and the alarm is sounded by the medical staff of the ward.

Table 4. Alarming time data

Quality level of the fire alarm system Alarming time ta
Level A1 0 min
Level A2 2–5 min
Level A3 5 min–1h

Pre-evacuation time is reserved for reactions of occupants, i.e. double-checking

for the threat or gathering personal belongings or looking for the exit. This
constituent of RSET depends on alarming, management and building complexity
levels. The management level for the hospital under consideration was M3, since
no trained security personnel is present in the building. The complexity level was
assumed to be B3, since the building is multi-storey and its layout is complex.
According to Table 5, pre-evacuation time should be assumed to be more than
15 minutes. For the sake of the analysis, it was assumed a priori to be 16 minutes
for 1st percentile and 19 minutes for 99th percentile (including 1 minute for
wayfinding).

Table 5. Response time

Modifications to First percentile Last percentile
Behaviour category
possible scenarios tp (1%) tp (99%)
M1 B1 A1-A2 0.5 2
D: Medical care M2 B1 A1-A2 1.0 3
M3 B1 A1-A3 > 15 > 15
– For B2, add 0.5 min to find the exit.
Comments: – For B3 add 1 min to search for exit.
– M1 – audible warning system required.
70 Zeszyty Naukowe SGSP 2025, No. 93(1)

All parameters for RSET calculations were summarized in Table 6. According to

the standard two scenarios where considered – for the first occupants including
possible congestions and for the last occupants. Mid-values are also communicated
with relevant formulas, where required.

Table 6. Summary of RSET calculations

Parameter Factor (or comment) Value Unit

Detection
Human detection 35 s
time

Alarming
Warning dissipated by the medical staff 300 s
time

Pre-travel ∆tpre (1%)=16 min 960 s

time ∆tpre (99%)=19 min 1140 s
Horizontal route

m
k 1.4
s
m2
a 0.266
people
people
D 0.22
m2
m
Vs 1.32
Averaged s
velocity Vertical route

m
k 1.23
s
m2
a 0.266
people
people
D 0.22
m2
m
Vs 1.15
s
Zeszyty Naukowe SGSP 2025, No. 93(1) 71

Horizontal route

L 62.3 m
m
Vs 1.32
s

47 s

Travel time Vertical route

L 16.8 m
m
Vs 1.16
s

15 s

Sum 62 s
The flow of evacuees through the local narrowing

m
Vs 1.32
s
people
D 0.22
m2
people
0.29
ms

Flow time
0.8 m

people
0.23
s

P 33 people
people
0.23
s

143 s
72 Zeszyty Naukowe SGSP 2025, No. 93(1)

RSET (1st percentile)

1500 s
Required
25 minutes
safe escape
time RSET (99th percentile)

1537 s
25.62 minutes

Where:
– time to detection
– time to a general alarm or warning
– pre-travel time for the 1st percentile
– pre-travel time for the 99th percentile
k – velocity unaffected by the density people
a – empirical constant for density-dependant velocity reduction = 0.266
D – density of people on evacuation route
– walking speed
L – length
– walking time during travel time
– specific flow
– effective width of the narrowest passage on the escape route
P – population
– flow of people
– time required to flow through the exits (congestion)
– required safe escape time

3.2. Available Safe Egress Time

ASET was analysed in time and space domains – it was determined after what time
critical conditions were exceeded on evacuation routes in subsequent rooms of the
analysed space. Additionally, the maximum time until start of evacuation was
determined so that critical conditions in terms of visibility range were not exceeded
(Dorsz et al., 2024). The summary of times obtained as a result of the analysis is
presented in Table 7. The numbering of rooms refers to the designations in Fig. 1.
Zeszyty Naukowe SGSP 2025, No. 93(1) 73

Table 7. Time to exceed critical conditions, maximum response time and ASET in subsequent
spaces of the analysed facility
No. Room [s] [s] [s] ASET [s]
1 Dirt room and adjacent rooms 75 55 35 55
Corridor near the dirty room
2 100 125 -*** 100
(access to staircase no. 1)
3 Exercise rooms 110 485 60 110
4 Corridor to the hall 170 535 -*** 170
5 Patient rooms 230 -** 115 230
6 Hall (access to staircase no. 2) * 330 900 240 330
* The hallway became smoke logged only when the pavilion was not separated by smoke-proof doors.
** The condition was not exceeded until the estimated time of fire brigade intervention.
*** Determined only for rooms/halls, assuming a movement speed of 1.2 m/s.

The FDS program was validated against over a hundred series of experiments,
which were used to develop methods for estimating the uncertainty associated
with the obtained results (McGrattan et al., 2024b). For the analysis in question,
the uncertainty of smoke optical density estimation (and its transformation
– visibility range) was found to be 0.01%. For the temperature threshold, calculated
model uncertainty was at the level of 5%.

3.3. Evacuation drill – time measurements

Calculated ASET and RSET values were then compared with the actual time of
egress measured during the evacuation drill. The total number of 33 poseurs were
involved in the experiment: 10 persons with impaired mobility, 8 walking patients
with other injuries (including 1 stuck in the toilet), 7 visitors (including 1 pregnant
woman), 2 persons with hearing issues, 2 blind persons, 2 persons with spine
issues, 1 elderly and 1 foreigner. Time intervals were recorded between the start of
evacuation – alarm on the ward – and the first checkpoint located at the exit from
the room of evacuee origin (Fig. 2) and between the first checkpoint and the
second one located at the exit from the building (Fig.3). Travel time was calculated
as the sum of those two intervals (Fig. 4). Recordings have also been summarized
in Table 8.
74 Zeszyty Naukowe SGSP 2025, No. 93(1)

Figure 2. Distribution of time interval from evacuation start to egress from room of origin
for evacuees grouped according to impairment type

Figure 3. Distribution of time interval between egress from room of origin to egress from
the building for evacuees grouped according to impairment type
Zeszyty Naukowe SGSP 2025, No. 93(1) 75

Figure 4. Total travelling time for evacuees grouped according to impairment type

Table 8. Summary of results measured during evacuation exercises

No. Category Result
Evacuated 33
No. Evacuee category
1. non-ambulatory patients 10
2. patients with spine problems 2
3. hearing impaired patients 2
1 4. patients walking 7
5. elderly people 1
6. people with vision problems 2
7. people locked in the toilet 1
8. pregnant women 1
9. people visiting 6
2 Total evacuation time from the start of the exercise until its end 42 min (2520 sec)
3 Longest travel time between checkpoints 35 min (2100 sec)
4 Arithmetic mean of the travel time (from the drill start to evacuee egress) 18 min (1087 sec)
76 Zeszyty Naukowe SGSP 2025, No. 93(1)

Rescue and firefighting operations were not taken into consideration during these
exercises. The role of the fire service was limited to their presence at the scene and
coordination of the experiment.

4. Discussion
Based on the calculations performed acc. to PD 7974-6, RSET for the analysed
medical facility was equal to 1,537 s. Default values suggested by the standard were
used for velocity that does not include i.e. mobility issues. The analytical method
assumes a particular designed escape route to be followed by evacuees. Pre-
evacuation time was also assumed a priori as 16 (lower limit for alarming level A3
that includes wayfinding) and 19 minutes for 1st and 99th percentiles of reactions,
respectively. A comparison of the analytical results with experiment, where solely
travel time was measured as a total of 2,520 s, showed that total empirical travel
time was over 40 times higher. When referring to the averaged travel time recorded
in the experiment, the default analytical approach would underestimate travel
time over 17 times. It should be stressed that experimental data does not include
detection, alarming and pre-evacuation time, so the actual RSET could be much
higher and the distribution of discrepancies among RSET constituents is unknown.

ASET, assessed by a numerical simulation, was found to be 230 s for patient rooms
and 170 s for the ward corridor. Such a short time was due to the fast development
of the fire that was driven by material stored in the dirt room. Moreover, the doors
to the corridor were initially open and walls between the room where the fire broke
out and adjacent rooms were not bricked but lightweight cardboard structures.
The stringent assumptions are also supported by the lack of the smoke detection
and automatic fire suppression systems in the facility. Thus, when the threat is
finally recognized, it is already significantly developed.

When referenced to ASET, analytical RSET is exceeded over 6 times. Including

measurements from the evacuation drill instead of analytical travel time would
equal to 11 times for averaged travel time and 17 times for total time required to
move all of the ward patients to the outside. Obviously, without further technical
or operational improvements, the risk for human life in case of fire is unacceptable.

An operational solution for the issue at hand could involve the implementation of
delayed evacuation strategy. It could utilize the hallway as a temporary safe-zone
for patients evacuated from the ward area. However, when no smoke-proof doors
are applied between the hall and the ward, ASET calculated on the ward is 330 s.
According to the calculations and simplified comparison of RSET (total) to ASET
on the corridor, it would be exceeded over 4 times.

One of the temporary technical solutions proposed were smoke-proof doors

separating the ward from the hallway and other units of the facility. In the
Zeszyty Naukowe SGSP 2025, No. 93(1) 77

simulation where such a solution was adopted (row 6 in Table 7), even though
RSET values for patient rooms were not changed, the hallway was successfully
protected from smoke. Such a compartmentation, even if not full (only doors in
regular bricked wall, no fire-rated wall), could significantly release the pressure to
proceed with evacuation downstairs. Combined with the delayed evacuation
approach, the hallway could be used as a buffer zone, before further evacuation
through staircases, which could be especially challenging for the staff of medical
units.

Nevertheless, validating the analytical model proposed in PD 7974-6 using solely

measurements from evacuation exercises entails certain drawbacks and may be
prone to cognitive errors. The only factor that can be compared in case of the
experiment described in the paper is the travel time. The drill that was carried out
for the needs of this study did not fully reflect the real-case scenario and
measurements do not include all constituents of RSET. Another limitation of the
evacuation experiments in general is preparedness of people involved (staff and
poseurs/occupants). Although the drill in such a form has huge training and
educational value, any conclusions on human behaviour in a fire should be drawn
with caution.

5. Conclusions
In the article an overview was carried out of evacuation strategies and evacuation
conditions in medical facilities with particular emphasis on challenges characteristic
for this type of occupancy. An analysis of common calculation methods revealed
a limited applicability of those methods in their default form for such buildings.
This was especially the case for travel time. The study showed in a real-world case
study supported by empirical data that actual travel time for those facilities could
be on average 17 times longer than calculated in the model with its default
assumptions up to the 40 times for boundary case.

Regardless of informed and detailed analysis that can prove evacuation safety,
a proper design is required. In the current state RSET is greater than ASET over 17
times. However, with combination of appropriate technical and operational
solutions applied it was possible to meet safe evacuation criteria in the building
under consideration. Hence, particular attention should be paid to adapting the
hospital infrastructure to the needs of people with limited mobility, which may
significantly improve the effectiveness of evacuation. Moreover, regular evacuation
exercises that take into account the diversity of users are crucial to identifying
potential problems and increasing awareness among staff.

There are still gaps in knowledge on the evacuation of medical facilities. Although
some attempts have been undertaken to find parameters for adequate assessing
78 Zeszyty Naukowe SGSP 2025, No. 93(1)

people with disabilities in evacuation analyses, more empirical evidence is required.

At some wards the evacuation strategy is strictly imposed by the current state of
the patient (i.e. critical condition patient). Safe and scientifically valuable
evacuation drill in existing medical facilities is still an underexplored subject. Such
an experiments are difficult from the perspective of continuity of hospital operation
but so are interpretation of results and uncertainties linked to it. This area requires
further investigation since it can be an excellent source of overall empirical data
for models and analytical methods.

References
British Standards Institution, (2019). PD 7974-6 The application of fire safety engineering principles
to fire safety design of buildings. Part 6, Human factors: life safety strategies: occupant evacuation
behaviour and condition, Vol. 3, p. 49.
Chołuj, Ł., (2012). Bezpieczna ewakuacja a założenia scenariusza pożarowego (Safe evacuation and
established fire scenario). Bezpieczeństwo i Technika Pożarnicza, Vol. 3, pp. 127–130. In Polish
Dorsz, A., Zugaj, M., Łącki, W., Krauze, A. and Zawrotniak, A., (2024). Wytyczne weryfikacji instalacji
wentylacji strumieniowej garaży zamkniętych za pomocą symulacji CFD, Krasuski, A., (ed.).
Warsaw: Fire Univeristy. doi: 10.70402/apoz.2024.wytycznecfd
Elisala, N.T.A.M.M.M.T., (2020). Exploring persons with disabilities preparedness, perceptions and
experiences of disasters in Tuvalu. PLOS ONE, Public Library of Science, Vol. 15, No. 10, pp. 1–19.
doi: 10.1371/journal.pone.0241180
Hashemi, M., (2018). Emergency evacuation of people with disabilities: A survey of drills, simulations,
and accessibility. Cogent Engineering, Vol. 5, No. 1, pp. 1–20. doi: 10.1080/23311916.2018.1506304
Hopkin, C., Spearpoint, M. and Hopkin, D., (2019). A Review of Design Values Adopted for Heat
Release Rate Per Unit Area. Fire Technology, Vol. 55, No. 5, pp. 1599–1618. doi: 10.1007/s10694-
019-00834-8
Hurley, M., Gottuk, D., Hall, J.R., Harada, K., Kuligowski, E., Puchovsky, M., Torero, J., et al., (2016).
Appendix 3: Fuel Properties and Combustion Data. SFPE Handbook of Fire Protection Engineering,
5th ed. New York: Springer.
Kuligowski, E., Peacock, R. and Hoskins, B., (February 2010). A Review of Building Evacuation
Models, 2nd Edition, Technical Note (NIST TN). Gaithersburg, MD: National Institute of Standards
and Technology.
Kuligowski, E.D. and Omori, H., (2014). General Guidance on Emergency Communication Strategies
for Buildings, 2nd Edition. Gaithersburg, MD: National Institute of Standards and Technology.
doi: 10.6028/NIST.TN.1827
Kuziora, Ł., (2024). Stochastyczny model działań gaśniczych jednostek państwowej straży pożarnej
jako element oceny ryzyka pożarowego budynków. Doctoral dissertation. Warsaw: Fire University.
McGlown, K.J., (2001). Evacuation of Health Care Facilities: A New Twist to a Classic Model. Natural
Hazards Review, Vol. 2, No. 2, pp. 90–99. doi: 10.1061/(ASCE)1527-6988(2001)2:2(90)
McGrattan, K., Hostikka, S., Floyd, J., McDermott, R., Vanella, M. and Mueller, E., (2024a). Fire
Dynamics Simulator. Technical Reference Guide. Volume 3: Validation, Gaithersburg, MD, doi:
10.6028/NIST.SP.1018
McGrattan, K., Hostikka, S., Floyd, J., McDermott, R., Vanella, M. and Mueller, E., (2024b). Fire
Dynamics Simulator. User’s Guide, Gaithersburg, MD. doi: 10.6028/NIST.SP.1019
Ministry of Internal Affairs and Administration, (2010). Regulation of the Minister of Internal
Affairs and Administration of 7 June 2010 on fire protection of buildings, other construction
objects and lands, Poland (Polish Journal of Laws/Dz.U. from 2023 item 822 as amended).
Zeszyty Naukowe SGSP 2025, No. 93(1) 79

National Fire Protection Association, (2024). NFPA 204, Standard for Smoke and Heat Venting, 2024
Edition.
Proulx, G., (2002). Evacuation Planning for Occupants with Disability. doi: 10.4224/20378862
Purser, D.A. and McAllister, J.L., (2016). Assessment of Hazards to Occupants from Smoke, Toxic
Gases, and Heat. In Hurley, M.J., Gottuk, D., Hall, J.R., Harada, K., Kuligowski, E., Puchovsky, M.,
Torero, J., et al. (eds.), SFPE Handbook of Fire Protection Engineering. New York, NY: Springer
New York, pp. 2308–2428. doi: 10.1007/978-1-4939-2565-0_63
Sejm of the Republic of Poland (1991), Act on Health Care Facilities (Polish Journal of Laws/Dz.U.
from 2007 No 14 item 89 as amended).
Sejm of the Republic of Poland (2011), Medical Activities Act (Polish Journal of Laws/Dz.U. z 2024
item 799 as amended).
Thomas, P.H., (1981). Testing products and materials for their contribution to flashover in rooms.
Fire and Materials, Vol. 5, No. 3, pp. 103–111. doi: 10.1002/fam.810050305
Wolny, P., (2021). Holistyczne podejście do problemu ewakuacji. Łódź: Wydawnictwo Politechniki
Łódzkiej.