Proceedings of the 2013 International Conference on Applied Mathematics and Computational Methods in Engineering

Simulation of the course of evacuation in tunnel fire

conditions by FDS+Evac
Lukas Valasek, Jan Glasa
Institute of Informatics, Slovak Academy of Sciences
Bratislava, Slovak Republic
e-mail: lukas.valasek@savba.sk, jan.glasa@savba.sk

recommended the use of FDS for simulation of possible

consequences of fires in nuclear power plants [18]. FDS
allows to perform calculations on multiprocessor/multicore
computer systems. For visualization of simulation results, the
Smokeview [19] system which is a part of FDS is used. It
allows to visualize the results of 3D simulations of fire and
smoke in time and slices of selected quantities and to export
the visualized simulation results in the form of graphs, tables,
pictures and movies.

AbstractIn this paper, simulation of fire in a short 2-lane

road tunnel and its evacuation are described. Two traffic
situations in the tunnel and their impact on people evacuation in
fire conditions are analyzed. For simulation of the tunnel fire, the
FDS (Fire Dynamics Simulator) system, version 5.5.3 based on
advanced CFD (Computational Fluid Dynamics) fire model is
used. The evacuation of people in tunnel fire conditions is
modeled by the FDS agent-based evacuation module Evac,
version 2.3.1.
Keywordstunnel fire, people evacuation, FDS+Evac, CFD

I.

FDS also contains the evacuation module, Evac.

FDS+Evac [20] is a system, which simulates fire and its
impact on the course of evacuation and behavior of evacuees.
It allows to simulate both the fire evacuation and fire drill. In
FDS+Evac, CFD-based fire model and agent-based evacuation
model interact. The system is able to take into account the
information about fire (smoke) in every place of the space at
arbitrary time of calculation and model the impact of fire on
evacuation. Several other advanced evacuation simulation
systems, such as Pathfinder, buildingEXODUS, STEPS,
Simulex, etc., are available. Some of the existing systems are
able to import the information about fire from FDS or from
some other fire simulator and are able to utilize it partially for
evacuation modeling. The FDS+Evac system calculates 3D
simulation of fire on 3D computational meshes and 2D
simulation of evacuation on a single 2D computational mesh.

INTRODUCTION

Fires are catastrophic events which can cause material

damages, loss of human lives as well as the threat of human
health and environmental pollution. They are an indicator for
society that it is necessary to care about fire safety and prevent
society from possible fire consequences. Fire in a tunnel can
become such catastrophic event. Tunnels can be several
kilometers long and fire inside a tunnel can cause a large
destruction (loss of human lives, tunnel destruction and cars
damage). Tunnel fire can be dangerous for people during the
tunnel evacuation.
In the literature, several specific aspects of tunnel fire,
such as for instance the emergency ventilation action [1, 2, 3,
4], computer simulation of the course of fire and smoke and
their consequences [5, 6, 7, 8], and modeling of evacuation of
people in tunnel [9, 10, 11, 12], are analyzed. Numerical
calculation of computer simulation of tunnel fires requires
high computational power of computers, therefore, the
computational space must be divided into a number of
computational meshes and the calculation must be realized in
parallel [13, 14, 15]. At present, several program systems,
which are capable to simulate complex phenomena associated
with fire in closed or semi-closed structures, have been
developed. SMARTFIRE, FLUENT, SOFIE, JASMINE,
PHOENICS and FDS are examples of such systems.

In this paper, we use the FDS+Evac system for simulation

of fire in a short 2-lane road tunnel with simple ventilation
system and for evacuation of people in the tunnel. This
research utilizes our experience in the field of computer
simulation of fires in various environments and conditions, for
instance fire in forest [21, 22, 23], compartment [24], family
house [24], cinema hale [25, 26], automobile [15, 27], tunnel
[28, 29] and garage [30]. We have particularly studied the
impact of parallelization of calculation for accuracy and
efficiency of the tunnel fire simulation calculation [15, 27, 28,
29, 30].

In this paper, we used the FDS (Fire Dynamics Simulator)

system, version 5.5.3 [16, 17] which is based on the
knowledge about CFD (Computational Fluid Dynamics). It
was developed by National Institute of Standards and
Technology (NIST) in the USA. FDS has been verified and
tested by the U.S. Nuclear Regulatory Commission which

This paper is organized as follows. In Section 2, a 3D

model of a 180 m long 2-lane road tunnel, fire scenario and
two evacuation scenarios are described. In Section 3, the
results obtained by simulation of fire and evacuation in the
tunnel are described. Section 4 summarizes the main
conclusions and some of our future research plans.

288

Proceedings of the 2013 International Conference on Applied Mathematics and Computational Methods in Engineering

II.

DESCRIPTION OF FIRE TUNNEL AND EVACUATION

A. Tunnel model
We created a model of a single-directional 2-lane road
tunnel using Google SketchUp (see Fig. 1). The tunnel is 180
m long with the 10 m x 180 m x 7.2 m (width x length x
height) dimensions. The ventilation system of the tunnel
consists of two couples of jet fans placed about 1 m under the
tunnel ceiling at the distances of 47.4 m and 137.4 m from the
left tunnel portal. The fans are placed 3 m far from each other.
Their effective diameter and length is 0.9 m and 5.2 m,
respectively.

SCENARIOS

In this section, we describe a model of a tunnel, fire

scenario, ventilation action during the fire, representation of
input data for FDS simulation and two traffic situations in the
tunnel with different requirements for the tunnel evacuation.

B. FDS representation of tunnel, fire source and ventilation

The tunnel structure with vertical walls and a curved
ceiling (see Fig. 1) was represented in FDS using orthogonal
obstacles (OBSTRUCTIONs) from concrete (of the 20 cm
THICKNESS). In the curved ceiling representation, we used
the SAWTOOTH feature to smooth the ceiling surface in
order to prevent the origin of turbulent phenomena related to
the flow of gas around sharp corners (edges) of obstructions
from which the concrete ceiling was created.
We represented the jet fans in the tunnel standardly using
thin obstructions (OBSTRUCTONs with the 0 m
THICKNESS) with the POROUS=.TRUE. parameter. The
square fan cross-section corresponded to the circular crosssection of a standard tunnel jet fan shown in Fig. 1.
We assumed that the value of ambient temperature in the
tunnel was 20C. We also assumed a steady flow in the tunnel
at the beginning which consists of flow caused by the tunnel
structure as well as of the movement of vehicles through the
tunnel (traffic contribution).

Fig. 2 Ventilation action.

Fig. 3 HRRPUA of the fire source.

Fig. 1 3D tunnel model and its ground plan and the side and front elevations.

289

Proceedings of the 2013 International Conference on Applied Mathematics and Computational Methods in Engineering

We represented such a quasi-steady flow by ventilation action

so that all fans blew with the 6.25 m/s velocity during the first
60th s of simulation. After this manner, quasi-steady air flow at
the top part of the tunnel with approximately 2 m/s velocity
was created (see Fig. 2).
Fire in the tunnel started at the 50th s increasing linearly up
to its maximum value (10 MW) of heat release rate (HRR)
which was reached at the 55th s. Since that time, the value of
HRR was not changed until the end of simulation. In the
simulation, the fire was represented by a 2 m x 3 m surface
placed about 1.1 m above the road at the 92 m distance from
the left tunnel portal (see Figures 1 and 3), which produced
heat with the 1666.667 kW/m2 HRRPUA (heat release rate
per unit area). After detection of fire, we assumed the
following action of ventilation. At the 60 th s, all fans started to
work with the velocity increasing linearly from the value of
6.25 m/s to 25 m/s. The maximal velocity value was reached
at the 65th s and from that time it was not changed until the
end of simulation (see Fig. 2).
We did not assume any flammable materials that would
influence the course of fire in the simulation.
C. Evacuation scenario without vehicles with higher
capacity of passengers
In simulation, we assumed the following traffic situation
during the tunnel fire described above and the corresponding
evacuation scenario (it will be referred to as Scenario 1 in this
paper). The total number of vehicles (cars) which arrived
through the left portal into the tunnel was 24. Positions, in
which the individual cars stopped, are shown in Fig. 4. The
first car stopped at the 53rd s, i.e. three seconds after the fire
initialization. The next cars stopped at every second, therefore,
the last car stopped at the 76th s. The distances between
vehicles are shown in Fig. 4. The number of passengers of
individual cars, times of vehicles stop and evacuation times of
individual passengers are shown in Fig. 5. The total number of
passengers in this scenario was 65. The crews of cars in
Scenario 1 consisted of 1-4 people. We did not assume any
vehicles with higher capacity.
We assumed that all passengers knew (were familiar with)
the left tunnel portal, because they came through it. Passengers
of cars C1-C7 knew the portal as well as the exit (they saw the
exit before stopping the vehicle, see Fig. 4).
Fig. 4 Scheme of traffic in Scenario 1: positions of the C1-C24 cars in regard
of other cars and fire source.

The exit was about 1.5 m wide and was placed at the 73 m
distance from the left tunnel portal (see Fig. 4). We
represented it in FDS+Evac by a VENT object with the given
width, assigned evacuation mesh and the corresponding point
(X, Y, Z) placed in the middle of the exit. It was used by
evacuees (agents) to escape and its parameters were used as
input of the decision algorithm of agents and were used for the
calculation of preferred directions field which directed the
agent movement.

does not allow to consider low obstacles, which obstruct

agents in the movement but they do not obstruct agents to see
exit (i.e., to see the point (X, Y, Z) assigned to the exit). The
portal representation using a single exit with the portal width
(i.e., with a single point (X, Y, Z) assigned) would cause that
agents at side parts of the tunnel would not see the portal
because standing cars in the tunnel would obstruct them to see
the point (X, Y, Z). Therefore, we used the portal
representation using three exits, the width of which was
determined by the width of free spaces available for escape of

The left tunnel portal was represented in FDS+Evac, by

three individual exits of the 2 m width. They allowed agents to
escape through the portal. The current version of FDS+Evac

290

Proceedings of the 2013 International Conference on Applied Mathematics and Computational Methods in Engineering

Fig. 5 Description of vehicles evacuation in Scenario 1, where A, E, CH, F

and M is adult, elderly, child, female and male, respectively; C1,, C24 are
cars; AT is the vehicle stop time; ET is the individual passenger evacuation
time from vehicle; and LFD, RFD, LBD and RBD is the left front, right front,
left back, and right back door, respectively.

agents in the direction to the portal. Such representation made

agents able to see the portal escaping in the direction to it.
D. Evacuation scenario with vehicles with higher capacity of
passengers
In order to test the impact of vehicles with higher capacity
on the course of evacuation, we assumed 21 cars (1-4
passengers), one bus (30 passengers) and one transporter (9
passengers). In this scenario (see Fig. 6), we placed the bus B1
instead of the cars C2 and C4, and the transporter T1 instead
of the car C6. The rest of parameters of the traffic situation
remained unchanged. The passengers evacuation times and
vehicles stop times for B1 and T1 are shown in Fig. 7. The rest
of parameters for the passengers evacuation from the cars C1,
C3, C5, C7, C8,, C24 was the same as in Scenario 1 (see
Fig. 5).
In Scenario 2, we assumed that all passengers knew the left
portal (as in Scenario 1) and the passengers of the vehicles C1C7, B1 and T1 knew both the portal and exit (they saw the exit
before stopping vehicle). The exit was placed at the same
place as in Scenario 1 (see Fig. 6). The total number of
passengers in this scenario was 95.
Fig. 6 Traffic scheme in Scenario 2.

III.

SIMULATION RESULTS

The simulation was realized in parallel on a PC (6-core i73930K, 3.26 GHz, 64 GB RAM). The computational domain
was divided into three 3D computational meshes with the 10
cm mesh density on which the fire was resolved. The mesh
parameters fulfilled the conditions associated with efficient
calculation of the FDS pressure solver. One 2D computational
mesh was assigned for evacuation calculation. Each of these
computational meshes was assigned to one core. Thus, the
calculation was performed in parallel on 4 CPU cores. The
total computational time of the of the 180 s simulation of fire

Fig. 7 Description of vehicles evacuation in Scenario 2, where A is the adult,

M is the male, B1 is the bus, T1 is the transporter, AT is the vehicle stop time,
ET is the individual passenger evacuation time; and LFD, RFD, LBD and
RBD is left front door, right front door, left back door and right back door,
respectively.

291

Proceedings of the 2013 International Conference on Applied Mathematics and Computational Methods in Engineering

Fig. 8 Simulation of the course of fire and traffic situation at the 50th, 53rd, 54th, 55th, 56th, 90th and 99th s (Scenario 1).

and evacuation was 95.87 hours and 98.82 hours for Scenario
1 and Scenario 2, respectively.

B. Evacuation simulation results (Scenario 1)

The evacuation started at the 58th s (see Fig. 5) and ended
at the 180th s. The course of evacuation at some selected times
is shown in Fig. 9. From detailed analysis of the course of
evacuation, it follows that the passengers from the C1 and C7C24 cars used the left tunnel portal to escape, the passengers
from the C2 and C4 cars used the exit, and the passengers
from the C3, C5 and C6 cars used both the exit and portal.

A. Fire simulation results

The fire started at the 50th s and already at the 53rd s hot
gases hit on curved part of the tunnel ceiling and spread under
the ceiling (see Fig. 8). The quasi-steady flow in the tunnel
caused that the smoke was drifted more towards the right
tunnel portal than towards the left portal. Fig. 8 also illustrates
how the individual cars stopped during the period between the
53rd and the 56th s in Scenario 1. Since the 60th s, the
ventilation started to act reaching its maximum velocity of 25
m/s at the 65th s. The ventilation action caused that smoke
began to spread more rapidly towards the right tunnel portal.
At the 90th s, the C1-C4 cars were still threatened by smoke.
At the 99th s, the tunnel was devoid of smoke at the part of the
tunnel at the left from the fire source.

The detailed analysis showed that the passengers getting

out from the left doors of the cars stopped in the left tunnel
lane escaped towards the tunnel portal. This was caused by the
fact that they knew and saw the portal. Therefore, the portal
belonged to preferred exits unlike the exit which was not seen
by agents through stopped vehicles (the cars are not low
obstructions). Since the passengers from the C1-C7 cars knew
the exit (as well as the portal), it was possible to assume that a
part of the agents would choose the exit to escape rather than
the portal in the case that they would see the exit through cars
(low obstructions are to be involved in the next FDS+Evac
version). In the actual FDS+Evac version, it is possible to
achieve such behavior by setting the corresponding

The course of fire in Scenario 2 is similar to Scenario 1.

However, some differences were observed in the smoke
spread (smoke was slightly more drifted towards the right
tunnel portal and was exhausted sooner from the tunnel part at
the left from the fire source).

292

Proceedings of the 2013 International Conference on Applied Mathematics and Computational Methods in Engineering

properties of individual agents and exits.

In Fig. 11, the graph of using the exit and portal in time is
shown. Majority of evacuees used the portal (55 evacuees) and
minority of agents used the exit (10 evacuees). In Fig. 12, we
illustrate the efficiency of the agents evacuation through the
exit and portal. The portal was used by less than 85% of the
total number of evacuees. The exit was used by more than
15% of the total number of evacuees (65 evacuees).

Fig. 9 Course of evacuation at the 58th, 66th, 70th and 78th s.

Fig. 11 Using the exit and portal in time.

C7

C6

Fig. 10 Passenger from the C1 car in smoked environment.

A part of the agents which were in the middle of the tunnel

between two chains of vehicles (passengers from vehicles C2,
C3, C4 and C5) escaped thought the exit because they knew
and saw it. Other passengers escaped through the portal
because they did neither see nor know it. Some of them (the
passenger from the C1 car, see Fig. 10) did not see the exit
because of a dense smoke. Other passengers were inhibited to
see it by standing cars. The behavior of selected passengers (or
crews of selected vehicles) can be influenced so that they
choose the exit to escape similarly as it was mentioned above.
Fig. 10 illustrates the impact of smoke on behavior of the
agent escaping from the C1 car. The dense smoke obstructed
the agent to see the exit, therefore, he had to escape via the
portal.

Fig. 12 Efficiency of the agents evacuation through the exit and portal in time
(N is the relative number of evacuees in regard of their total number).

C. Evacuation simulation results (Scenario 2)

From analysis of the course of evacuation, it follows that
the passengers from the C1-C3 cars and the transporter T1
used the portal and exit, the passengers from the C4-C21 cars
used the portal, and the passengers from the bus B1 used the
exit to escape. The evacuation began at the 58th s and ended at
the 169th s. The course of evacuation in selected times is
shown in Fig. 13. Fig. 14 (top) shows a people jam originated
in front of the exit. The queuing at this place causes
emergency risk (unwanted contacts of evacuees, injuries) and
slowdown of evacuees movement. In Fig. 14 (bottom) an
agent from the transporter escaping towards the portal is
highlighted. This agent selected the portal as the best way to
escape taking into account waiting in the queue in front of the
exit. However, already at the 83rd s he turned round and
escaped to the exit because the queue in front of the exit was
reduced. The exit became the fastest way for him how to
escape.

According to analysis of the evacuation, the passengers

getting out from the right doors of the C2, C4 and C6 cars
standing in the right tunnel lane escaped through the exit and
the others escaped towards the portal. From settings of the
passengers of the C8-C24 cars, it follows that the portal was
their preferred exit (they saw and were familiar with it) rather
than the exit (it was seen only). In order to influence the
agents behavior, it is enough to change setting of the agents
(exits).

293

Proceedings of the 2013 International Conference on Applied Mathematics and Computational Methods in Engineering

In Fig. 15, the graph of using the exit and portal in time is
shown. Most of agents used the portal (54 evacuees). A
slightly less number of evacuees used the exit (41 evacuees).
In Fig. 16, the efficiency of the exit and portal is illustrated.
The portal and exit was used by more than 56% and by less
than 44% of the total number of agents (95 evacuees),
respectively.

Fig. 15 Using the exit and portal in time.

Fig. 13 Course of evacuation at the 58th, 60th, 70th and 80th s.

Fig. 16 Efficiency of the agents evacuation through the exit and portal in time
(N is the relative number of evacuees in regard of their total number).

IV.

CONCLUSIONS

In this paper, two scenarios of the evacuation of a short

single-directional 2-lane road tunnel in fire conditions were
described. The used FDS+Evac simulation system is based on
CFD fire model which is combined with people evacuation
model based on agent principle. The current version of the
FDS+Evac system has certain particularities which must be
taken into account to avoid incorrectness of input
representation of the tunnel, traffic situation in the tunnel,
people moving in the tunnel as well as the representation of
emergency exits. Such misrepresentation could distort the
results of computer simulation of the fire evacuation in the
tunnel. The tunnel portal representation and impact of fire and
traffic on the course of evacuation are illustrated. The
advantage of using FDS+Evac for simulation of tunnel
evacuation in fire conditions is the direct access of evacuation
module to the fire data provided by fire simulation.
Future investigations related to the impact of the tunnel
fire, ventilation and traffic as well as of the agents setting on
behavior of individual agents and/or groups of agents in the
tunnel in fire conditions will be needed. We plan to make such
analyzes using fine resolution meshes utilizing highperformance computing.
ACKNOWLEDGMENT

Fig. 14 Queuing in front of the exit at the 71st, 75th, 80th, 81st, 83rd and 86th s.

This paper was supported by the Slovak Scientific Research

Agency (project VEGA 2/0216/10).

294

Proceedings of the 2013 International Conference on Applied Mathematics and Computational Methods in Engineering

[1]

[2]

[3]

[4]

[5]

[6]

[7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

REFERENCES
R.O. Carvel, A.N. Beard and P.W. Jowitt, Fire spread
between vehicles in tunnels: effects of tunnel size,
longitudinal ventilation and vehicle spacing, Fire Tech.,
vol. 41, pp. 271-304, 2005.
R.O. Carvel, R. Guillermo and J.L. Torero, Ventilation
and suppression systems in road tunnels: some issues
regarding their appropriate use in a fire emergency, Proc.
of the 2nd Int. Tunnel Safety Forum for Road Rail, pp.
375-382, 2009.
Y.Z. Li, B. Lei and H. Ingason, Theoretical and
experimental study of critical velocity for smoke control
in a tunnel cross-passage, Fire Technology, vol. 49, pp.
435-449, 2013.
H. Ingason and Y.Z. Li, Model scale tunnel fire tests
with longitudinal ventilation, Fire Safety Journal, vol.
45, pp. 371-384, 2013.
S. Bari and J. Naser, Simulation of smoke from burning
vehicle and pollution levels caused by traffic jam in a
road tunnel, Tunnelling and Undeground Space
Technology, vol. 20, iss. 3, pp. 281-290, 2005.
L. H. Hu and N. K. Fong, Modelling fire-induced smoke
spread and carbon monoxide transportation in a long
channel: Fire Dynamics Simulator comparisons with
measured data, Journal of Hazardous Materials, vol. 140,
pp. 293-298, 2007.
J. Ji, C.G. Fan and W. Zhong, Experimental
investigation on influence of different transverse fire
locations on maximum smoke temperature under the
tunnel ceiling, Int. J. of Heat and Mass Transfer, 2012.
K. McGrattan and A. Hamins, Numerical simulation of
the Howard street tunnel fire, Fire Tech., vol. 42, pp.
273-281, 2006.
Y. Gao-Shang, P. Li-Min, Z. Jing-Hua and A. Yong-Lin,
Simulation of peoples evacuation in tunnel fire, J.
Cent. South. Univ. Tech., vol. 13, iss. 3, pp. 307-312,
2006.
P. Kucera and I. Bradacova, Modelling the evacuation of
people from a train on fire in a railway tunnel, Recent
Advances in Engineering, Proc. of the 3rd European Conf.
of Civil Engineering, Paris, pp. 196-201, 2012.
E. Ronchi, P. Colonna, J. Capote, D. Alvear, N. Berloco
and A. Cuesta, The evaluation of different evacuation
models for assessing road tunnel safety analysis,
Tunelling and Underground Space Technology, vol. 30,
pp. 74-84, 2012.
E. Ronchi, P. Colonna, S.M.V. Gwynne and D.A. Purser,
Representation of the impact of smoke on agent walking
speeds in evacuation models, Fire Technology, vol. 49,
pp. 411-431, 2013.
M.G. Vega and K.M.A. Diaz, Numerical 3D simulation
of longitudinal ventilation system, Memorial tunnel case,
Tunnelling and Underground Space Technology, vol. 23,
pp. 539-551, 2008.
V. Betta and F. Cascetta, Numerical study of the
optimization of pitch angle of an alternative jet fan in a
longitudinal tunnel ventilation system, Tunnelling and
Underground Space Technology, vol. 24, pp. 164-172,
2009.
L. Halada, P. Weisenpacher and J. Glasa, Computer
modelling of automobile fires (Chapter 9, pp. 203-228),
In Advances in Modeling of Fluid Dynamics (LIU, Ch.,
ed.), InTech Publisher, Rijeka, 2012.
K. McGrattan, H. Baum, R. Rehm, W. Mell, R.
McDermott, S. Hostikka and J. Floyd, Fire Dynamics
Simulator (Version 5), Technical Reference Guide, NIST
Special Publication 1018-5, NIST, Gaithersburg,
Maryland, USA, 2010.

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

[29]

[30]

295

K. McGrattan, R. McDermott, S. Hostikka and J. Floyd,

Fire Dynamics Simulator (Version 5), Users Guide,
NIST Special Publication 1019-5, NIST, Gaithersburg,
Maryland, USA, 2010.
K. Hill, J. Dreisbach, F. Joglar, B. Najafi, K. McGrattan,
R. Peacock and A. Hamins, Verification and validation
of selected fire models for nuclear power plant
applications, NUREG 1824, U. S. Nuclear Regulatory
Commission, Washington, DC, 2007.
G.P. Forney, Smokeview (Version 5) - A Tool for
visualizing Fire Dynamics Simulator data, vol. III,
Verification Guide, NIST Special Publication 1017-1C,
U.S. Goverment Printing Office, Washington, p. 52, 2009.
T. Korhonen and S. Hostikka, Fire Dynamics Simulator
with evacuation: FDS+Evac, Technical Reference and
User's Guide, VTT, Finland, 2009.
J. Glasa and L. Halada, On mathematical foundations of
elliptical forest fire spread model (Chapter 12, pp. 315333), In Forest Fires: Detection, Suppresion and
Prevention (E. Gomez and K. Alvarez, eds.), New York:
Nova Science Publishers, Inc., 350 p., 2009.
J. Glasa and L. Halada, A note on mathematical
modelling of elliptical fire propagation, Computing and
Informatics, vol. 30, no. 6, pp. 1303-1319, 2011.
J. Glasa and L. Halada, On elliptical model for forest fire
spread modeling and simulation, Mathematics and
Computers in Simulation, vol. 78, iss. 1, pp. 76-88, 2008.
P. Weisenpacher, P. Polednak, L. Halada, J. Glasa and L.
Valasek, Analysis of course of fire by computer
simulation, Proc. of the Int. Conf. on Fire Safety (M.
Senovsky, ed.), Valtice, 15 p., 2012.
J. Glasa, L. Valasek, P. Weisenpacher and L. Halada,
Use of PyroSim for simulation of cinema fire, Int. J. on
Recent Trends in Engineering and Technology, vol. 7, no.
2, pp. 51-56. ACEEE, New York, USA, 2012.
L. Valasek, The use of PyroSim for creation of the input
FDS geometry for cinema fire simulation, Recent
Advances in Systems Science and Mathematical
Modelling, Proc. of the European Conf. of Systems, Paris,
pp. 304-309, 2012.
P. Weisenpacher, J. Glasa and L. Halada, Parallel
simulation of automobile interior fire and its spread onto
other vehicles, Proc. of the Int. Congress on Fire
Computer Modeling, Santander, pp. 329-338, 2012.
P. Weisenpacher, L. Halada and J. Glasa, Computer
simulation of fire in a tunnel using parallel version of
FDS, Proc. of the 7th Mediterranean Combustion Symp.,
Assoc. Sezione Italiana del Comb. Inst., 11 p., 2011.
P. Weisenpacher, L. Halada, J. Glasa and V. Sipkova,
Parallel model of FDS used for a tunnel fire simulation,
Proc. of the Int. Conf. ParNum 11, Graz: University of
Graz, pp. 96-105, 2011.
P. Weisenpacher, J. Glasa, L. Halada, L. Valasek and M.
Dobrucky, The impact of car park fire on concrete
structure: parallel computation, Proc. of the Conf. on
Applications of Structural Fire Engineering, Prague,
2013.