Optimization of Egress Controls of Fire Emergency Management

Plans using Agent based Simulation: A Case Study of Ready-made

Garment Industry

Abdul Kaium

, Sikder Mohammad Tawhidul Hasan

, Saqib Mehmood

, Shakeel Ahmed

Anders Schmidt Kristensen

and Dewan Ahsan

Danish Centre for Risk and Safety Management, Department of Civil Engineering, Aalborg University, Esbjerg, Denmark

Department of Sociology, Environmental and Business Economics, University of Southern Denmark, Esbjerg, Denmark

Keywords: Readymade Garment, Fire Emergency Management, Fire Risk Analysis, Evacuation Simulation Model,

Agent based Simulation.

Abstract: Ensuring a profound occupational health and safety in Readymade garment industries is a challenge for the

developing countries despite having significant painstaking efforts. Recent statistics show that, the workers

of this sector are in vulnerable conditions due to their exposure to fire risk at the factories and this risk gets

worse during an emergency due to bottlenecks of egress components and clogging. Therefore, the emergency

management plans of the RMG sectors needs to be optimized for safety of workers. The research work of this

paper analyzes and explores the way of addressing underperforming emergency plans by optimizing

evacuation strategies. For this purpose, an RMG factory in Bangladesh is selected as a case study. Based on

the data provided by the factory management and survey regarding occupant’s perception, a simulation model

in an evacuation simulation is built. Fire risks at the factory are analyzed through risk analysis tools for worst

case scenario to assess the existing risks. The results reveal that the application of risk analysis tools along

with an agent-based evacuation can benefit the entire RMG industry by reducing the risks of bottlenecks and

clogging of egress controls that endangers lives of workers, thereby in optimizing the efficacy of fire

emergency management plans.

1 INTRODUCTION

The Apparel industry, also known as Readymade

Garment (RMG), of Bangladesh is making vital

contribution to the emerging economy as well as

acting as the major catalyst of the development of the

country. The readymade garment sector is accounted

for 81 percent of the total export earnings and is the

single biggest export earner for the country with

about four million workers (Hasan, M. and Mahmud,

A., 2017).

Being a developing nation, Bangladesh

government as well as the employers has not paid

much attention to ensure workplace safety in the

RMG industry (Barua, U. and Ansary, M.A., 2017).

As a result, this sector has continuously been facing

tragic incidents one after another. According to the

Centre for Policy Dialogue (CPD), a well-known

research organization working with the occupational

health and safety in Bangladesh, 161 safety incidents

have taken place in the RMG sector resulting 3875

injuries and 1303 deaths from November 2012 to

March 2018 (Solidarity Centre, 2018).

The Tazreen Fashion fire in 2012, one of the tragic

incidents in the Bangladesh apparel industry, in 2012

that caused about 112 dead and more than 200 injuries

followed by the Rana Plaza collapse in 2013 resulting

1134 dead and 2500 injuries awoke the whole country

as well as the world to the poor working conditions in

the Readymade Garment industry of Bangladesh

(Barua, U. and Ansary, M.A., 2017). The government

as well as other stakeholders like the ILO, the Accord,

the Alliance etc. have taken various initiatives to

make the workplace safe.

This research paper has put emphasis on the

implementation of the operational emergency

management principles using risk analysis tools and

simulation of an evacuation model with worst fire

scenario to make proper emergency management

strategies and proper recommendations for the

384

Kaium, A., Hasan, S., Mehmood, S., Ahmed, S., Kristensen, A. and Ahsan, D.

Optimization of Egress Controls of Fire Emergency Management Plans using Agent based Simulation: A Case Study of Ready-made Garment Industry.

DOI: 10.5220/0008117003840391

In Proceedings of the 9th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH 2019), pages 384-391

ISBN: 978-989-758-381-0

selected factory building as well as for the RMG

industry in Bangladesh.

2 OBJECTIVES

The aim of the study is to find out an effective

solution of reducing the exposure of risks to workers

in case of fire evacuation in RMG factories in

Bangladesh. This paper tries to analyze the

anticipated potential fire scenarios based on ISO

16733-1 and how the Pathfinder simulation software

can help assess the risk of fire evacuation and give a

scope to optimize the emergency management plan.

3 METHODOLOGY

An RMG factory is selected for the research purpose.

Then, based on the data provided by the factory

management, the existing fire risks were analyzed

using risk analysis tools.

As a part of data collection, a survey was

conducted to have the occupant’s perception about

current emergency protocol in the factory.

After analyzing the fire risks and considering

previous data regarding the RMG sector, risk ranking

criteria is prepared to make fire scenarios in

accordance with ISO 16733. Later, an event tree

analysis is performed to analyze the uncertainties

followed by a risk profile which presents the

probability of acceptance of risks and consequences.

To mimic the actual evacuation process at the

factory building, Pathfinder, simulation tool, is

incorporated based on management data and the

survey that was conducted by face to face interview.

4 FIRE RISK ANALYSIS

According to the Fire Service and Civil Defense,

Bangladesh (FSCD), 70% of the fire incidents

occurred due to electrical short circuit, 10% for

smoking cigarettes, 8% for machinery spark, 7% for

overheat and 5% for welding spark. Machinery spark

arises due to the friction and lack of proper

maintenance (FSCD, 2019).

Figure 1: Hazards causing fire incidents in Bangladesh

(FSCD, 2019).

4.1 Hazard Identification

As a part of risk assessment, hazards that can cause

fire at the factory have been identified and listed as

follows:

Table 1: Hazard identification.

Hazards

Description

Mitigation

measures

Electrical

cables or

wiring

Electrical short circuit

may happen due to

poor quality cables and

wiring or extra load on

cables.

Good quality of

electrical

materials

should be used,

and regular

maintenance is

required.

Smoking

Smoldering cigarettes

can ignite fire in the

smoking area or at the

building.

‘No Smoking’

sign should be

visible at the

building.

Frictional

mechanica

l parts

Extra heat (overheat)

due to frictions of

machines can ignite

fire

Proper

maintenance of

the machines is

essential for

smooth

production.

Tube light

spark

Spark may arise from

the fluorescent tube

light used everywhere

in the building e.g.

warehouse.

Tube light

should be

checked in

regular basis.

Chemical

ignition

Different types of

chemicals are being

used in the dyeing

section and they can

ignite fire.

Chemical

should be used

with proper

care and MSDS

should be used

strictly for

chemicals.

Yarns,

fabrics

Yarns, fabrics are the

flammable materials

and they can be a good

source for ignition.

Floors should

be neat and

clean and extra

care should be

given to the

sparks and any

other ignition.

70%

10%

Electrical short circuit

Smoking

Welding spark

Overheat

Spark (other types)

Optimization of Egress Controls of Fire Emergency Management Plans using Agent based Simulation: A Case Study of Ready-made

Garment Industry

385

4.2 Risk Matrix

Risk matrix, a useful subjective and semi quantitative

approach to analyze the risks (Aven, 2008), is applied

on the hazard identification for possible fire incidents

at the factory building (Table 1).

Key:

High risk

Moderate risk

Low risk

Negligible risk

Figure 2: Risk ranking matrix for the causes of fire (based

on ISO 16733-1).

In the matrix (Figure 2), the hazards namely

electrical cables and tube light spark are placed in the

high-risk zone which is marked by blue color. These

hazards are serious threats to the fire safety of the

building and need immediate attention for mitigation

to avoid serious consequences. Smoking and fabrics

or yarns are placed in the moderate risk zone and they

should be minimized in a short period of time and top

management should be informed about these hazards.

And finally, chemicals and machines friction can be

mitigated through regular checkup.

5 EXISTING EMERGENCY

MANAGEMENT

Emergency management has four cycle namely

prevention and mitigation, preparedness, response,

and recovery (George D. Haddow; Jane A. Bullock;

Damon P. Coppola, 2006). This section briefly

describes these four cycles in the context of fire

related laws and rules.

Table 2: Emergency management at the factory.

Phase

Existing measures

Prevention

and

Mitigation

Smoke detectors, sprinkler system etc.

have been installed as per BNBC. If

there are 500 occupants, there should be

at least two staircases in the building as

per the BNBC. In our factory, they have

4 staircases.

Preparedness

The authority has prepared themselves

with in house fire team in each floor

who will try to put out fire immediately,

rescue team who will be assisting

mainly disabled or pregnant women or

who get stuck in evacuation, first aiders

who will provide first aid to the injured

ones, fire extinguishers, hose pipe, hose

reel etc.

Response

In-house firefighter will be the first

responder and the fire brigade or FSCD

will be the second responder.

Recovery

Mainly the fire brigade, in-house fire

fighters, in-house rescue team will be in

the recovery stage if there is any fire

incident. Workers and local people also

help in response stage. Local hospitals

play vital role in case of casualties in

emergency.

6 DESIGN FIRE SCENARIOS

6.1 Fire Safety Objectives

The main objective for the fire safety is to provide life

safety to all occupants in the building. The fire safety

management should be such a way that no occupant

will be exposed to the fire. Occupants must be able to

evacuate the building before fire reach critical

condition. Hence, the fire safety objectives are as

follow:

 Life safety of occupants in the building

 Reduce the property damage

 Reduce environmental impact

 Continuity of factory operations

6.2 Selection Process of Fire Scenarios

ISO has guidelines regarding the selection process of

fire scenarios for the evacuation simulation. Based on

ISO 16733-1, the following guidelines have been

discussed to select fire scenario (ISO, 2015).

SIMULTECH 2019 - 9th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

386

Table 3: Selection process of fire scenarios.

Location of

fire

In case of readymade garments factory

of Bangladesh, ware house is the most

potential fire location (Khandoker,

M.A.R., Mou, R.J., Muntaha, M.A. and

Rahman, M.A., 2018)

Type of fire

Fire involving solid materials like

fabrics, clothes, hard board etc. as well

as chemicals.

Potential

fire hazards

As discussed in Table 1.

Systems

affecting

fire

Automatic fire detection and alarm

system, manual alarm, automatic

suppression system (sprinkler), and in-

house firefighters.

Occupant’s

response

Familiar, wake and used to go with the

fire drills. But, there may be new

occupants, visitors who are not familiar

with the building as well as fire safety

evacuation.

Event Tree

Deterministic approach has been made to analyze the

uncertainties affecting any fire event. The event tree

analyzes all the potential outcomes based on the

initiating event- starting of a fire at the building. In

the event tree, all the existing barriers in the building

have been utilized to control the fire incident.

The event tree (Figure 3) has been made based on

the data provided by the factory management, but it

was not possible to check the correctness of the

information used. If any fire incident takes place,

there is automatic detection and alarm system

installed at the factory and the probability of working

this system is 85%. If the automatic system fails, there

is 80% probability that this manual alarm works.

Sprinklers are being installed and there is 70%

probability that they work in case of fire. There are

in-house fire team working in every floor. They can

suppress fire initially and probability is 60%. The

factory management observed during the fire drill that

80% of occupants behave correctly during fire.

In the following event tree (Figure 3), the

consequences have been made based on the

assumptions and subjective judgment. The number of

consequences is the number of occupants who may be

exposed to the critical condition and does not

necessarily mean casualties or fatalities. For example,

in case of fire, if every barrier works and occupants

behave correctly, then there will be no one to be

exposed to the fire. Sometimes people get panicked

just hearing the alarm sound. That is why, there are

some consequences in case of fire suppressed in the

event tree.

In the following event tree (Figure 3), all the

possible fire scenarios and impacts have been

calculated.

Figure 3: Event tree of fire at the factory building.

6.3 Risk Profile

Risk is defined as the multiplication of the likelihood

of an event and the severity of consequences (Aven,

2008). Risk profile has been made from the event tree

after calculating the probability and consequence of

each scenario. The consequences have been sorted in

increasing order with corresponding probability and

then cumulative probability has been calculated.

Using these impact and cumulative value, the risk

profile or staircase function has been drawn and it

ranges from probability 1 to 0.

Figure 4: Risk profile curve of the event tree.

Here, the risk profile (Figure 4) shows that the

individual risk is 0.52 that means if there is a fire at

the building, the estimated probability that one or

more occupants are exposed to critical condition, is

0.52. In other words, there is 52 percent chance that

Optimization of Egress Controls of Fire Emergency Management Plans using Agent based Simulation: A Case Study of Ready-made

Garment Industry

387

one or more occupants will be exposed to critical

condition if there is a fire at the building.

To calculate the medium risk, probability and

consequences of each scenario in the event tree has

been multiplied and then summed. The medium risk

is 1.15 that means, if there is a fire, there will be an

average of 1.15 person exposed to critical conditions.

The acceptance curve shows that the management

should pay heed to controlling the high probability

with zero consequences and high consequences with

lower probability zones.

6.4 Fire Scenarios

Fire can be ignited from the electrical short circuit or

fluorescent tube light sparks in the warehouse where

all the finished goods, yarns etc. are being stored.

According to statistics, most of the fire incidents in

the apparel industry of Bangladesh starts from the

warehouses. As it was mentioned above that, 70% of

the fire incidents caused from electrical short circuit

(See Figure 1), that is why likelihood of this fire

scenario has been set to high. In this case, the fire is

ignited in the warehouse during the daytime. In

daylight hours, 883 occupants are in the building. The

warehouse is in the fifth floor with 2359.55 square

meter area and four exits leading to stairs. If fire

ignites from any corner of this room, adjacent rooms

in the same floor will be affected immediately. The

fire and smoke will be spread out to the sixth and

seventh floor shortly as the smoke is upper bound in

nature. There could be a lot of fire scenarios, hence

only this scenario is further discussed as worst case.

7 SIMULATION MODELING

After the analysis of fire risks at the factory and the

required information from the survey, evacuation

model has been built. The simulation model is built

based on the actual drawing and floor layout of the

factory building provided by the management. Then

the blueprint is produced in .dwg file and further

exported to .pth file for extracting floors and doors,

stairs etc. in Pathfinder.

In simulation model, factory building

characteristics, occupant’s factors and designing of

fire scenario have been considered. The building is

eight-storied having span of 37161.94 square feet or

3452.46 square meters occupying 883 occupants in

total. It has different floors having different layouts

comprise knitting, linking, labeling section along

with large warehouses. There are egress components

such as stairs, lifts by which occupants evacuate

themselves effectively as well as proper ventilation

system which takes out the heat and smoke from the

building and acts as one of the passive fire protection

systems.

Occupant factors has different components such

as the number of occupants, age, gender, height and

shoulder width, and behaviors of the occupant during

fire evacuation. Figure 5 and Figure 6 depict the

occupants’ profile, behavior, walking speed,

waypoint etc. in the simulation model.

Figure 5: Occupants profile, behavior and walking speed

(1st to 4th floor).

Figure 6: Occupants profile, behavior and walking speed

(5th to 7th floor).

8 RESULTS AND VALIDATION

8.1 Simulation Results and Analysis.

Table 4 shows that the maximum time in steering

mode is around 343.40 seconds while 405.20 seconds

in the SFPE mode.

SIMULTECH 2019 - 9th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

388

Table 4: Results from the simulation model.

Steering

SFPE

Min

34.7

34.2

Max

343.4

405.2

Average

169.7

176.2

StdDev

76.5

91.3

The figure 7 shows that the time of exit for all

occupants is 343.40 seconds. The time is taken by

first occupant to exit is around 34.70 seconds is

shown in horizontal part of the graph just before

sliding down. The steady state of the first occupant

from the initial point is due to pre-movement time.

After that it slides down at a constant state which

means occupants evacuate themselves without any

obstacles thus does not create any density in the end.

Figure 7: Time of exit for all occupants at the building.

Figure 8 illustrates that, fire breaks out in the

spot at the warehouse in 5th floor. The two red circles

indicate that two exits on the right side of the

warehouse are considered to be blocked due to

smoke, gas and heat. Thus, all the occupants seek

another waypoint which are marked red straight line

and seek to safe exits marked yellow circle.

Occupant’s speed in this floor those who work closest

to fire initiating point is affected. Fire fighters stay

there to extinguish fire as first responders if needed

having the sprinkler system works at the same time.

Figure 8: OCCupant’s Waypoint Due to Fire in 5th Floor.

Occupants of the two floors which are just above

5th floor do not choose two exits which are blocked

in 5th floor due to fire and smoke because of the

panic. The panic due to fire lead occupants to choose

the same way points as occupants in the 5th floor seek

to leave through safe exit on the other side of the

floor.

Waypoints are set up for the safe evacuation

purpose (Figure 8). In this kind of situation,

occupants might be advised to follow the safe way

points for safe exit.

The red marks in Figure 9 indicate that in 2

floor,

there is bottleneck at the exit. This congestion leads

to overcrowd at the exit and thus creates panic which

might turn out to be stampede instead of safe

evacuation.

Figure 9: Density Heat Map in 2nd Floor.

8.2 Validation with Fire Drill Data

Pathfinder simulation tool simulates the evacuation

model, verifies the model and validates it with actual

fire drill data in terms of total evacuation time

(Ahmed, S., Mehmood, S. and Kristensen, A.S.,

2019).

The evacuation time from the fire drill is 295

seconds (data was taken from the management)

whereas the total evacuation time in Pathfinder

simulation (steering mode) is 326.6 seconds. There is

a bit difference between two data, because most of the

fire drills in most readymade garment factories are

announced beforehand that is why there is no pre-

movement time.

8.3 Validation with Fundamental

Diagram

The fundamental diagram test represents the speed-

density profile of the occupants in the model. This test

validates the following function:

  

(1)

Where, speed is a function of density.

For this validation purpose, the 2

floor of the

factory is chosen to examine how the density affects

the occupant’s speed which reflects the real

evacuation scenario. Figure 10&11 represent speed-

Optimization of Egress Controls of Fire Emergency Management Plans using Agent based Simulation: A Case Study of Ready-made

Garment Industry

389

density and specific flow-density, and data are

represented over time intervals until the steady state

arrives.

The simulation is run on steering mode. In

steering mode, there is no boundary layers in doors

thus does not have any specific flow rate. Here in

steering mode, each occupant uses steering system

and keep a reasonable distance from others to avoid

any kind of obstruction. Each occupant has own

specific goal and acts as independent agent.

Occupants interact with other occupants.

Figure 10 demonstrates that there is a negative

relationship between speed and density so that speed

of the occupants is reduced when there is high

density.

Figure 10: Speed-Density profile.

Figure 11 shows that maximum specific flow is

obtained where there is density of 2 person/m

. It gets

lower when the density is high.

Figure 11: Specific Flow-Density profile.

8.4 Verification with Hand Calculation

The following hand calculation is based on equations

in accordance with Engineering Guide to Human

Behavior in Fire (Galea, 2003) and considering

geometry of the factory building:

If T

= the time it takes first occupant to reach

controlling component, T

= the time it takes to 883

occupants to exit through the controlling components

and T

= the time it takes the last occupants move from

the controlling to component to exit. Then,

  



 



 



(2)

















(3)

















  



(4)

v= 



  

(5)

Where d, v, p, F

, w and BL means density (1.88

pers/m

), velocity, number of occupants (883),

maximum specific flow (1.32 person/s-m), actual

width of door (32 inch), boundary layer (6 inch)

respectively, and k

(1.40, 1.08 for corridors and stairs

respectively) and a (0.266) are constant.

The total evacuation time by hand calculation is

423.22 seconds. The evacuation time from simulation

is 405.20 seconds without pre-movement time closest

to 423.22 seconds. This difference is acceptable

according to simulation developers (Thunderhead

engineering, 2018).

9 DISCUSSION

This study shows that fire can be broken out due to

hazards exist in the factory. Despite having fire drills

in the factory, it lacks the real evacuation scenario

since fire drill is pre-determined in most of the cases

and it does not mimic the existence of fire.

The simulation result shows that there could be lot

more improvement in terms of further optimization of

the emergency management system. For instance, in

case of fire, occupants in the warehouse could be

advised to choose other exits for safe evacuation.

Simulation result clearly shows the option for safe

evacuation when other exits are blocked due to fire

and smoke.

The result also shows that there could be clogging

in any of the exits. In this case, occupants could be

advised to go to other exits when they see that the

exits they try to get through are highly dense and thus

prevent stampede due to panic.

10 CONCLUSION AND

RECOMMENDATIONS

Evacuation for a hundred percent export-oriented

factory is a difficult task to carry out in emergency.

Fire in the factory possesses high risk and has an

impact on human lives and properties. As an integral

part of the emergency management approach, the fire

SIMULTECH 2019 - 9th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

390

drill carried out at the factory which does not reflect

the true evacuation process. In the drill, fire and

smoke are not created and most importantly, exits are

not blocked during the fire drill at the factory. But

exits near the fire will be blocked and halt the

occupants from safe egress. In Pathfinder simulation,

in case of fire in warehouse, exits were blocked near

fire location and occupants were assigned to use

specific exit route through waypoint to safe exit.

Pathfinder provides some extent of behavior based on

parametric input from the data from survey. The

simulation gives the Required Safe Egress Time and

identifies hazardous area in terms of congestion in

exits of the floors due to high density which helps

assessing the risk. One important drawback is that,

Pathfinder cannot predict the real behavior of

occupants in case of emergency. Behavior of

occupants varies largely based on the experience,

training, safety culture etc. However, the Pathfinder

simulation largely opens the scope of optimizing the

current emergency management approach in terms of

adopting different strategy in emergency

preparedness, response, mitigation and recovery. To

optimize the emergency strategy, following measures

can be taken:

Relocating the warehouse to some other places

would be the optimal choice for the management of

the factory to avoid huge economic loss and loss of

lives.

The findings from simulation like clogging,

behavior etc. should be taken into consideration

during fire drill in mitigation and prevention phase of

emergency management.

Quantitative risk analysis tools may be used to

assess the fire risks.

The number of occupants in 2nd and 3rd floor

should be reduced.

Authority should engage the local people,

hospitals and law enforcement agencies in the

response and recovery stage.

There should be an incident command system at

the factory for handling any emergency at any time

(day or night).

This research using risk analysis tools and

Pathfinder simulation software can help the entire

Readymade garment industries along with other

industries and any building structures to assess the

risk and to optimize the emergency management in

case of fire evacuation.

ACKNOWLEDGEMENTS

The authors would like to express their gratitude to

Ms Annette Kaalund-Jørgensen, Capacity

Development Advisor, Danida Fellowship Centre for

her kind support.

REFERENCES

Ahmed, S., Mehmood, S. and Kristensen, A.S., 2019.

Critical Analysis of Computer-based Evacuation

Simulation of an Educational Institution. s.l., in

Proceedings of the 11th International Conference On

Computer Modeling and Simulation (pp. 26-30). ACM.

Aven, T., 2008. Risk analysis: Assessing uncertainties

beyond expected values and probabilities. Chichester,

UKT: Wiley.

Barua, U. and Ansary, M.A., 2017. Workplace safety in

Bangladesh ready-made garment sector: 3 years after

the Rana Plaza collapse. International Journal of

Occupational Safety and Ergonomics, 23(4), pp. 578-

583.

FSCD, 2019. Accident, Rescue and Yearly Statistics.

[Online] Available at: http://www.fireservice.gov.bd/

site/files/e4d115ed-d51b-4bfc-b941-2092c1a5c916/

Accident,-Rescue-&-yearly-Statistic [Accessed 12 My

2019].

Galea, E., 2003. SFPE Engineering Guide to Human

Behavior in Fire. s.l.:s.n.

George D. Haddow; Jane A. Bullock; Damon P. Coppola,

2006. Introduction to Emergency Management. 5th ed.

225 Wyman Street, Waltham, MA 02451, USA:

Elsevier.

Hasan, M. and Mahmud, A., 2017. Risks Management of

Ready-Made Garments Industry in Bangladesh.

International Research Journal of Business Studies,

10(1), pp. 1-13.

ISO, 2015. Fire safety engineering -- Selection of design

fire scenarios and design fires -- Part 1: Selection of

design fire scenarios. [Online] Available at:

https://www.iso.org/standard/63140.html [Accessed 12

May 2019].

Khandoker, M.A.R., Mou, R.J., Muntaha, M.A. and

Rahman, M.A., 2018. Numerical simulation of fire in a

multistoried ready-made garments factory using

PyroSim. s.l., In AIP Conference Proceedings (Vol.

1980, No. 1, p. 050026). AIP Publishing..

Solidarity Centre, 2018. Bangladesh Safety Incident Chart.

[Online] Available at: https://

www.solidaritycenter.org/wp-content/uploads/2018/

04/Bangladesh-Safety-Incident-Chart.4.9.18.pdf

[Accessed 12 May 2019].

Thunderhead engineering, 2018. Pathfinder user manual.

s.l.:403 Poyntz Avenue, Suite B, Manhatten, KS 66502

USA.

Optimization of Egress Controls of Fire Emergency Management Plans using Agent based Simulation: A Case Study of Ready-made

Garment Industry

391