625

1 INTRODUCTION

Until this time water transportation is still very liked

by Indonesian citizen because it is more efficient and

also water transportation is free from traffic jam. But

the number of ship accidents that occurred caused

some people became a doubt to make a ship as a man

transportation in determining the preferred mode of

transportation for cross-island travel.

Even based on the results of the investigation data

from The National Transportation Safety Committee

(KNKT) in 2021 water transportation mode became the

biggest contributor with most of the fatalities are in

accident. KNKT investigated as many as 19 cases, with

a total death toll of dead and missing reaching 342

souls. From 19 cases investigated by KNKT there is

several prominent accidents, one of them is from KMP.

Yunicee who is passenger on Ro-ro ship Ketapang-

Gilimanuk who sank in Bali strait on June 29th, 2021 at

midnight.

When the ship has an emergency condition or

which allows dangerous conditions, passenger

evacuation became the first thing that must to do to

prevent the occurrence of many casualties, both death

and missing. But unfortunately, poor evacuation

planning in the ship became one of the factor that can

cause many victims/toll who not safe when the ship

accident happened.

Analysis of Inflatable Liferaft Layout Effectiveness

Towards The Evacuation Process for Passenger Ships

Based on IMO MSC.1/Circ. 1533

I.P. Mulyatno, H. Yudo, S.A.Prasanti & W. Amiruddin

Diponegoro University, Kota Semarang, Jawa, Indonesia

ABSTRACT: The inflatable liferaft layout applied to passenger ships for the effectiveness of the evacuation

process must be based on IMO MSC.1/Circ. 1533 regulations with the maximum evacuation duration is n ≤ 60

minutes. Based on the data from KNKT, in 2021, sea transportation became the biggest contributor to accidents

with 342 people dying and missing. Liferaft is one of the main safety tool used during an emergency to save the

people and leave the ship. This study used Thunderhead Pathfinder software which was Agent Based Evacuation

Simulation combined with 3-D simulation results. The modeling was conducted with two types of layout liferaft

and two scenarios of dangerous conditions, the first was a fire in the engine room and the second is the ship

experiencing a 20° of heel. The results of this study indicate that there was a difference in the total evacuation

duration between the existing layout and the layout that has been changed according to the writer's suggestion.

In fire conditions there is a difference of 1 minute 18 seconds in case 1, 1 minute 16 seconds in case 2, 1 minute 38

seconds in case 3, and 22 seconds in case 4. In the heel condition there is a difference of 1 minute 19 seconds in

case 1 and 1 minute 25 seconds in case 2.

The results of the evacuation simulation modeling with the liferaft layout on the navigation deck that have been

modified according to the writer’s suggestion in all cases are getting a value of n ≤ 60 minutes and also have

complied with the IMO MSC.1/Circ. 1533 regulations.

http://www.transnav.eu

the International Journal

on Marine Navigation

and Safety of Sea Transportation

Volume 16

Number 4

December 2022

DOI: 10.12716/1001.16.04.03

626

SOLAS conference in 1995 it has been established

that all evacuation procedures on Ro-ro passenger

ships must be completed in 60 minutes. Time

effectivity that used during evacuation process

depending on the number of passengers and the

distance traveled to embarkation corridor. The attitude

and reaction that passengers show when they hear the

dangerous alarm will be different, these things are

depend on the background experience and the

knowledge of the passenger in dealing with the

situation under pressure when are in the group.

Human behavior and walking speed under list and

dynamic condition of ships are very important factors

in analysis for Ro-ro passenger and large passenger

ships [1].

The Maritime Safety Committee, at its seventy-first

session (19 to 28 May1999), noted that under SOLAS II-

2/28-1.3, Ro-ro passenger ships constructed on or after

1 July 1999 are required to undergo an evacuation

analysis at an early stage of design [2]. Regarding to

SOLAS regulation II-2/13 about the provision of

evacuation routes so that passenger can quickly and

safely get to the meeting point, the ship must follow the

following conditions:

− Safe escape routes shall be provided;

− Scape routes shall be maintained in a safe condition,

clear of obstacles; and

− Additional aids for escape shall be provide as

necessary to ensure accessibility, clear marking, and

adequate design for emergency situations [3].

In this case the evacuation route will direct

passengers to a place where ship safety equipment is

provided.

Liferaft is one of the man safety equipment that

used to emergency situation to save themselves and

leave the ship. As specified in MSC/Res. 809, all

inflatable liferaft must fulfill the requirements of

paragraph 4.2, namely: The lowering speed for a fully

equipped fast rescue boat with its full complement of

persons on board should not exceed 1 m/s. [4]. So the

placement of the liferaft must be in the right position

so that it can be effective when an emergency occurs.

This is because the layout of the liferaft will greatly

affect the length of time it takes necessary for the

evacuation of passengers. Therefore, the author will

conduct further research on the effect of liferaft layout

on the effectiveness of the passenger ship evacuation

process. This research was conducted by simulating

using Thunderhead Pathfinder software which is an

Agent Based Evacuation Simulation combined with

Simulation result in the form of 3-Dimensional

animation.

2 METHOD

2.1 Research Object

This research acquired the data from crossing ship 600

gross tone (GT) belonging to Directorate General of

Land Transportation that operate in track Singkil –

Banyak Island. It was the types of Ro-Ro crossing ship

with IMO 9926817 and the BKI Regulation Number 191

213 0043. It was constructed in 2019 in PT Citra Bahari

Shipyard.

Table 1. Principal Dimensions KMP. Aceh Hebat 3

________________________________________________

NO Name Measure Unit

________________________________________________

1. Length Overall (LOA) 54.50 Meter

2. Length of Perpendiculars (LPP) 47.25 Meter

3. Weight (B) 13.00 Meter

4. Heght (H) 3.45 Meter

5. Draft (T) 2.45 Meter

6. Speed ( Trial Speed) 2.45 Meter

7. Passenger 212 Person

8. Crew 24 Person

________________________________________________

With a lot of the capacity, this crossing ship could

accommodate 15 trucks and 6 sedans. On this crossing

ship detents 14 liferafts that were each of it could

accommodate 25 people.

2.2 Regulation

International Maritime Organization (IMO) published

the standard term related the ship’s passenger

evacuation, as referred it was provisions of Safety Of

Life At Sea (SOLAS) that related with the ship’s safety

and the total of lifebuoy with all the characteristic.

Thereafter on 2016 International Maritime

Organization (IMO) published MSC.1/Circ. 1533 that

contained about “Revised Guidelines on Evacuation

Analysis for New and Existing Passenger Ships”. The

calculation of the performance standards of the total

maximum duration evacuation that must be compiled

with is:

( ) ( )

1,25

R T E L n+ + + 

(1)

( )

30 E L min+

(2)

In performance standars:

For ro-ro passenger ship, n = 60 minute; and for

passenger ships other than ro-ro passenger ship, n = 60

if the ship has no more than three main vertical zones;

and 80, if the ship has more than three main vertical

zone [5].

Figure 1. Performance Standards

1. Response duration (R) = 10 minutes for night case

and 5 minutes for day case.

2. Total travel duration (T) = duration it takes for all

persons on board to move from where they are

upon notification to the assembly stations.

3. Embarkation and launching duration (E+L) =

maximum 30 minutes with the regulation Safety Of

Life At Sea (SOLAS) III/21.1.3.

627

4. Overlap duration = 1/3 (E+L).

5. n value (minutes).

Based on the MSC.1/Circ.1533 this provided the

parameter to facilitate evacuation simulation, that is

related the population categorize which is explained

the composition of the population in term of age,

gender, physical attribute, and response duration. The

population consist of the following combination:

Table 2. Populations Composition

________________________________________________

Population Group Percentage Amount of

Passengers of Passengers (%) Passengers

________________________________________________

Female < 30 years 7 15

Female 30 – 50 years 7 15

Female > 50 years old 16 34

Female > 50 years,

Mobility Impaired (1) 10 21

Female > 50 years,

Mobility Impaired (2) 10 21

Male < 30 years 7 15

Male 30 – 50 years 7 15

Male > 50 years 16 34

Male > 50 years, M1 10 21

Male > 50 years, M2 10 21

________________________________________________

Population Group Percentage Amount of

Crew of Crew (%) Crew

________________________________________________

Crew Females 50 12

Crew Males 50 12

________________________________________________

Total 236

________________________________________________

For the purpose of conducting an evacuation

analysis, the initial distribution of passengers and crew

on board should be considered. In this study, the

researcher uses IMO MSC.1/Circ. 1533 guidelines for

the following case:

1. Case 1 (primary evacuation case, night)

Passengers in cabin with maximum berthing

capacity occupied; 2/3 of crew members in their

cabins; of the remaining 1/3 of crew members:

1) 50% are in their respective workplaces.

2) 50% are spread over each deck.

2. Case 2 (primary evacuation case, day)

Public spaces, as defined by SOLAS regulation II-

2/3.39, will be occupied to 75% of maximum

capacity of the spaces by passengers. Crew will

distributed as follows:

1) 1/3 of the crew are in cabin.

2) 1/3 of the crew are in public spaces.

3) the other 1/3 are in their respective workplaces.

3. Case 3 and 4 (secondary evacuation, night and day)

These cases use the same population demography

as the primary evacuation case with the difference

that one stair on a ship that has a large capacity for

passengers to pass during an evacuation is

considered unusable in this case simulation.

In this study, case 3 and 4 will only be used in a fire

scenario.

For each of the gender group specified in table 2,

walking speed must be modeled as a statistical

distribution which has minimum and maximum

values, as follow:

Table 3. Walking Speed on Flat Terrain

________________________________________________

Population Group Walking Speed

Passengers Min (m/s) Max (m/s)

________________________________________________

Female < 30 years 0.93 1.55

Female 30 – 50 years 0.71 1.19

Female > 50 years 0.56 0.94

Female > 50 years, M1 0.43 0.71

Female > 50 years, M2 0.37 0.61

Male < 30 years 1.11 1.85

Male 30 – 50 years 0.97 1.62

Male > 50 years 0.84 1.4

Male > 50 years, M1 0.64 1.04

Male > 50 years, M2 0.55 0.91

________________________________________________

Population Group Walking Speed

Crew Min (m/s) Max (m/s)

________________________________________________

Crew Females 0.93 1.55

Crew Males 1.11 1.85

________________________________________________

The walking speed on stairs were given by the

category of gender, age, and direction of travel up

which has the values as follow:

Table 4. Walking Speed on Stairs

________________________________________________

Group of The Passenger Walking Speed on Stairs

Population Min (m/s) Max (m/s)

________________________________________________

Female < 30 years 0.47 0.79

Female 30 – 50 years 0.44 0.74

Female > 50 years 0.37 0.61

Female > 50 years, M1 0.28 0.46

Female > 50 years, M2 0.23 0.39

Male < 30 years 0.50 0.84

Male 30 – 50 years 0.47 0.79

Male > 50 years 0.38 0.64

Male > 50 years, M1 0.29 0.49

Male > 50 years, M2 0.25 0.41

________________________________________________

Group of The Crew Walking Speed on Stairs

Population Min (m/s) Max (m/s)

________________________________________________

Crew Females 0.47 0.79

Crew Males 0.50 0.84

________________________________________________

Previous research conducted by Trika Pitana et al.,

showed that the differenciate in total evacuaton time

between walking speed data in IMO and research is not

too significant, it could mean that the data is relevant

IMO if applied to the case of evacuation in Indonesia

[6].

2.3 Accident Scenario

This study was conducted with two dangerous

conditions that triggered the evacuation. The

conditions are a fire in the engine room caused by a

leak in the fuel pipe and the ship experiencing a 20

heel. This scenario is defined based on primary cases

and secondary cases according to IMO

MSC.1/Circ.1533.

2.4 Data Processing

From the data that has been obtained, several

evacuation simulations will be carried out by moving

the position of the liferaft to determine the placement

of the liferaft where the evacuation simulation process

shows the most effective results. The following are the

steps taken in processing the data:

1. Reading the General Arrangement of KMP Aceh

Hebat 3 to find out the placement of liferaft.

628

2. Redrawing the General Arrangement in

Thunderhead Pathfinder software.

3. Conduct an evacuation simulation using a liferaft

layout according to the real design.

4. Conduct an evacuation simulation by moving the

liferaft according to the researcher’s suggestion.

5. Calculate the total duration of evacuation in each

simulated case.

6. Determine the most optimal liferaft layout for the

passenger evacuation process.

3 RESULT AND DISCUSSION

3.1 Evacuation Simulation Modeling

The simulation modeling in this research was

conducted to analyze the effectiveness of the inflatable

liferaft layout on the evacuation process on KMP Aceh

Hebat 3. Where the goal is to get liferaft placement

position that will produce the most effective total

duration of evacuation. This research was conducted

by doing simulation using Thunderhead Pathfinder

which is Agent Based Evacuation combined with 3-D

simulation results. Simulation will be done 2 times,

namely the condition of the existing ship (real design)

and when the liferaft placement has been changed

according to the author’s suggestion.

The model is very influential in calculating the

evacuation time. This is about the characteristics of the

model that can represent real conditions in the field.

This simulation is used to calculate the total travel

duration (T) in the evacuation process. Modeling refers

to the general arrangement of KMP Aceh Hebat 3.

The following are the stages of modeling in the

Thunderhead Pathfinder software:

1. Import file general arrangement of the ship to

Pathfinder through Menu “Model” on the toolbar

then click “Add a Background Image” on the dialog

box that appears.

2. Redrawing every room that exist on the ship. In the

accordance with the general arrangement. In the

redrawing of the room, it is also equipped with the

provision of door access, so that the passenger can

get out of the room to a predetermined evacuation

point.

3. Combining all the decks and making the stairs that

later will be connecting each deck and the crew.

Every deck will be arranged upwards according to

the order and height corresponding to the general

arrangement.

4. Adding agents representing passengers and crew as

objects of the evacuation simulation process. The

agents to be added have their respective

characteristics according to predetermined

parameters such as walking speed on flat terrain

and walking speed on stairs as presented in table 3

and table 4.

5. Furthermore, after all data is entered, the

distribution of the agent is adjusted according to the

rules of IMO MSC.1/Circ.1533 as presented in table

6. The next step is to run a simulation of the

evacuation process modeling. “Run Simulation”

must be pressed to be able to rn this simulation

process. Then a dialog box will automatically

appear regarding the evacuation simulation

process.

7. After the process is running and all agents have

been evacuated, the process will automatically stop

and display a video about the total travel duration

(T) running time.

Figure 2. Evacuation Simulation Process

Based on the picture above, it can be seen that in the

video of the evacuation results will be presented some

information related to the number of passengers who

reach the muster point in unit time. Each case will have

a different result according to the characteristics of the

simulated case. In each simulated case, the results will

displayed in graphical form.

Figure 3. Evacuation Simulation Results Graph

3.2 Evacuation Simulation

As explained above, the liferaft laying scenario for this

research conducted with two kinds of layout liferaft.

The first layout of the entire liferaft is on the front

navigation deck according to the existing conditions on

the ship.

Figure 4. Layout 1 (Existing Condition)

While the second layout, liferaft is deployed on the

navigation deck in order to get the right evacuation

time. The transfer of the liferaft namely, 2 liferafts are

moved to the back closer to the exit stairs one each on

the starboard and portside, 2 liferafts moved to the

center closer to the passenger area one each on the

starboard and portside, then 2 liferafts remain at the

629

front of the navigation deck one each on the skateboard

and portside.

Beside aiming to get a faster evacuation time

compared to the first layout, the second layout also

aiming to prevent queues or accumulation of

passengers at one point on master point.

Figure 5. Layout 2 (Liferaft Layout Changes)

3.2.1 Fire Scenario

This modeling is carried out to calculate the

evacuation time in the event the ship experiences a fire

in the engine room. In this research the source of the

fire trigger is the occurrence of leaks in the fuel pipe.

Based on the research about the danger of fire that

coordinated by International Maritime Organization

(IMO) identified the main source of the fire trigger in

the engine room. In this research also present

calculation of frequency of fires caused by self-ignition

of flammable liquids in fuel oil and diesel oil systems,

which constitute 60% of the overall hazard [7].

Figure 6. Fire Triangle

On this case when the fuel molecules evaporate on

an open surface, the steam will mix with the

surrounding air and the heat from the main engine will

produce a diffusion air. Then, assumed that fire will

grab the pvc pipe on the main engine which then

causes a fire.

The fire modeling is done by combining between

software Thunderhead Pathfinder and Pyrosim

software to cause smoke effect in fire simulation, with

the following process:

1. Import the ship’s general arrangement autocad file

to the Pyrosim software. Make sure that the design

in autocad has been made with a scale of 1:1 so that

it can depict on a real scale with real conditions.

2. Then, the picture import result as in the form of 2

dimensions will be converted into 3 dimensions by

extruding the ship design line from autocad into a

wall with an adjusted thickness to approach the real

conditions.

3. Then, details of the composition of the materials

used on the ship are carried out. This is intended to

be able to resemble the real conditions that exist in

the field.

4. Models are converted into customized materials,

the model is added to a fire source to simulate a fire.

As explained above, the fire scenario in this study is

assumed that the initial material burned is pvc pipe

with Heat Release Per Area (HARRPUA) value is50

kW⁄m

. The heat exposure are considered as the

advisable value to be used in the methodology [8].

Assuming the width of the burned area is 2,4 m

5. The next step is to determine the reaction used for

the fire source during the fire simulation. The

reaction used is a Polyurethane reaction using the

standard “SPFE Handbook, GM 27” with the

following composition details:

1) Carbon atoms 1,0

2) Hydrogen Atoms 1,7

3) Oxigen Atoms 0,7

4) Nitrogen Atoms 0,08

6. The next step was mesh making. This serves to limit

the fire affected area in this model. In this study the

restricted mesh was one full ship with different

open ventilation on board so that the smoke

produced was not trapped within the specified

mesh limit.

7. The last step was simulate the modeling by pressing

“Run Simulation” on the toolbar.

8. After the process was complete, then it will

automatically stop and display a video of fire

simulation results.

Figure 7. Modeling That Has Been Adapted to The Type of

Material

Figure 8. The Place of Fire Source

Figure 9. The Result of Fire Simulation Process

After making the model on Pyrosim software

complete, then the next step was to combine the model

with the previous model that was made in the

Thunderhead Pathfinder software by importing the

630

Pyrosim model into the Pathfinder file. Then both

models will be adjusted to become a single unit.

Figure 10. The Result of Import Pyrosim Model into

Pathfinder

Figure 11. The Result of Merged Model

The picture above was a final picture of evacuation

modeling on fire case. The model was ready to simulate

and would produce total score travel duration (T) in

the simulated case.

The result total travel duration (T) that gain from

the simulation above as follow:

Table 5. The Result of Total Travel Duration on Fire Case

________________________________________________

The total Travel Duration (T) (Second)

Layout 1 Layout 2

________________________________________________

case 1 447.5 385.0

case 2 392.8 332.3

case 3 742.5 663.8

case 4 619.3 601.5

________________________________________________

3.2.2 Heel Scenario

The evacuation simulation modeling in the case of

a ship in a tilted or heeled state was not much different

from previous case. Agent Basic Model Simulation

(ABMS) modeling has limitation namely it can’t

manipulate the speed according to the dynamic tilt

angle of the ship. This is due to the modeling that

cannot fully represent every degree the ship will

overturn, but can represent the tilt of the ship at a

certain angle that was relevant to use.

This modeling was modeled with the difference in

movement speed and the changing muster point. For

the heel case, muster point that used was one from the

starboard or the portside only. It is because when the

ship is tilted, which can be used for the process of

evacuating passenger and crew, the liferaft that used

was on the lowest side of the ship. Because if the liferaft

is launched on the higher side it is possible that the

liferaft will hit the wall of the ship that is not in its

normal position or even keel.

Evacuation simulation modeling in this case was

carried out in a 20° heel conditions. For walking speed

data in heel conditions was taken from previous

research conducted by Refan Trisna Wijaya in 2016.

The researcher did the experiments with sampling

from drilling test on KM Gunung Dempo owned by PT

Pelni (Persero). From the experimental research

obtained the following results:

Table 6. Walking Speed in Heel Conditions [9]

________________________________________________

Population Group Walking Speed

Passengers Min (m/s) Max (m/s)

________________________________________________

Female < 30 years 0.39 1.03

Female 30 – 50 years 0.47 1.02

Female > 50 years 0.39 0.87

Male < 30 years 0.45 1.34

Male 30 – 50 years 0.56 1.13

Male > 50 years 0.37 1.07

________________________________________________

Population Group Walking Speed

Crew Min (m/s) Max (m/s)

________________________________________________

Crew 0.68 1.04

________________________________________________

Evacuation simulation is run with layout of liferaft

placement as in the case of standard evacuation.. Then

the results of the travel duration (T) are as follows:

Table 7. Travel Duration Results in Heel Conditions

________________________________________________

Total Travel Duration (T) (second)

Layout 1 Layout 2

________________________________________________

Case 1 639.3 576.3

Case 2 586.5 519.0

________________________________________________

3.3 Evacuation Time Calculation

After all cases get their respective travel duration

(T) values, then the next step is to put in the travel

duration (T) values that have been obtained into the

performance standards for the total maximum

evacuation duration. So that each case can be searched

for the total evacuation time

( ) ( )

1,25

R T E L n+ + + 

( )

30 E L min+

For example, the calculation of case 1 in the fire

scenario where the ship is in an existing condition in

the night case. In the simulation process, the total travel

duration (T) in this case takes 447.5 seconds or 7.45

minutes in equivalent. The calculation of the total

evacuation duration is as follows:

( ) ( )

1,25

R T E L n= + + + 

( ) ( )

1,25 10 7,45 30

= + +

Total evacuation duration = 41,81 minutes

3.3.1 Fire Scenario

Here are the following results of the total

evacuation duration obtained from the evacuation

simulation under standard evacuation condition:

Table 8. Total Evacuation Duration in Fire Conditions

________________________________________________

Total Evacuation Duration (minute)

Layout 1 Layout 2

________________________________________________

Case 1 41.81 40.52

Case 2 34.43 33.17

Case 3 47.97 46.33

Case 4 39.15 38.78

________________________________________________

631

The table above shows the results of calculating the

total required evacuation duration in fire conditions.

From the calculation results obtained the differences in

results in case 1 is 1,30 minutes, case 2 is 1,26 minutes,

case 3 is 1,64 minutes, and case 4 is 0,37 minutes.

3.3.2 Heel Scenario

The results of the total evacuation duration

obtained from the evacuation simulation in heel

conditions are here as follows:

Table 9. Total Evacuation Duration in Heel Conditions

________________________________________________

Total Evacuation Duration (minute)

Layout 1 Layout 2

________________________________________________

Case 1 45.82 44.51

Case 2 38.47 37.06

________________________________________________

The table above shows the results of calculating the

total required evacuation duration in heel conditions.

From the calculation results obtained the differences in

results in case 1 is 1,31 minutes and case 2 is 1,41

minutes.

3.4 Effective Evacuation Time

Based on the results from the calculation of the total

evacuation duration in all cases, it was found that the

difference in the results of the evacuation time was

quite big between the ship in existing conditions and

when the liferaft placement had been changed.

Here are the comparison graphs of the total

evacuation duration obtained from the evacuation

simulations that have been carried out, as follows:

Figure 12. Comparison Chart of Total Evacuation Duration

From those results, it can be concluded that layout

2 (layout of liferaft that has been changed according to

the writer’s suggestion) is more optimal than layout 1

(layout of liferaft with existing conditions).

4 CONCLUSION

Based on the analysis and simulations that have been

carried out with IMO MSC.1/Circ. 1533 standards

regarding the evacuation process on passenger ships,

along with standards and other sources with variations

in evacuation of ships in fire and heel conditions, it can

be concluded that each identified case will show a

different total evacuation duration depending on the

characteristics of each case. In this study, it was found

that layout 2 (layout of liferaft that has been changed

according to the writer’s suggestion) is more effective

than layout 1 (layout of liferaft with existing

conditions). This is proved by the differences in the

total duration of evacuation between layout 1 and

layout 2 in all cases. In the fire conditions, there is a

difference of 1 minute 18 seconds in case 1, 1 minute 16

seconds in case 2, 1 minute 38 seconds in case 3, and 22

seconds in case 4. In the heel conditions, there is

difference of 1 minute 19 seconds in case 1 and 1

minute 25 seconds in case 2. Also, from the modeling

results of passengers evacuation of KMP Aceh Hebat 3,

the entire evacuation durations obtained have fulfilled

the IMO MSC.1/Circ. 1533 rules with maximum

evacuation time for passenger ship is n ≤ 60 minutes.

REFERENCES

[1] D. Lee, J. H. Park, and H. Kim, “A study on experiment of

human behavior for evacuation simulation,” Ocean

Engineering, vol. 31, no. 8–9, pp. 931–941, Jun. 2004.

[2] International Maritime Organization (IMO),

“MSC/Circ.909 – Interim Guidelines for A Simplified

Evacuation Analysis On Ro-Ro Passengers Ship,” 1999.

[3] IMO, “SOLAS-Consolidated-Edition,” 2018.

[4] International Maritime Organization (IMO),

“MSC/Res.809 – Recommendation for Canopied

Reversible Liferaft, Automatically Self-Rghting Liferaft

and Fast Rescue Boat, Including Testing, On Ro-Ro

Passenger Ship,” 1997.

[5] International Maritime Organization (IMO), “MSC.1-

Circ.1533 – Revised Guidelines on Evacuation Analysis

for New and Existing Passenger Ships (Secretariat),”

2016.

[6] T. Pitana, B. K. Artana, D. Prasetyawati and N. Siswantor,

"Observation Study the Walking Speed and Distribution

of Ship's Passenger as Basis for Passenger Evacuation

Simulation," Applied Mechanics and Materuals, vol. 862,

pp. 232-237, 2017.

[7] A. Charchalis and S. Czyz, “Analysis of fire hazard and

safety requirements of a sea vessel engine room.,” Journal

of KONES powertrain and transport, vol. 18, no. 2, 2011.

[8] J. Hietaniemi and E. Mikkola, “Design Fires for Fire Safety

Engineering,” 2010.

[9] R. T. Wijaya, “Analisa Evakuasi Penumpang dan ABK

Kapal PT. Pelni dalam Kondisi List dan Kebakaran,”

2016.