Appendix 1 - Method to Determine the Travel Duration (T) by Simulation Tools for the Advanced Evacuation Analysis
Clasification Society 2024 - Version 9.40
Statutory Documents - IMO Publications and Documents - Circulars - Maritime Safety Committee - MSC.1/Circular.1533 – Revised Guidelines on Evacuation Analysis for New and Existing Passenger Ships – (6 June 2016) - Annex 3 - Guidelines for an Advanced Evacuation Analysis of New and Existing Passenger Ships - Appendix 1 - Method to Determine the Travel Duration (T) by Simulation Tools for the Advanced Evacuation Analysis

Appendix 1 - Method to Determine the Travel Duration (T) by Simulation Tools for the Advanced Evacuation Analysis

1 Characteristics of the models

1.1 Each person (p) is represented in the model individually.

1.2 The abilities of each person are determined by a set of parameters, some of which are probabilistic.

1.3 The movement of each person is recorded.

1.4 The parameters should vary among the individuals of the population.

1.5 The basic rules for personal decisions and movements are the same for everyone, described by a universal algorithm.

1.6 The time difference between the actions of any two persons in the simulation should be not more than one second of simulated time, e.g. all persons proceed with their action in one second (a parallel update is necessary).

2 Parameters to be used

2.1 In order to facilitate their use, the parameters are grouped into the same 4 categories as used in other industrial fields, namely: GEOMETRICAL, POPULATION, ENVIRONMENTAL and PROCEDURAL.

2.2 Category GEOMETRICAL: layout of escape routes, their obstruction and partial unavailability, initial passenger and crew distribution conditions.

2.3 Category POPULATION: ranges of parameters of persons and population demographics.

2.4 Category ENVIRONMENTAL: static and dynamic conditions of the ship.

2.5 Category PROCEDURAL: crew members available to assist in emergency.

3 Recommended values of the parameters

3.1 Category GEOMETRICAL

3.1.1 General

The evacuation analysis specified in this annex is aimed at measuring the performance of the ship in reproducing benchmark scenarios rather than simulating an actual emergency situation. Four benchmark cases should be considered, namely cases 1, 2, 3 and 4 (refer to paragraph 4 for detailed specifications) corresponding to primary evacuation cases (cases1 and 2, where all the escape routes should be assumed to be in operation) and secondary evacuation cases (cases 3 and 4, where some of the escape route should be assumed to be unavailable).

3.1.2 Layout of escape routes – primary evacuation cases (cases 1 and 2): Passengers and crew should be assumed to proceed along the primary escape routes and to know their ways up to the assembly stations; to this effect, signage, low-location lighting, crew training and other relevant aspects connected with the evacuation system design and operation should be assumed to be in compliance with the requirements set out in IMO instruments.

3.1.3 Layout of escape routes – secondary evacuation cases (cases 3 and 4): Those passengers and crew who were previously assigned to the now unavailable primary escape route should be assumed to proceed along the escape routes determined by the ship designer.

3.1.4 Initial passenger and crew distribution condition. The occupant distribution should be based upon the cases defined in chapter 13 of the FSS Code, as outlined in section 4.

3.2 Category POPULATION

3.2.1 This describes the make-up of the population in terms of age, gender, physical attributes and response durations. The population is identical for all scenarios with the exception of the response duration and passenger initial locations. The population is made of the following mix:

Table 3.1 – Population's composition (age and gender)

Population groups – passengers Percentage of passengers (%)
Females younger than 30 years 7
Females 30-50 years old 7
Females older than 50 years 16
Females older than 50, mobility impaired (1) 10
Females older than 50, mobility impaired (2) 10
Males younger than 30 years 7
Males 30-50 years old 7
Males older than 50 years 16
Males older than 50, mobility impaired (1) 10
Males older than 50, mobility impaired (2) 10
Population groups – crew Percentage of crew (%)
Crew females 50
Crew males 50

All of the attributes associated with this population distribution should consist of a statistical distribution within a fixed range of values. The range is specified between a minimum and maximum value with a uniform random distribution.

3.2.2 Response duration

The response duration distributions for the benchmark scenarios should be truncated logarithmic normal distributionsfootnote as follows:

  • For Cases 1 and 3 (Night cases):

  • 400 < x < 700

  • For Cases 2 and 4 (Day cases):

  • 0 < x < 300

  • where x is the response duration in seconds and y is the probability density at response duration x.

3.2.3 Unhindered travel speeds on flat terrain (e.g. corridors)

The maximum unhindered travel speeds to be used are those derived from data published by Andofootnote which provides male and female walk rates as a function of age. These are distributed according to figure 3.1 and represented by approximate piecewise functions shown in table 3.3.

Figure 3.1 – Walking speeds as a function of age and gender

Table 3.3 – Regression formulation for mean travel speed valuesfootnote

Gender Age (years) Speed (m/s)
Female 2 - 8.3 0.06 * age + 0.5
8.3 - 13.3 0.04 * age + 0.67
13.3 - 22.25 0.02 * age + 0.94
22.25 - 37.5 -0.018 * age + 1.78
37.5 - 70 -0.01 * age + 1.45
Male 2 - 5 0.16 * age + 0.3
5 - 12.5 0.06 * age + 0.8
12.5 - 18.8 0.008 * age + 1.45
18.8 - 39.2 -0.01 * age + 1.78
39.2 - 70 -0.009 * age + 1.75

For each gender group specified in table 3.1, the walking speed should be modelled as a statistical uniform distribution having minimum and maximum values as follows:

Table 3.4 – Walking speed on flat terrain (e.g. corridors)

Population groups – passengers Walking speed on flat terrain (e.g. corridors)
Minimum (m/s) Maximum (m/s)
Females younger than 30 years 0.93 1.55
Females 30-50 years old 0.71 1.19
Females older than 50 years 0.56 0.94
Females older than 50, mobility impaired (1) 0.43 0.71
Females older than 50, mobility impaired (2) 0.37 0.61
Males younger than 30 years 1.11 1.85
Males 30-50 years old 0.97 1.62
Males older than 50 years 0.84 1.4
Males older than 50, mobility impaired (1) 0.64 1.06
Males older than 50, mobility impaired (2) 0.55 0.91
Population groups – crew Walking speed on flat terrain (e.g. corridors)
Minimum (m/s) Maximum (m/s)
Crew females 0.93 0.55
Crew males 1.11 1.85

3.2.4 Unhindered stair speedsfootnote

Speeds are given on the base of gender, age and travel direction (up and down). The speeds in table 3.5 are those along the inclined stairs. It is expected that all the data above will be updated when more appropriate data and results become available.

Table 3.5 – Walking speed on stairs

Population groups – passengers Walking speed on stairs (m/s)
Stairs down Stairs up
Min. Max. Min. Max.
Females younger than 30 years 0.56 0.94 0.47 0.79
Females 30-50 years old 0.49 0.81 0.44 0.74
Females older than 50 years 0.45 0.75 0.37 0.61
Females older than 50, mobility impaired (1) 0.34 0.56 0.28 0.46
Females older than 50, mobility impaired (2) 0.29 0.49 0.23 0.39
Males younger than 30 years 0.76 1.26 0.5 0.84
Males 30-50 years old 0.64 1.07 0.47 0.79
Males older than 50 years 0.5 0.84 0.38 0.64
Males older than 50, mobility impaired (1) 0.38 0.64 0.29 0.49
Males older than 50, mobility impaired (2) 0.33 0.55 0.25 0.41
Population groups – Crew Walking speed on stairs (m/s)
Stairs down Stairs up
Min. Max. Min. Max.
Crew females 0.56 0.94 0.47 0.79
Crew males 0.76 1.26 0.5 0.84

3.2.5 Consistency of travel speed

The unhindered travel speeds of each evacuee on flat terrain and on stairs (down and up) are consistent within the respective ranges specified in tables 3.4 and 3.5.

3.2.6 Exit flow rate (doors)

The specific unit flow rate is the number of escaping persons past a point in the escape route per unit time per unit width of the route involved, and is measured in number of persons (p). The specific unit flow ratefootnote for any exit should not exceed 1.33 p/m/s.

3.3 Category ENVIRONMENTAL

Static and dynamic conditions of the ship. These parameters will influence the moving speed of persons. Presently no reliable figures are available to assess this effect; therefore, these parameters could not yet be considered. This effect will not be accounted for in the scenarios (cases 1, 2, 3 and 4) until more data has been gathered.

3.4 Category PROCEDURAL

For the purposes of the four benchmark cases, it is not required to model any special crew procedures. However, the distribution of the crew for the benchmark cases should be in accordance with 4.

3.5 It is expected that all data provided in paragraphs 3.2 and 3.3 will be updated when more appropriate data and results become available.

4 Detailed specifications (scenarios) for the 4 cases to be considered

For the purpose of conducting the evacuation analysis, the following initial distributions of passengers and crew should be considered as derived from chapter 13 of the FSS Code, with the additional indications only relevant for the advanced evacuation analysis. If more detailed data considering the distribution of crew is available, the distribution may deviate from the following specifications:

4.1 Cases 1 and 3 (night)

Passengers in cabins with maximum berthing capacity fully occupied; 2/3 of crew members in their cabins; of the remaining 1/3 of crew members:

  • .1 50% should be initially located in service spaces;

  • .2 25% should be located at their emergency stations and should not be explicitly modelled; and

  • .3 25% should be initially located at the assembly stations and should proceed towards the most distant passenger cabin assigned to that assembly station in counterflow with evacuees; once this passenger cabin is reached, these crew are no longer considered in the simulation. The ratio between the passenger and counterflow crew should be the same in each main vertical zone.

4.2 Cases 2 and 4 (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 be distributed as follows:

  • .1 1/3 of the crew will be initially distributed in the crew accommodation spaces (cabins and crew day spaces);

  • .2 1/3 of the crew will be initially distributed in the public spaces;

  • .3 the remaining 1/3 should be distributed as follows:

    • .1 50% should be located in service spaces;

    • .2 25% should be located at their emergency duty locations and should not be explicitly modelled; and

    • .3 25% should be initially located at the assembly stations and should proceed towards to the most distant passenger cabin assigned to that assembly station in counterflow with evacuees; once this passenger cabin is reached, these crew are no longer considered in the simulation. The ratio between the passenger and counterflow crew should be the same in each main vertical zone.

5 Procedure for calculating the travel duration T

5.1 The travel duration, both that predicted by models and as measured in reality, is a random quantity due to the probabilistic nature of the evacuation process.

5.2 In total, a minimum of 500 different simulations should be carried out for each of the benchmark cases. This will yield, for each case, a total of at least 500 values of tA.

5.3 These simulations should be made up of at least 100 different randomly generated populations (within the range of population demographics specified in paragraph 3). Simulations based on each of these different populations should be repeated at least 5 times. If these 5 repetitions produce insignificant variations in the results, the total number of populations analysed should be 500 rather than 100, with only a single simulation performed for each population.

5.4 The minimum number of 500 different simulations can be reduced when a convergence is determined by an appropriate method, such as the one shown in appendix 3. The total number of different simulations should be in this case not less than 50.

5.5 The value of the travel duration for each of cases 1 to 4: the value tI is taken which is higher than 95% of all the calculated values (i.e. for each of cases 1 to 4, the durations tA are ranked from lowest to highest and tI is selected for which 95% of the ranked values are lower).

5.6 The value of the travel duration to comply with the performance standard T is the highest of the four calculated travel durations tI (one for each of cases 1 to 4).

5.7 The procedure for calculating the travel duration for cases 5 and 6 should be based on the same principles as for cases 1 to 4.

6 Documentation of the simulation model used

6.1 The assumptions made for the simulation should be stated. Assumptions that contain simplifications above those in paragraph 3.2 of the Guidelines for the advanced evacuation analysis of new and existing passenger ships, should not be made.

6.2 The documentation of the algorithms should contain:

  • .1 the variables used in the model to describe the dynamics, e.g. walking speed and direction of each person;

  • .2 the functional relation between the parameters and the variables;

  • .3 the type of update, e.g. the order in which the persons move during the simulation (parallel, random sequential, ordered sequential or other);

  • .4 the representation of stairs, doors, assembly stations, embarkation stations, and other special geometrical elements and their influence on the variables during the simulation (if there is any) and the respective parameters quantifying this influence; and

  • .5 a detailed user guide/manual specifying the nature of the model and its assumptions and guidelines for the correct use of the model and interpretations of results should be readily available.

Parent topic: Annex 3 - Guidelines for an Advanced Evacuation Analysis of New and Existing Passenger Ships
Copyright 2022 Clasifications Register Group Limited, International Maritime Organization, International Labour Organization or Maritime and Coastguard Agency. All rights reserved. Clasifications Register Group Limited, its affiliates and subsidiaries and their respective officers, employees or agents are, individually and collectively, referred to in this clause as 'Clasifications Register'. Clasifications Register assumes no responsibility and shall not be liable to any person for any loss, damage or expense caused by reliance on the information or advice in this document or howsoever provided, unless that person has signed a contract with the relevant Clasifications Register entity for the provision of this information or advice and in that case any responsibility or liability is exclusively on the terms and conditions set out in that contract.