FSEG LOGO FIRE SAFETY ENGINEERING GROUP The Queen's Anniversary Prize 2002 The British Computer Society IT Awards 2001 The European IST Prize Winner 2003 The Guardian University Awards Winner 2014
The Faculty of Architecture, Computing & Humanities

FIRE-EXIT : Formulation of Immediate Response and Evacuation Strategies through Intelligent Simulation Assessment and large-scale Testing

This is an EU funded project under Framework 5 Growth Programme (contract GRD2-2001-50055) which runs from 2001-2005.

Maritime safety is one of the central concerns of the European Commission’s transport policy. This notes the extent of losses at sea (currently on average 230 vessels and over 1,000 lives every year), the additional impact on marine pollution and its international dimension.

Council Directive 98/18/EC of 17 March 1998 on safety rules and standards for passenger ships (Official Journal L 144 , 15/5/1998, pp.1-115) aimed to introduce a uniform level of safety of life and property on new and existing passenger ships and high-speed passenger craft, when both categories of ships and craft are engaged on domestic voyages, and to lay down procedures for negotiation at international level with a view to a harmonisation of the rules for passenger ships engaged on international voyages. Specifically, this Directive sets out requirements applicable to new ro-ro passenger ships constructed on or after 1 July 1999. For new Class B, C and D ro-ro passenger ships constructed on or after 1 July 1999, escape routes shall be evaluated by an evacuation analysis early in the design process.

The analysis shall be used to identify and eliminate, as far as practicable, congestion which may develop during an abandonment, due to normal movement of passengers and crew along escape routes. Including the possibility that crew may need to move along these routes in a direction opposite to the movement of the passengers. In addition, the analysis shall be used to demonstrate that escape arrangements are sufficiently flexible to provide for the possibility that certain escape routes, assembly stations, embarkation stations or survival craft may not be available as a result of a casualty.

FIRE EXIT proposes to equip the marine industry with a Ship Evacuation Simulator that is a quantum leap in the level of reliability, realism and design utility of today’s ship evacuation software. The FIRE EXIT ship evacuation simulator (maritimeEXODUS) will be capable of addressing issues of mustering, ship motions, fire and abandonment. In achieving these aims the FIRE EXIT project does not intend to re-invent existing technology. Rather, it will take as its starting point the very leading edge of the current state-of-the-art in ship evacuation simulation (maritimeEXODUS), fire simulation (SMARTFIRE) and large-scale experimental facilities (SHEBA, amongst others) and significantly enhance these capabilities. FIRE EXIT fits squarely into the objectives of 3.2.1: Efficient, safe, and environmentally friendly ships and vessels.

The results and deliverables of FIRE EXIT address six principal areas of investigation

  • the development of probabilistic design methods in the fire risk area consistent with the Safety Case approach and the new fire protection regulations, which specify how safety equivalence may be achieved,

  • the development of an interface to CFD software predicting the spread of fire and the creation of a module to derive from there the pattern of smoke spreading during the evacuation and its impact on evacuation and abandonment,

  • the creation of the physical mechanisms to enable realistic and valid large-scale test facilities to simulate ship motions and smoke spreading in real-time, and to establish performance data for Life Saving Appliances,

  • the development of AI-based software which

  • will convert the output of the simulation runs into a reduced set of human-intelligible figures of merit, and

  • will be able to trace back to its source any failure to achieve a sufficiently high assessment recommend improvements leading to a higher assessment

  • the development of Concept Design software with:

    • an XML based interface to the simulation module, enabling conceptual designs to be tested at an early stage, and

    • a Safety Assessment module which will take the results of simulation runs, the assessment provided by the AI module and the Classification Society Fire Safety Rules and create in a principled way an evacuation-safe design.

FIRE EXIT will be created by a Consortium derived from a number of different countries which will include:

British Maritime Technology, United Kingdom: One of the world's leading maritime and engineering consultancies providing a wide range of advanced design, risk management, experimentation and software systems - primarily to environmental, marine, offshore, defence and civil engineering markets.
BMT Fleet Technology Limited, Canada: BMT Fleet Technology Limited, a British Maritime Technology company, performs applied research, development and engineering for the transportation, environment, resource/energy, fabrication and manufacturing sectors, including government and the military.
Det Norske Veritas, Norway: Established in 1864, DNV is an independent foundation with the objective of safeguarding life, property and the environment and is a leading international provider of services for managing risk.
Marine Institute, Canada: World-class Centre of Excellence in education and advanced fisheries and marine technology.
Grandi Navi Veloci is the youngest cruise line in the Gruppo Grimaldi, founded in 1992, as the result of a decision to introduce a new notion of ships in the Mediterranean-ships that could respond to the need for fast passenger, car and merchandise transport while combining the comfort and luxury of an cruise ship.
METTLE Groupe, Italy: Mettle offers important and wide competence in Engineering and Innovation for maritime technology, Design and Re-engineering including Logistics, Supply and Demand Chain Management, Intermodal Transport, Shipbuilding and Ship Design, Intelligent Transport Systems, Automatic Control, Marine Environment, Safety and Risk Analyses, together with broad experience of full-scale marine and land operations.
National Research Council Canada, Canada: NRC, Canada's premier science and technology research organization, is a leader in scientific and technical research, the diffusion of technology and the dissemination of scientific and technical information.
TRIBON Solutions, Sweden: The world leader in shipbuilding CAD/CAM/CIM systems.
University of Greenwich University of Greenwich, United Kingdom: The University of Greenwich is a major modern UK university. Its three campuses extend from Maritime Greenwich to Medway,and aims to combine the best of the "old" and "new" university traditions, combining strong regional links and a mission for access and lifelong learning with research excellence and an international role.

The idea behind the development comes from the final end-users of the system: a ship operator, a shipyard, a classification society and a ship designer. Apart from the end-users, the consortium also includes a CAD vendor and major research organisations which will both carry out theoretical development and a comprehensive experimental programme comprising model tests, large-scale tests and full scale ship trials.

The scientific and technological output from FIRE EXIT (by way of the maritimeEXODUS model) will be used by ship designers during the concept phase, shipping regulators for the certification of ship design and by ship operators for training on shore and at sea. In the early stages of the design process FIRE EXIT (by way of the maritimeEXODUS model) will bring important issues of safety, evacuation, staffing and procedures to the fore of ship design in a manner that will be quantifiable and reproducible. In a similar process, ship regulators will be able to quickly assess a proposed design, including the crew procedures and determine whether proposed designs meet acceptable standards. Finally, ship operators will be able to assess safety provision on-board as conditions including number, type and location of passengers, number of crew etc., change. On-board versions of the software running in real time will allow ship operators to train crews and even re- direct fire fighting and passenger evacuation activities in response to the on- board situation.
The project is separated into six core work packages, each of which contributes significantly to the overall success of the project:
WorkPackage 1 - Accident Scenario Definition - DNV
Goals - to identify and generate a prioritised list of potential risks that need to be addressed by maritimeEXODUS; to identify the distribution of risk, thus allowing attention to be focused upon high risk areas; to propose effective and practical risk control options that the optimisation and design software should be able to analyse; to identify benefits and costs associated with the implementation of the risk control options; recommendation on how software could be used in design and design approval.

WorkPackage 2 - SimulationUniversity of Greenwich
Goals - define in detail what the project requires from the simulation module and what changes must be made to existing software; create and implement mechanism to allow fire hazard data to be seamlessly imported into the maritimeEXODUS from SMARTFIRE; create behaviour model to simulate individual and group behaviour under smoke and list conditions within maritimeEXODUS; create model to simulate individual and group behaviour under dynamic motion conditions within maritimeEXODUS; create behaviour model to simulate behaviour under both smoke and dynamic motion conditions within maritimeEXODUS; modify present abandonment model within maritimeEXODUS to incorporate data from trials.

WorkPackage 3Large-Scale RigFTL
Goals - develop the performance and technical specifications for the changes required to SHEBA to allow motions and smoke to be introduced; design and implement and test the system changes on SHEBA for smoke introduction and dispersion; design, acquire components and install the revised hydraulics system and associated structural changes to permit motions on SHEBA; engineer and implement the changes to the SHEBA infrastructure to permit the full range of SHEBA motions; design and develop and implement the control system required to provide full SHEBA motions.
WorkPackage 4Interpretation & OptimisationBMT
Goals - definition of the requirements and specification of the interpretation & optimisation module; design of the interpretation and optimisation module; develop the prototype interpretation module; design & develop the final optimisation module.


WorkPackage 5Concept DesignTRIBON
Goals - define requirements and specifications for a seamless bi-directional link between concept design and evacuation simulation; design of the annotation facility in concept design; creation of the annotation facility; integration with both the remainder of the concept design software and with the simulation and interpretation & optimisation modules.


WorkPackage 6Testing & TrialsFTL
Goals - To collect data on the behaviour and mobility capability of passengers of all ages and mobility, under conditions of smoke and static list; to collect data on the mobility and behaviour of passengers of all ages and mobility under conditions of ship motion and to validate the program; to collect data on the limiting sea conditions for successful launch of Lifesaving Appliances and on the increased times for settling LSAs when there are waves; to collect data on the times for the various components of abandoning via a variety of LSAs using a full scale rig; to execute ship evacuation trials to validate the FIRE EXIT model (maritimeEXODUS); testing of software and evaluation by end-users.

These packages interact in order to address key components in the use, application and result analysis during the application of an evacuation model. These components can be represented by a number of key questions:

  • What scenarios should be modelled?
  • What data is required to perform the necessary simulations?
  • Is this data available and if not provide it where possible?
  • How easy is it for the engineer to configure the model for use? Provide a mechanism to simplify this process – in this case a method by which to simplify importing the general arrangement of the vessel into the simulation model.
  • Is the simulation model able to represent the behaviours expected during an evacuation? Modify the simulation in order to better represent the evacuation process and to incorporate the data generated within the model.
  • Is the simulation model able to accurately simulate the environmental conditions to which an evacuee might be exposed? Modify the simulation in order to better represent the environmental conditions, by providing direct access to results produced using a sophisticated CFD model.
  • What type of results would the user require? Is the model currently able to produce such results? If not, produce a module to configure the results produced into a format that is readily understood.
  • Does the model perform satisfactorily? Generate a data-set that provides sufficient detail for a number of the components within the model to examined.

The main FSEG contribution to FIRE EXIT is presented below. This contribution can be broken down according to the sub-tasks associated with Workpackage 2.

T2.1 UOG, in collaboration with the end-users, defined in detail the modifications required in the software to allow it to accommodate smoke and dynamic motions and to streamline the process of input and output of data.
T2.2 A Volume Averaging module was created by UOG, which took 3D discretized data from SMARTFIRE and constructed the zone input required by maritimeEXODUS.

Several examples have been produced in order to demonstrate the progress that has been made during this task.
Results produced by SMARTFIRE model.
Results produced by the maritimeEXODUS model once the configured SMARTFIRE data has been imported.


T2.3 This task incorporated the data from the smoke and list trials, after the video data had been analyzed and the impact that the list and smoke had on the behaviour of individuals and groups had been determined. A Behaviour Model component has been prototyped in order to simulate the observed behaviour.


Quantitative Behavioural Model Qualitative Behavioural Model

Effectively, the data has been analyzed to determine the impact of static heel and smoke upon the travel speeds that were attained and the behaviours that were evident during the trial. This has then been incorporated into the evacuation model in order to better represent the behaviour observed.

Example of smoke trial footage
Incorporation of this data within maritimeEXODUS- example 1
Incorporation of this data within maritimeEXODUS- example 2


T2.4 This task will run in two discrete time lapses. In the first phase, the basic capabilities were developed in the software allowing dynamic motions to be modelled in maritimeEXODUS.

Development of model within maritimeEXODUS

Then in the second phase, after the first large-scale trials are concluded, a more detailed representation of the behaviour evident and the quantitative results produced will be implemented.
Example of the type of data that has been collected.

The video data will then be analyzed and the impact of dynamic motions on passengers (both on individuals and on groups) will be modeled. Software will be created to encapsulate the models abstracted from this analysis.


T2.5 Similarly to T2.4, this task will also run in two discrete time lapses. In the first period, the basic capabilities will be developed in the software which will allow smoke plus dynamic motions to be accommodated in the simulation software. In the second period, after the first large-scale evacuation trials (T5.2) are concluded, the video data will be analysed and the impact of smoke in addition to dynamic motions on passengers (both on individuals and on groups) will be modelled. Software will be created to encapsulate the models defined by this analysis. UOG will lead and FTL will ensure the work is compatible with the development in WP3.

Development of model within maritimeEXODUS


T2.6 After the conclusion of the Model and Full-Scale Abandonment Trials at facilities of IMD and MI the data will be analyzed and used to modify the abandonment model to take into account individual and group behavior. UOG will conduct the work.

To demonstrate the type of process that is expected of this project, a simple example has been produced. This has involved a number of stages, each involving a different partner in order to exemplify the advances that have been made within the project.
Normally the general arrangement of the vessel would be provided to the engineer in a format that includes a vast amount of information that would not be required for the simulation process to take place. Not only is this information redundant, it precludes the accurate representation of the space available for the passengers during an evacuation.

As is apparent from the picture below, the description of the cabin section contains far too much detail for the deck to be represented accurately.

Before this could be used, the extraneous lines would have to be removed manually in order for the occupiable space to be meshed accurately.


Lines have been deleted from the GA Space now available for the geometry to be meshed accuarately

The manipulation of overly cluttered drawings such as that shown, would delay the modelling of the vessel by potentially several days. The TRIBON package has now been modified so that it now produces EXODUS-ready general arrangements with the information that is not required, already removed from the file.

Standard Tribon view General Arrangement of the file
3-dimensional view of stripped file 2-dimensional view of stripped file

This file, produced by TRIBON, can then be exported directly to maritimeEXODUS and then used directly in the simulation process.


Once the file has been imported into the model, the geometry can be meshed, the scenario defined and the simulations performed.
maritimeEXODUS produces a vast set of data, enabling the user to understand the conditions that developed during the simulation. A software module has been developed by BMT to filter this information and produce and XML report allowing the user to interrogate the data in a more ‘human-friendly’ format.

This module is able to read the data exported from the EXODUS model into the ASCII-based *.sim file. This file (or a set of such files) is converted into an XML report which can then be read via an internet browser (such as Mozilla).
This module has converted the raw data produced by EXODUS into a report that presents judgements based on the results produced. the XML format allows the user to interrogate the judgements made in order to get more information.

For instance the assumptions on which the judgements are made can be viewed;

the data can be presented in a graphical format;

or the values being described can be viewed in a more traditional manner.


See publications # 166, 165, 164, 152, 145.

Copyright © 2003- University of Greenwich | Privacy and cookies | Accessibility | Contact the Webmaster