Fire safety measures for road and rail tunnels

A wide range of factors need to be considered when it comes to preventing tunnel fires. From tunnel design to actual fire fighting – every element could prove critical in saving lives.

In recent years tunnel fires have been in the news, reflecting, or stimulating, public concern. This public concern is no doubt in part due to the scale of the losses, both in human life, property and consequential losses to business and transport systems. It may also in part be due to the nature of tunnels; the risks associated with transportation are put into one of the worst environments for a fire – underground and potentially many miles from safety. And unlike other difficult environments, such as industrial or offshore structures, tunnels are used by the ordinary public when travelling by train or car. The more notable tunnel fires have been on the road network and include;

• Storebaelt Tunnel, Denmark/Sweden, June 1994 (tunnel under construction), no fatalities.
• Pfander, Austria 1995, 3 dead, 4 injured.
• Isola della Femmine, Italy 1996, 5 dead, 20 injured.
• Ekeberg, Norway 1996, No fatalities.
• Gotthard, Switzerland, 1997, No fatalities.
• Mont Blanc Tunnel, France/Italy, March 1999, 41 dead.
• Tauern Tunnel, Austria, May 1999, 12 dead.
• Gotthard, Switzerland, October 2001, 11 dead.
• Frejus, France/Italy, June 2005, 2 dead.
• 155 people died in the Kaprun Funicular Tunnel, Austria, in November 2000.

In the UK, our experience of serious tunnel fires (and tunnel-like fires) is relatively limited but include the Summit Tunnel fire (December 1984, no fatalities), Kings Cross (November, 1987, 31 fatalities) and the Channel Tunnel fire in November 1996 (also no fatalities), all of which involved the rail infrastructure.

Tunnel fires involve all aspects of fire safety science, engineering and management. The designer or engineer who is considering fire safety as part of the tunnel design has to have an appreciation of a wide range of disciplines if they are to provide a safe tunnel.

Engineering systems

In general, the two main fire related areas of concern for tunnel engineers have been structural safety and ventilation, in particular smoke ventilation.

The structural fire protection in a tunnel serves a number of purposes, which includes protection of the means of escape and/or safe havens (refuges), assisting (or protecting) rescue efforts, avoiding major business interruption, and seeking to avoid complete or partial loss of the tunnel.

The designer needs to consider the scenarios, the risks, the role in protecting life or in protecting the structure as a system, and there may be crucial commercial and/or cost decisions that inform this process. Decisions must be made regarding the design fires, their size and location, how much protection (if any) to apply, and how much damage is ‘acceptable’. There is an increasing body of knowledge available to assist in the characterisation of the design fire, and computer modelling tools have been available for some time to assist in the assessment of heat and smoke flow.

Similarly, the provision of tunnel equipment is an essential part of the fire safety system, and will include protecting or assisting the means of escape, with lighting, emergency lighting and communications, power for essential safety systems, and provision of fire fighting facilities.

Many tunnels will be provided with smoke control systems, which need to be properly protected against direct fire attack, hot smoke, and power loss, but also will be integrated with the operation of any ventilation equipment; in both the Channel Tunnel and Mont Blanc fires the inappropriate operation of ventilation systems appears to have had a negative impact on the actual or potential outcome of the incident.

Communications systems are of critical importance; in both the Channel Tunnel and Mont Blanc fires the loss of communication cables through the tunnel appear to have had an impact of the handling of the incident. Other aspects of equipment performance that are of importance include:

• The need to ensure that communication channels do not become overloaded,
• Applying the appropriate detector logic and detector response,
• Being able to identify the location of the incident in the tunnel,
• Ensuring that the speed of response of essential equipment is adequate,
• Avoiding spurious problems that confuse the response to an incident,
• Avoiding false alarms.

All of these issues need to be considered during design and, where appropriate, approved products and services should be specified.

Fire Fighting

Fire fighters face particular problems in dealing with tunnel fires which include long communication lines, the difficulty of getting to the fire, the distance from facilities and support, the growth in the size of fire (as a result of these delaying factors), and problems of joint action (language and procedures) when two countries are involved.

However, recent events have highlighted the need to integrate fire safety with other emergency response strategies. This requires coordination between network staff, trained staff and emergency services, adequate and appropriate access for emergency services, and the provision of services (e.g. water, communications). As well as fire, the threats that may need to be considered include CBRN (chemical, biological, radiological, nuclear attack), explosions, derailments, collisions, earthquakes (although not in the UK) and other incidents, some or all of which may involve a consequential fire. The need to design safety systems that can cope with any, or any selection, of these types of incident, requires careful thought.

Planning

Careful planning, in advance of any formal design, is becoming increasingly important. In selecting the fire safety philosophy and designing the fire safety systems for a tunnel there are a range of aspects to consider.

Of these it is the selection of the design scenarios which is often most controversial. The need to select one or more events based on the location of the fire, its size, rate of growth and smoke and toxic gas production will be fundamental to the design of, for example, structural protection and ventilation (for smoke control). Often such events will be selected on the basis of probability, and the acceptable values will require different criteria for life safety or property protection. In adopting such a risk-based approach there are likely to be differences of view from the client and/or operator, the regulator or enforcer and the general travelling public.

In adopting a risk-based approach to the fire protection of a tunnel system there are therefore a number of issues to consider, which include:

• The construction of decision flow-charts and how to anticipate every possible (or probable) event
• Carrying out a risk analysis – and determining what assumptions are appropriate and how the results should be interpreted
• How to balance the needs for life protection and property (asset) protection
• The selection of fire protection and detection products to ensure that they work as claimed
• The difficulty of doing full ‘emergency’ simulations, for example as part of the commissioning process, to ensure that engineering systems and human procedures will work properly if needed
• The problems of maintaining good practice (e.g. ensuring that emergency doors are kept closed over the life of the tunnel)
• The need for regular checks of safety systems, and ensuring that faults are rapidly corrected
• How to consider how the incident will cascade; almost every major incident is made up of lots of interacting events, all of which are individually unlikely
• The management implications. It may be this issue which can prove critical in the event of an incident.

Management

There are a number of management issues which can have a significant impact on the outcome of an incident, and need to be carefully considered at the design stage, at the initial opening of a tunnel, and maintained during its operational lifetime. These include:

• Ensuring an appropriate response by security or safety staff
• Ensuring that staff (both one’s own and others using the system) are well trained in a range of emergency procedures
• Recognising and responding to a variety of passenger responses and behaviours during an incident
• Being able to react and respond to the speed of events during an incident, (for example, opening emergency exits)
• Arranging effective communications between relevant agencies and emergency services
• Arranging for detection and alarm provisions to be continually checked and reassessed once the tunnel is operation
• Providing for communication and control systems to be routinely tested under ‘realistic’ ‘emergency’ conditions, and ensuring that key personnel are not overloaded
• Ensuring that all safety equipment is subject to routine, and documented, maintenance, testing and repair.

There is therefore a need for the design engineer to understand how the effectiveness of his designs will be influenced by, and influence, safety management, and engineers need to be educated to identify all of the management implications, implicit and explicit, in their fire safety designs. It is to include the design of signage and communicating with vehicle occupants, disabled access and other disability issues, interactions with other tunnel systems and emergency planning and advise on control rooms.

Designing for real people; understanding human behaviour

A successful emergency plan exploits the behaviour of the people involved; in an emergency it is not realistic to expect those people to undergo a rapid learning experience, for example, in how to open a safety hatch. The first few minutes in an incident are critical and it is essential to provide the means to motivate vehicle occupants to make their escape. Research findings have demonstrated the importance of ‘cues’ and communication between vehicle occupants, as well as communications between staff and passengers.

An understanding of human behaviour and evacuation responses in the various possible emergency situations, and in the unfamiliar environment of a rail tunnel, is necessary if successful safety system is to be provided. Issues that the designer, and later the operator, need to consider include movement speeds in smoke, the intelligibility of signage, communication between people, communicating with train occupants, communicating with tunnel occupants and disabled egress and other disability issues. It is important to learn from actual incidents, false alarms, drills and other types of incident.

New challenges

A number of trends are adding to the challenge of ensuring safety from fires in tunnels. There are more tunnels – rail and road –being built, and longer tunnels are being built, which are increasingly critical to the whole social and commercial infrastructure. Traffic densities are increasing, with little spare capacity or safety margin, and larger, and more combustible, loads are being transported. In addition new and potentially more hazardous fuels are increasingly being introduced, such as LPG and hydrogen. Within Europe, tunnel safety regulations are being developed for road tunnels (through the EC Directive on minimum safety requirements for tunnels in the Trans-European Road Network) and rail tunnels (through the EC Interoperability Directives on the safety of the Community’s railways).

Research is needed to reduce the risk of fire and to find ways to make people escape quickly in the event of an incident. The data, assumptions, models and design methodologies we use must be well founded and reflect what happens in the ‘real’ world.

Conclusions

We need to learn lessons from real incidents; to ensure that such events are less likely to reoccur, to identify the ‘near miss’ features, and it is essential that information from real fires is fed into the fire science and design knowledge base. We need to maintain and develop the dialogue between tunnel designers, vehicle designers, fire investigators, fire scientists and fire engineers.

Fire safety engineers and designers need to ensure that they identify all of the management implications, implicit and explicit, in their designs. They should consult fire safety managers, or representatives of eventual fire safety managers, from the earliest stage of the design process, and ensure that the concerns of these managers are taken into account. Long-term issues, such as change of use and maintenance, must be considered, and should include consideration of exceptional situations (such as equipment failure). The management implications of the design should be stated and recorded, and made available to later managers. The designer or engineer needs to be as sure as possible that managers will be given appropriate powers and resources, and their fire safety role will be an explicit and accountable part of their job. Or he must ensure that his design makes no such management demands.