Level crossing safety in the UK compares favourably with that in other European countries. However, collisions at level crossings are still the largest single cause of train accident risk.
Level crossings are an open interface between the road and the railway, so there is increased potential for pedestrian and road user behaviour to affect train operations and put themselves in danger. The risks, however, can be reduced by the safe design and engineering management of level crossings, thereby reducing the number of serious and fatal incidents.
Risk control should, ideally, be achieved through the elimination of level crossings in favour of bridges, underpasses or diversions. However, with the majority of the rail network designed over 100 years ago, such elimination can be very difficult and expensive to achieve. Even if resources could be made available, there is often not the available space to construct a safer alternative to a level crossing.
Level crossings connect communities, and people in those communities often want their crossings to remain open, even when a case for closure on safety grounds has been made. Where elimination is not possible, the risks to users and the operational railway should be reduced so far as is reasonably practicable.
A challenging role
There are approximately 6,000 level crossings in use on the mainline rail network. Trains are now typically more frequent and travel at higher speeds than ever before while, at the same time, more road traffic crosses the railway and larger farm vehicles with better soundproofing are using occupational crossings. In addition, more pedestrians carry electronic equipment that can distract them when crossing the railway.
Ed Rollings is head of level crossings engineering at Network Rail, so the development and introduction of innovatory level crossing controls, and the technology that is being used to reduce level crossing risk, falls to him.
Level crossings are unique in a number of ways. The layout, configuration and use of level crossings vary from location to location. Infrastructure managers have little direct control over the variety of users that are entitled to cross, and some of the features and functionality of level crossings have been specified in legislative requirements for many years, without the benefit of modern technology.
The total cost of ownership of level crossings is significant in annual maintenance and delay minutes. There is also an operational cost in signaller and crossing keeper hours to manage and control level crossings; whether through CCTV monitoring and control, level crossing telephone requests to cross, or permanently staffed crossings manually operated. All this, with the requirement to provide systems more cost effectively, and quicker, makes Ed’s role very challenging, but very important.
Network Rail’s vision is to achieve ‘zero harm’ at level crossings and the policy is to make them safe so far as is reasonably practicable. To achieve this, Network Rail is investing in research and innovation in collaboration with industry, to identify new and emerging technologies that will deliver cost effective protection.
The aim is to introduce technology solutions that provide an active interface to users, providing safe crossing information and encouraging safe behaviours at level crossings. Another challenge is that the active interface, in whatever form, should ideally not rely on infrastructure-based train detection as the industry moves toward train-based position and location systems in support of European Train Control System (ETCS) Level 3.
Crossings as a system
The head of level crossings engineering manages a small team of engineers who have extensive knowledge and experience of signalling engineering with respect to level crossings. Where necessary, they have access to other engineering disciplines, such as track, structures, telecommunications and electrical power, as well as operations, risk and human factors experts. This provides a holistic approach to level crossing engineering and operation, which is an important point.
Ed’s team manages a programme of work to specify the policy and standards that will deliver improvements to level crossing protection. The programme covers short-term improvements to existing systems, response to accident and incident investigation recommendations, research into human factors and new technology and management systems. The scope also provides, where applicable, assurance regimes to confirm the policy and standards implementation.
In addition, the head of level crossings engineering is accountable for responding to formal and Rail Accident Investigation Branch (RAIB) investigation recommendations relating to level crossing assets. This includes reviewing and responding to draft reports, representation at national recommendations review panels, analysis of problems and the development of solutions. Proposed solutions will normally be mandated through a special inspection notice, unless of a significant or strategic nature. Significant or strategic changes will be developed as a ‘project’ and managed accordingly.
A sample of level crossing sites is subject to engineering verification assurance by the team, normally as part of verification being undertaken by the parent discipline of the asset sub system concerned, such as track, signalling, telecoms or electrical power. This may also include accompanying the route-based level crossing managers on their routine inspections and to follow through with a review of the effectiveness of their inspections.
The legal framework governing safety at level crossings is complex, sometimes out-dated (some legal requirements are Victorian in origin) and overly prescriptive in places. When a railway route was authorised by an act of parliament, the legislation would have described the requirements for any level crossings, in terms of gates, operator and signage. When the method of protection of a crossing changes, then the act of parliament has to be modified by a Level Crossing Order. The Office of Rail and Road (ORR) is responsible for authorising Level Crossing Orders (on behalf of the Secretary of State for Transport), and then in inspecting against them to ensure that the measures that are set out in the Order are actually in place and being complied with. Therefore, it is important that the ORR is involved at an early stage with any level crossing innovation.
Technical strategy development
Level crossings are provided with many different forms of protection, ranging from full barriers and obstacle detectors linked with protecting signals to simple warning boards at footpath crossings. They can be categorised in a number of ways according to function, usage or method of operation. However, for the purposes of the technology strategy, level crossings can be categorised to be one of the three types – passive, automatic or controlled.
Passive level crossings
These are operated by the user and there is no interface to the signalling system. They comprise:
• Footpath and bridleway (FP + BW) – with or without telephone and/or Miniature Stop Light (MSL);
• User Worked Crossing (UWC) – with or without telephone and/or MSL;
• Open Crossing (OC) – with warning boards to make users aware of a level crossing and to instruct them to only use the crossing when they have determined no train is approaching by direct observation or telephone, along with ‘whistle’ boards for approaching trains.
This type of crossing presents a high risk and, over the last two years, seven of the ten accidental fatalities that have occurred at level crossings have been at FP level crossings, and one was at a UWC with telephone.
Historically, the only technology safety enhancement has been a telephone provided at some crossings with poor sighting of approaching trains. These connect to the controlling signal box, but can be a weak method of protection in some locations. The signaller may have to deal with a large number of crossings and, in very long signal sections, may not know where a train is with respect to the crossing. Furthermore the user is unlikely to be conversant with railway voice protocol. With the introduction of larger signalling control areas, telephone protection of crossings is unsustainable.
Recent product introductions include Overlay Miniature Stop Lights (which are no longer miniature!) branded VaMoS supplied by Schweizer, and the EBI Gate 200 from Bombardier. A number of these are planned for deployment across the network. The EBI Gate has been totally re-engineered by Bombardier to remove a design deficiency and a programme of work is underway to roll out the new version to all the original pilot deployments.
Where a passive crossing is provided with manual gates, there is a risk that users who make regular use of the crossing may be tempted to leave the gates open. A power-operated gate opener (POGO), which reduces the time and effort to open and close the gates, is available and uses a solar power supply. It avoids a user having to cross and re-cross in order to close all the gates and reduces the risk from gates being left open and of being unnecessarily on the railway. This supports a Network Rail theme of improving safety by making crossings easier and more convenient for users.
The Supplementary Audible Warning Device (SAWD), supplied by Covtec, detects when a train is approaching by using a radar sensor and reports via a wireless link to provide an audible warning at the crossing, which sounds like a train horn. The system is solar-powered and is currently a safety overlay to existing protection arrangements. It provides an interim solution for locations where local ambient noise or the night time quiet period reduces the effectiveness of the train horn.
Network Rail, however, is developing Project Meerkat, (dependable warning for footpath crossings). The requirement is for a system with safety integrity, such that it could (subject to gaining safety approval) replace whistle boards and the requirement for level crossing telephones at footpath and similar crossings. A functional specification has been prepared and Network Rail is to invite industry to submit proposals.
Where the telephone remains as the sole means of protection, it should be able to report its condition and operation with a sufficient level of integrity appropriate to the safety function provided.
Automatic level crossings
As the name suggests, there is no intervention between the level crossing and the user or operator. There are three (soon to be four) main types:
- Automatic Half Barrier Crossing (AHB);
- Automatic Open Crossing Locally monitored (AOCL) and AOCL with Barrier (AOCL +B);
- Automatic half Barrier Crossing Locally monitored (ABCL);
- A new concept level crossing (AHB+) that will, when available, be an automatic level crossing.
AHB crossings, when used correctly, are a very efficient method of crossing with respect to minimal barrier down time and inconvenience to road users. However, AHBs are not linked to the signalling system and there is no monitoring to determine if the crossing is clear, nor signals close by to stop trains if the crossing is occupied. This is why full barriers are not provided in order that users of the crossing can’t be trapped between barriers and can exit the crossing before a train approaches. Unfortunately, this increases the risk of misuse with users weaving around the barriers.
Network Rail is developing a new concept crossing, known as AHB+. This is a cost-effective solution to reduce pedestrian and vehicle weaving risk without the need for a full rebuild of the crossing system.
To address the risk to sight-impaired users, provision is being considered to alter the audible warning devices so they to continue to sound once the barrier descends and until it raises. AHBs will not normally be renewed like for like in anything other than exceptional circumstances, although alternative designs may result in increased delays to road traffic.
AOCL crossings are locally monitored by the train driver – a white light shown on the approach to the crossing confirms that the road lights are working. The approach speed is such that trains should be able to stop if the crossing is occupied. However, users can still pass through the lights and onto the crossing just as the train arrives.
A significant number of AOCL level crossings have had an entry barrier added (AOCL+B) to provide another layer of protection, and this will continue where there is a safety/economic justification and engineering feasibility. Some sites have been constrained by surrounding buildings and topography, rendering barriers of any kind impossible without land purchase, demolition and remodelling.
Controlled level crossings
The level crossing operation is controlled / supervised by a signaller or level crossing operator. These include:
• Manually Controlled Barrier Crossing (MCB);
• Manned Gated Crossing (MGC);
• Train Staff Operated Crossing;
• MCB with Closed-Circuit Television (MCB-CCTV); and
• MCB fitted with Obstacle Detection (MCB-OD) – with minimal supervision
With MCB-OD, radar is used to determine that the crossing is clear before the barriers are lowered, then another sweep is carried out to check that the crossing is still clear and nothing has been trapped inside the barriers. Only once this has been confirmed are the protecting signals allowed to clear. Network Rail enhanced the design with a LiDAR (light detection and ranging) system to confirm that a fallen person or small child would still be detected.
Numerous installations of MCB-OD have been installed and the concept has been a success. The preferred option is to implement automatic obstacle detection at MCB, MGC and MCB-CCTV crossings at the time of renewal or during major re-signalling works or re-control, where supported by a business case.
Relay-based interlockings have been used to control MCB-OD crossings. However, this has resulted in some unreliability and timing issues due to the additional relays required. In future, Network Rail is keen to introduce PLC-based interlockings and approved products are already available from a number of suppliers. PLC (programmable logic controller) interlockings can be located in smaller lineside locations rather than large equipment rooms. This will result in both a capital and maintenance saving, as well as power savings for heat and lighting.
Often, when a level crossing renewal is planned, the risk assessment prior to commencement of design may identify that an upgrade to the method of protection is required, for example the replacement of an AHB with an MCB-OD or similar. However, this should take into account that the barrier down time will be greater; which could increase complaints by road users and the risk of misuse – nothing is straightforward when designing level crossings!
Network Rail is commencing the development of the next generation of MCB-OD detection systems. This will be based on the lessons learned from the first generation and the company is planning to invite industry to submit proposals for the detection technology that may, for example, include video-analytic capabilities. Ed Rollings was very clear that Network Rail is open to the type of technology proposed and will simply be specifying the performance, reliability and safety requirements.
Some local residents can be concerned that removing the crossing operator increases risk. However, technology solutions are not prone to boredom, forgetfulness, overlooking trains or other deficiencies of the Mk1 human brain.
The Equality Act 2010 replaced a range of legislative instruments with one piece of legislation covering a wide range of different characteristics. The Act codifies the need to consider the likely or actual effects of policies, programmes and developments on different sections of society.
A key element in implementing the Equality Act is the Public-Sector Equality Duty (PSED), which requires public bodies to consider all individuals in shaping policy, in delivering projects and services and in relation to their own employees. Network Rail is, therefore, exploring cost-effective ways of making level crossings equally accessible to all potential users to meet the requirements of the Act where applicable.
Whenever Network Rail considers the opportunity to close a level crossing (either by extinguishing the rights of way, providing a diversionary route across the railway, or replacing the level crossing with a bridge or underpass), it first completes a Diversity Impact Assessment. This considers the diverse range of needs that the local community has in relation to mobility, sight and hearing, and the impact that closing the level crossing will have on its users.
New level crossing products shall be designed for reliability and developed to support the Reliability Centred Maintenance (RCM) approach, with intrusive routine maintenance reduced to a minimum. In locations where the business case for a mains power supply is not viable, systems need to be sustainably self-powered.
Home Office-approved fixed red light digital safety camera systems, and the issuing of fixed penalty notices, can be used to discourage users from ignoring warning lights, and low-cost video monitoring systems for non-safety critical applications are also being considered.
Commercially available low-cost barrier systems and other means of protecting crossings, including the sourcing of additional suppliers for level crossing barrier operation machine,s are being investigated, together with technologies to reduce the risk of pedestrians or vehicles being struck by lowering barrier booms.
Modern self-sustainable power supplies, low-power LED technology and wireless mobile data are key for solutions specifically for remote or rural sites. Level crossing signage requires more intuitive pictogram style signs to reflect the needs of a diverse range of users, including people whose first language is not English, and to improve their understanding of level crossing signage.
Investigating the interlocking of highway road traffic lights with the signalling system at areas where drivers running red lights is a known problem, or at high volume road junctions near to level crossings, is under way. The team is also engaging with the automotive industry to seek out opportunities to benefit level crossing management by providing warning information directly to car drivers, such as through the ‘connected car’ programme
Technology solutions that automatically monitor any change in the usage at level crossings will allow the gathering of information useful in managing the increased and/or modified risk. These can be installed as required, or provided as standard if they can be provided cost effectively.
Train arrival prediction systems to dependably predict the time of a train’s arrival at a level crossing have been available for some time, but performance is variable and not consistent enough to provide ‘safe to cross’ information. The Network Rail strategy is to continue to develop a means of accurately and consistently predicting the arrival of trains at crossings that overcomes the limitations of current systems. Network Rail is exploring the possibility of using GPS in conjunction with other methods to achieve this.
Finally, to provide yet another initiative to improve level crossing systems, Network Rail is sponsoring a postgraduate doctoral research student at the University of Birmingham to research level crossing protection using advanced academic research techniques. The tools arising from this work will inform technology developments, selection and strategic investments in future developments.
Note on terminology: ‘Deliberate misuse’ is reserved for those events where intentional behaviour has been confirmed. Otherwise, instances of not complying with warnings and signs are categorised as ‘human error’.
This article was written by Paul Darlington