Mainline - Network Rail

Overview: we scrutinised Network Rail's health and safety management systems (SMS) closely, as it moves from 'ad hoc' and 'standard' to a more 'predictable' level of management maturity. Overall its management maturity improved marginally over 2014-15. 

We found significant variations in RM3 elements across routes, which indicates that Network Rail had not yet implemented its SMS consistently or shared good practice effectively.

Successes this year included the continued low level of Potentially High Risk Train Accidents (PHRTAs), which are a credit to Network Rail's maturing leadership and collaboration with train operators and its safety 'deep dive' reviews and audit. There were no fatal or major injuries and 23 minor injuries from the 41 mostly minor train accidents in 2014-15 – the lowest levels of harm for a decade. However, there was an ongoing rise in freight train derailments, which in part, were caused by substandard track geometry and its dynamic interface with freight wagon suspension and uneven wagon loads.

As our enforcement action shows, we found too many significant examples of failures to identify or control risks to the workforce effectively by Network Rail and its contractors, particularly around construction activities. Network Rail's recent in-depth reviews have helped to improve understanding of specific risk areas, their causes and how effectively they are being managed. These lessons must now be learnt and used.

Evidence: Network Rail must now look at the evidence from our RM3 assessment where it received only a level 1 'ad hoc' or level 2 'managed' and target improvements. 

Level crossings

Overview: the harm posed by level crossings continued to remain low, as it has since 2010-11, but overall harm to crossing users increased 22%, primarily because of the two additional fatalities compared to 2013-14. 

However, the high level of pedestrian incidents over recent years highlights the need for the industry's focus on enhancing crossing users' intentional or unintentional misunderstanding or misjudgement of crossing risks (the House of Commons Transport Committee's level crossing safety report issued in May 2014 recommended use of 'deliberate misuse' rather than 'misuse' to differentiate between wilful negligence and impaired human decision-making). There was not enough progress in reducing overall harm, but there was a decline, down three to seven, in vehicle collisions with trains at level crossings. 

Of the 5,974 crossings on the mainline network in June 2015, 72% are 'passive', where the user makes the decision about whether to cross, 14% are manual crossings, where there is some form of control and 14% have some form of automatic user warning controls. Footpath crossings form the biggest level crossing fatality risk to pedestrians.

It is important that risk reduction momentum is maintained, including the focus on crossing closures, down-gradings and improving users' understanding of the risks. Further strategic improvement will be driven by DfT's implementation of the Law Commission's level crossing safety report recommendations originally made in 2013.

Evidence: we focus on level crossing safety because of the high levels of potential harm they present, especially to crossing users. Level crossings account for 8% of overall modelled mainline train accident risk. 

There were 10 level crossing fatalities, eight involved pedestrians, one was a car occupant and one a motor cyclist. There were five other non-fatal collisions between trains and road vehicles at level crossings, down three, and reductions in the levels of recorded 'near miss' events. This continues the lower level of incidents involving road vehicles since 2010-11. 

The trend in level crossing pedestrian fatalities has remained high over the last few years, as those involving road vehicle occupants has declined. Modelled risk shows that of the harm at crossings, 57% involved pedestrians and 30% involved road vehicles being struck by trains. Reported incidents of near misses between trains and cyclists or pedestrians at level crossings have reduced slightly.

Over Control Period 5 (2014-19), our inspection will focus on passive crossings, as these pose the biggest risk to pedestrians. Central to future risk reduction will be the roll out of more cost effective active warning technologies to passive crossings, such as footpath, bridleway and user-worked crossings. There was a slight rise in pedestrian fatalities at passive crossings over the last two years and active technologies should help reduce this risk. We challenged Network Rail's failure to follow its change management process when altering the specification of new obstacle detector crossings. This resulted in the retention of low-level obstacle detection at crossings.

Other effective initiatives included the introduction of an improved individual crossing risk assessment process and the implementation of new technology. As well as the use of 15 British Transport Police (BTP) operated mobile safety vehicles, the development of red light safety cameras at 10 crossings to deter deliberate crossing misuse, and the installation of audible warnings for pedestrians at more than 100 crossings.

Network Rail is developing a strategy to improve risk management at level crossings. It includes consideration to make all passive crossings effectively active by enhancing them with active risk control. We required Network Rail to improve its process to focus on qualitative risk assessment by its crossing managers. This is improving the industry's understanding of crossing use and effectiveness of customised mitigation measures.

Crossing closures and downgrading form an important part of this ongoing risk reduction strategy: 383 are due for renewal and 345 crossings with wig-wag lights will be upgraded to LED lights. In 2014-15, Network Rail closed a total of 118 crossings, of which 25 were closed using the £74m of the ring-fenced CP5 funding. This will achieve 21% of the planned 25% risk reduction. Network Rail plans to close around 250 crossings over CP5 using ring-fenced funding. The remaining funds will be used to commission new technologies at user-worked and footpath crossings. 

We continue to work closely with RSSB by contributing to their research to reduce risks to pedestrians at crossings and to enhance signage and warning systems at private and public crossings. 

We processed 114 level crossing Orders from Network Rail, as crossings were renewed or upgraded during new signalling schemes – this is a high number. These should deliver sustained long term benefits. We are in the process of assessing a further 150 crossing Orders.

Infrastructure risks

Overview: some progress was made in improving drainage asset knowledge and in the stewardship of bridge, tunnel and viaduct civil assets, but Network Rail needs to do more to manage track geometry and to ensure the long-term safety and sustainability of all its assets.

Drainage: our inspections found improving mainline drainage asset knowledge, but it has been too slow. We found variations in the approach, quality and completeness of different routes' drainage management plans, which are needed to address the drainage asset under-investment legacy. Network Rail must do the necessary work to inspect and maintain its drainage assets.

Track: Network Rail's current approach is largely based on track renewal and refurbishment work to deliver long-term improvement and reactive routine maintenance work to correct track geometry faults. It needs to be better and more sustainably managed. Nationally, we remain unconvinced that Network Rail has done sufficient analysis of the accuracy of its work banks to enable us to check it has sufficient resource (labour, access, material, equipment) to maintain its asset. 

However, at this stage we are satisfied that immediate safety risks arising from poor track geometry is being controlled, but in an inefficient and largely reactive way that sometimes does not address the underlying causes of faults and misses opportunities to address identified weaknesses. This increased the reliance on routine inspection and reactive maintenance activities to manage risk – see pages 31-32.

Switches and crossings: our November 2014 enforcement on Network Rail demonstrated that while good central review and development of technical solutions supported by clear process is important, equally important is their implementation in the field, supported by suitable mentoring and monitoring activity. Our future inspections will focus on the potentially increased pressures on maintenance delivery units as a result of the under delivery of planned renewals and refurbishment, and shortfalls in mechanised maintenance, such as tamping and stone blowing.

Earthworks: Network Rail has refined its contingency arrangements in the event of severe weather, but this is not a sustainable long-term response. There must be an overall improvement in earthwork asset condition. The first step has to be better knowledge of its asset condition.

Structures: there was a growing backlog in structures examinations. Network Rail must halt this trend and ensure it is adequately resourced to inspect the condition of its civils portfolio -the physical features, such as bridges, tunnels and earthworks, on which railways are built. 

Off-track and vegetation management: our inspections found that vegetation conditions and its management varied across routes. Following our interventions over 2011-13, we found that the impact of the leaf fall season in 2014 was now mostly managed consistently, but was off the pace in discrete areas.


Drainage: following on from our work in 2013-14, we served a national improvement notice on Network Rail in February 2015 to drive improvements in drainage capacity and degraded performance, because they increase potential landslip, and therefore train derailment risk.

Track: crucially, further reductions in repeat track twist events are largely dependent on renewal and refurbishment volumes being maintained and supported by more effective maintenance interventions. Keeping the right balance between maintenance and renewal activity will be essential. If renewals are not delivered to plan, it puts increasing pressures on maintenance delivery units to identify and manage defects. Repeated monitoring and repair activity is inherently less reliable than permanent repair. 

Switches and crossings: we are monitoring the roll out of Network Rail's new design of tubular stretcher bar, developed as a result of the Grayrigg derailment in 2007, through field inspections of employee competence. 

Earthworks: we routinely monitor the development of Network Rail's five year activity plan aimed at improving asset management by focusing on risk assessment and reviewing changes to the earthworks management standards.

Structures: we recently looked closely at the management of advertising hoardings attached to structures, the inspection and assessment of operational property, the safety of metallic structures with concrete encased beams and the maintenance of signal posts. In November 2014, a signal fell across the track on the Western route and was struck by a high-speed train but caused no injuries. Another collapsed after corrosion, on the Anglia route in April 2015. We are also focused on ensuring appropriate risk control measures are put in place to manage known weaknesses in station footbridges.

Off-track and vegetation management: Network Rail has surveyed vegetation management on all its routes and we are awaiting its results. In the interim, we have continued to press Network Rail to develop deliverable plans and will monitor its revised business plan for any indications of reversing planned volumes. Our assessment of the sector's low-adhesion prevention and rail-head enhancement work over autumn 2014 found that vegetation management had improved on previous seasons. It remained an issue in some areas, such as long stretches in Anglia, which led to a heightened potential safety risk and knock-on effects on service performance.

Safety by design

Overview: our work is focused on duty holders' failures to take opportunities to eliminate or reduce risks at the design stage, especially during new-build and refurbishment projects. We found evidence in large infrastructure projects of a failure to consider optimal risk reduction, and preferably its elimination, at the design stage. Our enforcement over 2014-15 showed the industry remains some way off the pace in applying the safety by design principles. 

Evidence: Network Rail's major infrastructure projects present significant opportunities to design out or reduce risks. Some safety enhancements may only be realistic if implemented at the starting point of an infrastructure's life-cycle. 

We took enforcement on Network Rail in April 2014 because of its failure to have suitable and sufficient assessments of the risks to passengers, public and staff at two Western upgrade sub-projects. And again in January 2015, due to the failure of its North West Electrification phase 1 project to comply with electrical safety standards, to which it had committed. 

We propose to formalise our enforcement role to enable us to deal more effectively with safety by design issues at an earlier stage in major new-build infrastructure projects. With that in mind, we have increased our engagement on the safety aspects of HS2.

Infrastructure worker safety risk

Overview: overall workforce harm declined 3%, but when normalised by the 3% decline in workforce hours worked, showed no significant change. Undoubtedly, the big challenge for the industry is culture and behavioural change for infrastructure workers to help implement planned safety improvement initiatives. 

Evidence: There were four workforce fatalities, one of which was on metro infrastructure. This compared to three mainline fatalities in 2013-14. Of those four, two were electrocuted - one fell on to the third rail at a train depot and one contacted the overhead line (non-mainline) – we are currently investigating both. Two were killed in occupational road incidents. 

There were some minor reductions in non-fatal injuries. Of the 175 workers who suffered major injuries, 100 involved infrastructure workers. Of the overall harm to the workforce, 40% involves infrastructure workers who work on or near the running line - a disproportionately high level of harm given the relatively low number of infrastructure workers.

The ongoing national roll out of the safe work (the new control of work permit procedure and safe work leader) initiatives undoubtedly sets the right ambitions to improve infrastructure worker safety. We recognise that this is the solution Network Rail has identified to secure long-term cultural change. Therefore, we have focused our efforts on their development and implementation.

Occupational road safety

Overview: Network Rail's recent focus on reducing risks from the operation of its road fleet appears to show some benefits, with reductions in incidents without injuries and those with injuries or where the emergency services were called. 

Evidence: two infrastructure workers were killed and four received major injuries in occupational road accidents while working. There were 104 minor injuries, of which 24 resulted in staff taking three or more days off from their usual duties. Overall harm reduced 7%. Most occupational road incidents involved Network Rail's infrastructure workers and contractors. 

There have been five occupational road fatalities in the last three years. It is estimated to represent about 4% of the overall harm to the workforce. There were several off-duty fatal road accidents involving railway employees driving home after long shifts, such as two recent multi-fatality traffic accidents in Scotland and Western involving off-duty railway contractor staff. 

The inquest into the deaths of two rail welders in a road traffic collision at Newark in 2013 highlighted the problems of fatigue and driving risk. RSSB and wider industry partners have been active in promoting good practice. Their work and the Health and Safety Executive's Driving at Work guidance provide useful risk management resources.