Derailment inquiry exposes risk management shortfalls

A great divide exists between those who take proactive action to prevent undesirable events and those who just muddle along and wait until the bad thing happens before belatedly doing anything about it. Construction abounds with examples of both approaches, but unfortunately the outcome of the heads in the sand policy can be fatal.

The potential dangers of the simply reactive approach are clearly seen in the findings of a railway derailment inquiry in Scotland that killed three people last year. The Rail Accident Investigation Branch (RAIB) has produced an interim report into the derailment of the 06.38 Aberdeen to Glasgow passenger train that ran into drainage gravel washed onto the track at Carmont, Aberdeenshire, at 09.38 on 12 August last year, resulting in the deaths of the driver, a conductor and a passenger. All six others in the train were injured.

Other inquiries are being made jointly by Police Scotland, British Transport Police and railway regulator the Office of Rail and Road. Liability for what went wrong has still to be established, but the report raises serious questions that all companies and individuals working in the rail sector need to ask.

What has emerged so far reveals a systemic failure in the risk management and inspection procedures that underpin the maintenance regime of our ageing railway infrastructure, at least on that part of the railway network, but probably UK wide. In this incident heavy rainfall overnight had caused damage to many parts of the railway network in Scotland, but this had not led to any increased cautionary measures on the line this train ran on.

Network Rail has an extreme weather procedure but this required a weather hazard to have been notified the previous day if special measures were to be adopted the next morning. Despite news of heavy rain causing problems elsewhere there was no procedure in place for considering what this might imply for as yet not obviously affected areas.

A landslip had been reported on the line ahead and the train halted for a period, and resumed its journey at slow speed in the affected area before being allowed to pick up again to normal speed and then hitting the gravel washout at 73mph. The RAIB report says that a new drainage system had been installed at Carmont in 2012, (by Carillion, with design input from Arup & Partners Scotland) but only part of the system was recorded on Network Rail’s drain maintenance database. So part of the drainage system had not been inspected at all since 2012. The RAIB report says lack of an effective drainage inspection regime meant that crucial indications of future problems would not have been detected.

The drain’s design and installation are still being investigated but the Carmont report makes it clear that there is little resembling a joined up approach to the maintenance management of the drainage and associated earthworks on our railways. Railway engineers of old who prided themselves on their in depth knowledge of local infrastructure would shudder at this approach.

New procedures have been ushered in since the accident and no doubt the safety level has been increased since then. But there is no clear sign yet that the head in the sand approach that assumes all is well until events prove otherwise has been fully abandoned.

Nick Barrett
Editor