The Weekly Reflektion Week 03 / 2020

The subject of this week’s Reflektion is the Ladbroke Grove rail crash in London, England.
Reflekt will be organising a breakfast seminar in April 2020 where we will consider the three ‘R’s’, Reliability, Robustness, Resilience. Watch out for more information on our web site.

Do you have an effective action tracking process?

Have performance targets diluted the perceived importance of safety in your operation? 

On the 5th October 1999, just 2 years after the Southall rail crash (see Reflektion week 48/2019), a Thames Trains train travelling west from Paddington passed a signal at red and collided with a Great Western high speed passenger train (HST) travelling in the opposite direction from Cheltenham to Paddington. They collided at Ladbroke Grove, 3,2 km from Paddington, with a relative speed of 130 mph (210 km/hr). 31 people, including both train drivers, were killed and and 258 injured. The diesel fuel from the engines ignited, causing a fireball which in turn caused separate fires in the wreckage.

The westbound train should have stopped at the red signal and waited for the eastbound train to pass. Both drivers died in the accident so the reasons why the westbound train did not stop are not known. An inquiry found that driver training was deficient, with the westbound train driver, in addition, having qualified only 13 days previously. British Rail was privatised, between 1994 and 1997, and the driver training period had been shortened, and, the inquiry found, trainers did not follow the syllabus as they considered it ‘not fit for purpose’. The testing of trainees was found to be unstructured and unstandardized, with no clear pass/fail criteria. British Rail did not allow drivers with less than 2 years experience to drive through the Paddington area as it was perceived to be a challenging area.

Prior to the accident, the signal the westbound train failed to stop at, had been passed at red on 8 occasions in 6 years. The inquiry found that there were two main probable reasons for not registering the red signal. Firstly, the day of the accident had bright sunshine which was behind the driver, shining directly into the signal itself, which may have made it difficult to see. Also, as a result of electrification of the line in 1994, overhead electrification equipment partially obstructed the driver’s view of the signal.

All new or altered signals, according to procedures, should have been reviewed for sighting issues by a ‘signal sighting committee’. This had not been done for the Paddington signals. An audit in March 1999 reported this lack of review, and a follow-up audit in September 1999, one month before the accident, registered that nothing had been done. Railtrack were responsible for the track, signalling, tunnels and other infrastructure. The responsibility of Railtrack employee with formal responsibility for action tracking ended when the action had been accepted by the department responsible for execution.

By February 1998, 4 separate groups had been set up with the aim of reducing trains passing red signals. The inquiry described Railtrack as having an ‘endemic culture of complacency and inaction’. Decisions were found to have been delegated upwards in the organisation, with the management to whom the actions were delegated not closing them out. In addition, the culture was one of reactive management, not proactive management.

As in the Southall rail crash, the train was fitted with the automatic warning system (AWS) which required acknowledgement of alarms, such as passing a red light. A system called ATP, automatic train protection, actually engaged emergency brakes if such an alarm was ignored. This system would have prevented both Southall and Ladbroke Grove accidents. The westbound train was not equipped with the ATP system. An evaluation of the ATP system in 1996 concluded that the safety benefits were not great enough when related to its cost. After the Clapham Junction rail crash in 1998, a recommendation was made to install the ATP system in all trains. After this accident at Ladbroke Grove, this cost-benefit evaluation was re-visited, and the inquiry compared the ATP system with an upgraded AWS system termed TPWS, which enabled trains travelling under 70 mph (112 km/hr) to stop if a red signal was passed. This TPWS system was thought to be significantly less effective than the ATP. Although expressing considerable reservations about the effectiveness of TPWS, the inquiry concurred with its adoption over ATP.

The privatisation of British rail had the effect of fragmenting the industry, leading to numerous complex interfaces and problems with management, development and implementation of large-scale projects. This fragmentation also hindered common research and development projects across the different companies. The inquiry also found a focus on performance targets which diluted the perceived importance of safety.

In 2003, as a result of the poor rail safety record, the Rail Safety and Standards Board was established, and in 2005, the Rail Accident Investigation Branch followed. This clearly separated standard setting, accident investigation and regulatory functions for the railway industry.

Thames Trains were fined GB£2 million as a result of the accident.

Do you find any parallels in your organisation, for example, following up the results and recommendations from audits and reviews? Perhaps the follow-up of personnel training, both quality of the training provided and the verification that people do go through the training process could be improved.   

Reflekt AS