It is a fact universally acknowledged that a station platform is invariably in need of a good buffer-stop….
On the 31st of January 2013 a QR commuter train slammed into the end of platform barrier at the Cleveland street station, overrode it and ran into the station structure before coming to rest.
While the media and QR have focused their attention on the reasons for the overrun the failure of the station’s passive defenses against end of track overrun is a more critical concern. Or to put it another way, why did an event as predictable as this, result in the train overriding the platform with potentially fatal consequences?
As luck would have it the media pictures of the event provide most of the answer to that question. In the picture above you can see the concrete block that was used as the ‘buffer stop’ at the end of the line. The only problem of course is that it neither buffered (absorbed the train’s kinetic energy) or even stopped the train. What the block actually did was to rotate when struck by the train, thereby providing a ramp up which the train climbed and launching the whole mess into the station concourse. In contrast most modern stations, such as Helsinki below, provide buffer stops that are capable of absorbing the kinetic energy of the train without derailing it.
Interestingly the climbing effect has been recognised in train design for many years with the standard design response being the inclusion of anti-climber plates on the faces of adjoining cars that lock together when a train concertinas in an impact, to prevent cars from climbing over each other, and similarly anti climbers can also be incorporated into the design of buffer stops to prevent a train overriding the buffer.
What’s unfortunately quite clear from the post accident pictures is that no such anti-climbing mechanism was in place at the station. That lack, in combination with the failure mode of the concrete barrier ensured that the results of a ‘simple’ overrun would inevitably turn into a catastrophic and potentially lethal accident.
What’s not quite so obvious is what would have happened if the barrier had worked as intended and the train had simply smashed into a concrete wall. What would have been the injuries sustained by the passengers and train crew in that scenario? Again the consequences could have been potentially lethal.
The solution? Quite simply to install a properly designed buffer stop whose interface prevents climbing and which can absorb the kinetic impact of such a collision. None of which is new, the basic principles having been around in the rail industry for a long while. And of course it’s not as if this sort of accident hasn’t happened before in Australia.
Considering this incident from a design perspective this accident raises all sorts of interesting issues for Queensland Rail over whether the concrete barrier had been actually designed, how that design had been verified and whether the design had been carried out or supervised by a competent engineer as defined by the Registered Professional Engineers Queensland Act scheme (RPEQ 2002). A further question that this incident raises is what role, if any, did risk assessment play in the decision that a concrete barrier was appropriate in this application.
From a safety engineering perspective this is a great example of a hazard inadvertently introduced at an interface and is the sort of hazard that should be exposed during a System Hazard Analysis (SHA).The conduct of a formal analysis of interfaces for potential hazards is IMO a mandatory safety activity when an interface is also split between two separate organisational entities, in this case QR’s metro operations and the track infrastructure maintainers and providers.
Finally one might question the effectiveness of both Queensland Rail and the Queensland state rail safety regulator’s safety management in their apparently overlooking the published lessons learned in the Packhenam and Sandringham Overruns in Victoria. My somewhat jaundiced comment is that if this report was even looked at it was probably dismissed on the basis that what happens in another state jurisdiction could have little bearing in Queensland.
I’ll be interested to see what form the accident investigation takes and where it leads the ATSB investigators…