Ffestiniog Incident

The RAIB released a report into a level crossing incident on the Ffestiniog Railway a while ago. The incident itself occurred back in January.

As with several other reports from the RAIB, it references previous incidents, in this case that there have been three at this level crossing. As well as other incidents that offer similar learnings,

The summary makes several comments regards system safety and rules compliance. It suggests that rules compliance will ensure safety.

In the case of the Ffestiniog, the driver has two sets of operating conditions to contend with – when passenger trains are running, a signal works, and when passenger trains are not running, there is no signal – a classic error-provoking situation – same circumstance (a level crossing) four different rules (one for braked and another for unbraked trains, then another when the railway is carrying passengers, and another when the railway is not carrying passengers). A 75% probability of failing?

Whilst there is some degree of truth to the statement at “rules compliance will ensure safety”, blind compliance does not necessarily mean that the system will be safe. As once quoted to me, the rule book may prevent the same accident from happening twice, but rarely are the factors which leads to similar accidents the same. Rules have been developed based on past experience of the organization: Not necessarily incidents involving the individuals who, in the real world, are required to interpret and apply them. The importance of engagement and consultation with front line workers when systems processes and rules are developed and reviewed cannot be understated (in Australia, this is a requirement under Rail Safety National Law, as well as WHS legislation).

The key message, in my view, is contained in the second line of the first dot point – “the reason why a rule exists is not always obvious, and may have been forgotten as time passes”. In complex systems, this point raises two key themes;

1 – are people trained to comply with the rules, or are they trained to understand how the system is intended to work, and what the “rules” mean? And

2 – has the context for which the relevant rule was written changed? If it has, then the rule needs to be reviewed.

A topical example is ECP – a very different system to traditional Westinghouse air brake.

As Astronaut Karol Bobko notes – “the fastest way to get yourself killed on a manned space flight is to not follow standard operating procedure – the second fastest way is to get yourself killed is to always follow standard operating procedure”.

Have a great weekend, and thanks for reading

.Note - Comments reflect the Personal Professional opinion and experience of the Author, not those of his employer or other agencies