ACCIDENT INVESTIGATION USING THE TRIPOD METHOD

Abstract

Accidents provide important learning points to prevent future mishaps. In its approach to accident prevention one technique that Shell uses is the Tripod accident analysis tree.

In the past there has been a tendency in all organisations to blame accidents on the people who suffer them and to see unsafe acts as arising from the stupidity, carelessness or recklessness of particular individuals. Accident investigations frequently cite 'human error' as the cause and many people still see such errors as the beginning and end of the accident sequence. The three more tangible components of such accident investigations are unsafe acts, failures of defences and the accidents themselves. Such unsafe acts or 'active failures' are often seen as the most important part of the accident story. In reality the story has its beginnings much earlier in areas such as Design, Procedures, Training or Maintenance Management, areas known as 'latent failures',

To better understand the underlying causes of accidents, in 1988 Shell commissioned a behavioural research program called Tripod. Tripod does not address unsafe acts and hazards directly, but places them in context in a model of an accident which seeks to define the underlying organisational factors that provoke unsafe acts, and that allow errors to turn into accidents. The research has established that latent failures may be categorised into a small number of disruptive processes. These latent failures are incorporated into an accident diagram in the form of a 'tree', and improve our understanding of the causes of accidents, and direct our recommendations for improvement into areas that are likely to be the most effective.