3.1.1 The cause
A small digression will be made here to question the use of the word ‘cause’ and to reflect on the view expressed by Kletz (1988, p. 2). He argues that the word has an air of finality about it, and is concerned that finding the cause discourages further investigation. He cites an example that the cause of a pipe failure was corrosion – which suggests that we know why the failure occurred. He draws the analogy of the cause of a fall being gravity – suggesting that nothing more can be done about it and repetition of the incident cannot be avoided. While the causal statements may be true, they must not stop further questioning:
Was the specification of the material for the pipe correct?
Was the specified material used in practice?
Was the pipe used under the operating conditions assumed by the designers?
Did the designers ask for corrosion monitoring?
Was corrosion monitoring carried out in practice?
What was done with the results – did action follow?
Kletz further argues that the association of cause with blame introduces an atmosphere of defensiveness. The result of this is a ready acknowledgement that an incident may have been preventable by better design or operation, but there may be reluctance to attribute it to poor design or operation.
Whatever view of the causal structure of incidents is taken, the aim should be to look beyond the immediate cause and to identify more fundamental ways of avoiding the hazard and of improving the management system, especially through attention to multiple causes and their interactions.
As a further guide, here are some questions you may consider about incidents and the identification of their causes.
If the incident involved equipment failure, what was it that failed?
How can failure be prevented or made less likely to occur?
How can failure or its onset be detected early?
How can failure be controlled or the consequences minimised?
What does the failed equipment do?
What can be done instead of using this equipment?
What material was involved – leaked, decomposed, ignited, exploded …?
How can such an event be prevented?
How can such an event or its onset be detected?
What is the material used for in the process?
Do we need to have so much of the material present at any given time?
What alternative materials could be used?
What alternative processes are available?
Who could have carried out their role more effectively? Look at the whole life cycle of process or product – design, construct, operate, maintain, train, inspect, etc.
What could they have done better?
How can they be helped to perform better?
What function is served by the operation involved in the incident?
Why is it done?
What alternatives are available?
How could it be done differently?
Who else could provide the function?
When can it be done?
This checklist implies that there are many ways of preventing an incident being repeated. You may also perceive the interrelationships between many of the possible causes of an incident. Now we will explore some of the general issues in assessing risk.