The analysis of incidents is often considered to be part of the investigation. This may make some sense, but has a number of potential problems. In particular starting the analysis before the investigation is complete can lead people to 'jump to conclusion,' which may result in them collecting evidence that supports a conclusion that may not be valid. Therefore, although they may well be carried out partly in parallel, it is useful to differentiate between investigation and analysis.
An analysis of an incident involves looking at the evidence collected to identify the causes of an incident. These causes are generally broken into two distinct categories:
* Immediate causes - features of premises, plant, substances, procedures and people that created a hazard or contributed to the incident. Often considered as unsafe conditions and actions
* Underling causes - failures of planning, risk assessment, control, cooperation, communication, competence, monitoring and review that resulted in the immediate causes being present and/or not dealt with. These are typically management and organisational failures.
As a result of the analysis it is important that recommendations are developed to address the underlying causes. This means it is not only the exact incident that can be prevented, but that a general improvement is safety can be achieved. To do this it is often necessary to consider previous incidents to identify any trends that indicate a wider problem than may be apparent from a single incident.
As with investigation, it is usually best if a team carry out the analysis. Once again competence in analysis tools and techniques should be held by the team.
Thursday, November 23, 2006
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment