Investigating a serious accident or near miss
After a serious event, it is tempting to stop at the last action before it happened. My investigations go further back, to the organisational choices that made the event possible.
Which events?
- A serious or fatal accident.
- A high-potential near miss that could have killed someone.
- A series of events that keeps recurring despite action plans.
A serious accident never follows a simple mechanism.
Beyond the classic methods
Cause tree analysis, the 5 Whys, the Ishikawa diagram and its 5Ms, bow-tie analysis and FMEA each have their uses. They shed light on an event from a given angle. Yet all of them follow linear chains, from one fact back to the one before, as if a serious accident worked like a simple mechanism.
A serious accident grows out of many interactions. It involves compromises judged reasonable, deviations that became normal through tolerance, power relationships, and what gets said or left unsaid within the organisation. Classic tools miss this complexity. They steer the analysis towards the questions they know how to handle, rarely towards the uncomfortable ones.
My investigations step outside these frameworks. They draw on AcciMap, which places the event within the whole system, from the workstation up to leadership decisions and the regulatory framework. They also borrow from the methods of organisational sociology, and from the way inquiries into major disasters trace back to the decisions that prepared them. This is what brings the real precursors to light, where the usual tools stop at the last action.
How the investigation runs
- ScopingTogether with management, we set the scope, the people involved and how the findings will be shared.
- Reconstructing the factsInterviews, documents and site visits show how events unfolded as the people involved experienced them.
- Tracing back to the organisationI analyse the decisions, trade-offs and practices that prepared the event, often long before the day itself.
- DebriefThe findings are presented to management, with recommendations that also address the organisation.
The report concludes it was human error. What do you do?
It's the most comfortable option. It holds until the next accident, often in the same place.
I look for what made it possibleWhy an outside view
An internal investigation often stops where the responsibilities of those who commissioned it begin. An outside investigator can ask the questions nobody dares to ask internally, rigorously and with respect for the people involved.
The investigation can be commissioned by management or by the board, when it wants an independent analysis.
Its purpose is to understand and prevent. It does not replace judicial or administrative investigations.