From Major Accident Hazard to Maintenance Task
Tracing the chain that links a hazard study to a work order, and where that chain usually breaks.
13 min
Activities: 0 of 2
Asset integrity is often managed as a set of separate activities: inspection, maintenance, testing, repair. Each has its own team, budget and software. The risk is that nobody can show how a particular work order protects against a particular major accident. An integrity system you can defend is one where that line can be drawn in both directions.
The chain runs like this. A hazard study identifies a major accident hazard (MAH), such as a loss of containment of flammable gas leading to a vapour cloud explosion. A bow-tie or similar analysis identifies the barriers that prevent or mitigate it. The equipment that makes up those barriers is classed as a safety critical element (SCE). Each SCE has a performance standard. Inspection, maintenance and test tasks are written to show the performance standard is met. Results are reviewed, and failures are corrected.
The integrity chain
- 1
1
Hazard identification
Hazard identification (HAZID) and hazard and operability (HAZOP) studies identify credible major accident scenarios.
- 2
2
Barrier analysis
Bow-ties or layer of protection analysis show which barriers each scenario relies on.
- 3
3
Safety critical element list
Every item of equipment that forms part of a barrier, or whose failure could cause the event, is listed.
- 4
4
Performance standards
What each element must do, to what level, and how this is verified.
- 5
5
Assurance tasks
Inspection, test and maintenance routines in the maintenance system, each linked to a performance standard.
- 6
6
Results and correction
Test results compared against acceptance criteria; failures raised, assessed and corrected.
Where the chain breaks
Worked example: tracing a deluge valve
Take a deluge valve on an offshore separator module. The bow-tie for "gas release from the separator" shows deluge as a mitigation barrier against escalation by fire. The SCE list includes the fire water pumps, ring main, deluge valves and nozzles. The performance standard for the deluge valves states that each must open on demand from fire and gas logic, and that water must reach the most remote nozzle at a stated minimum pressure within a stated time. The maintenance routine must therefore test opening on a signal from the logic, measure time to water at the nozzle, and record residual pressure. A routine that only "exercises the valve manually" verifies none of these.
Worked example: tracing backwards from a work order
An auditor picks a routine work order: "Inspect pipe supports, area 3." She asks which performance standard it supports. Nobody can say. On investigation, the area contains a high-pressure gas line whose supports are critical to preventing fatigue failure at small-bore connections. The supports were not on the SCE list. The audit finding is not about pipe supports; it is that the element list does not reflect the hazards.
A missing link
During a review you find that the high-integrity pressure protection system on a gas export line is in the bow-tie as a key barrier, but its proof test routine in the maintenance system is not tagged as safety critical. It has been deferred twice by planners who did not know its status.
Rebuild the chain in order
- 1Identify the barriers for each scenario
- 2Link assurance tasks to each performance standard
- 3Write a performance standard for each element
- 4Review results against acceptance criteria
- 5List the safety critical elements
- 6Identify major accident scenarios