(111c) Determining the Effectiveness of the Human Safeguard

A team conducting a Process Hazard Analysis (PHA) reviews a vessel with a pressure alarm present – a common safeguard against many of the overpressure scenarios the vessel could experience. Everyone on the PHA team feels the alarm provides plenty of time for an operator to respond to any given overpressure scenario and continues with the PHA. This may be acceptable in some cases: the expected response time to that pressure alarm is sufficient for an operator to respond to a blocked outlet. But it could be a costly mistake in other cases: if a bypass on the fuel gas line is left open, the MAWP of the vessel is exceeded in under 60 seconds – not nearly enough time for the operator to respond before potential loss of containment.

This situation – where alarms are used as safeguards without proper vetting and consideration of operator response time to those alarms – is occurring more frequently throughout the industry. With the move toward more technology-driven and interconnected facilities, alarms have become one of the most prevalent safeguards in facilities. To appropriately use alarms as safeguards, sites must ensure that alarms have clear instructions on the correct response to the alarm and that there is adequate time to mitigate the scenario.

Without a clear idea of which alarms can or cannot be acceptable safeguards for given conditions or scenarios, facilities may leave themselves unprotected for scenarios deemed appropriately mitigated by a risk study. Alternatively, they could unnecessarily spend their budget installing more robust safety systems when operator intervention through alarm would be adequate.

This presentation will discuss ways to determine the amount of time an operator would have to respond to an alarm condition, ways to account for the human factors of alarm response, and examine appropriate best practices for the management of alarms as safeguards.