Fire and Gas detection and suppression systems have long formed an important part of the Loss Prevention toolkit for plants that process highly hazardous chemicals, both flammable and toxic. While every effort is made to prevent loss of containment of flammable and toxic chemicals, history has shown us that not only do these events occur, but those that are not prepared for them suffer major business?wrecking losses from which they may never recover.
Trends in Loss Prevention clearly show a move toward more and more quantitative methods. In general, this is a valuable trend because high risk situations cannot be simply dismissed as not credible due to the presence of safeguards that are qualitatively deemed appropriate. Furthermore, quantitative methods are helping to ‘right size’ risk reduction measures such as engineered safeguards, applying more resources to higher risk situations where they can provide the most benefit.
Unfortunately, the new methods that being employed are often misunderstood and misapplied to situations where their use is not appropriate. At present risk?based fire and gas detection is one of those situations. Using a very basic and very simple tool – such as Layer of Protection Analysis (LOPA) – to make decisions about the design basis for fire and gas detection and suppression systems, can be misleading and erroneously result in the decision not to implement such systems. The reality is much more complex. LOPA fails to correctly analyse risk mitigated by fire and gas systems for three key reasons-
- LOPA is a method that analyses the frequencies and probability of events where an event happens or does not, it is not suited to analysing the partial reductions in consequence the fire and gas detection and suppression systems provide
- LOPA is scenario?based and requires a priori knowledge of all the causes of an event, a priori knowledge of the occurrence of leaks and fires is not possible, and as such, statistical methods to determine fire frequencies and locations is required, which LOPA cannot process
- LOPA utilizes probabilities of failure to determine whether or not risk is tolerable – Simple use of probability of failure of a fire and gas detection and suppression system will lead to grossly inaccurate results as the preponderance of failures of these systems is the result of detectors not being located in the path of the release, as opposed to failure of the devices themselves.
There is an old saying in motor racing “When things go well it is good but when it goes wrong it goes terribly wrong”. The same apply in fires and/or explosions in petrochemical plants. We can name a few like Piper Alpha and now recently the Deer Park tank farm in Texas. Although we cannot state that none of these events would have happened if there was a gas mapping study done using the quantitative method. It will for sure be advantageous if a study with the percentage effectiveness required by the client and according to industry acceptable standard is done to mitigate these devastating incidents to protect lives, equipment and the environment.
Pinnacle Instruments now offers facilitation in conjunction with world leaders in Fire & Gas Mapping to render a complete quantitative package with regard gas mapping and supplying Honeywell gas detection.
Acknowledgements to Edward Marszal – Kenexis
Written by: Poon Schoeman