PT Notes
The Impact of Risk Uncertainties on SIL Determination
PT Notes is a series of topical technical notes provided periodically by Primatech for your benefit. Please feel free to provide feedback.
The IEC 61511 / ISA 84 standard requires that safety integrity levels (SILs) be determined for safety instrumented functions (SIFs) that make up safety instrumented systems (SISs). SIL determination entails the evaluation of the risks of a process and their comparison with appropriate risk tolerance criteria.
Various forms of uncertainty enter into the calculation of risk. For example, the models used in risk analysis are imperfect representations of the real world and result in modeling uncertainty in risk estimates. The data used in the calculations are often sparse, or not known very well, and result in data uncertainty. Currently, few companies have plant-specific data. Often, generic data are employed with their larger uncertainties. In some cases, uncertainties in non-SIS failure data are large enough that analysts could choose a lower part of the range in order to avoid the need for a SIS. Clearly, this is undesirable, but the effort involved in qualifying SISs to IEC 61511 / ISA 84 can encourage that approach.
Comparisons of risk estimates with risk tolerance criteria must address uncertainties. The validity of any decisions on required SILs depends on their consideration yet uncertainties are routinely ignored by many practitioners. Commonly, practitioners use best estimate values for failure rate data in SIL determinations but such values may be an order of magnitude or more away from the wings of their distributions. Consequently, decisions based on risk values using such data may be seriously in error unless the uncertainties in SIL determination are addressed. Uncertainties should be addressed in using any SIL determination method.
For more information, you can contact Primatech or consult the article:
The Interface of Functional Safety with Process Safety and Risk Analysis, by Paul Baybutt, Process Safety Progress, Volume 32, Issue 4, pages 346–350, December 2013.
The article is available at: http://onlinelibrary.wiley.com/doi/10.1002/prs.11640/abstract.
Click here for information on a related training course.
Click here for information on a related software tool.
