Uncertainty handling in safety instrumented systems according to IEC 61508 and new proposal based on coupling Monte Carlo analysis and fuzzy sets

摘要: Abstract Safety instrumented systems must be designed, built and operated to meet tolerable risk level as required regulatory agencies. This requirement is closely related their probabilistic performance measures which are either average probability of dangerous failure on demand (PFDavg) or frequency (PFH: Probability Failure per Hour). The object this work the SIS performances evaluation taking into account uncertainties different parameters that come play: rate (λ), common cause proportion (β), diagnostic coverage (DC), etc. leads an accurate therefore safe assessment safety integrity (SIL) inherent functions performed by such systems. aim in keeping with IEC 61508 standard respect handling uncertainty. In paper we first explain detail approach for Afterwards, propose combines (i) Monte Carlo analysis (MCA) (ii) fuzzy sets. Indeed, method appropriate when representative statistical data available (using pdf relating parameters), while latter applies case characterized vague subjective information membership function). proposed fully supported a suitable computer code developed under MATLAB environment.