Reliability Characteristics of a Safety Instrumented System via Markov Analysis

The Markov technique for determining the Reliability Characteristics (Probability of Failure on Demand, availability, Mean time to repair and Mean time between failures) of the Safety Instrumented System (SIS) on Offshore Metering and Custody Transfer Facility in Campeche Bay Mexico is presented in this work. Two important factors were considered based on the ISA TR84.0.02, Part 4: a) Common Cause Failure and b) the diagnostics capabilities were taking into account. The Markov technique methodology employed in order to assess and determine the Reliability characteristics of SIS was compared with Fault Tree Analysis previously analyzed, Part 3 of ISA TR84.0.02; for which the hazard and risk analysis mandated a SIL 2 for the SIS with two sets of sensors where each individual sensor can shut down the operation of the facility. Aspects like voting, redundancy or diversity bring about a full Markov model of a SIS, but it usually consists of intermediates states not significant for the assessment of the reliability characteristics. The quantitative results will mostly depend on the direct transitions to the fail-safe and fail-dangerous states. As a result, in most cases it was not necessary to present a fully developed model.