Automated safety and uptime analysis of safety instrumented systems

Programmable Electronic Systems are used in the process industry to perform very complex and sophisticated safety functions in a cost-effective, and flexible way. A typical Safety Instrumented System consists of sensors, logic solvers and actuators. There are many parameters influencing the performance of these safety systems, e.g., hardware, software, redundancy, voting, diagnostics or common cause and systematic failure aspects. Standards, such as the ANSI/ISA S84.01-1996 or the dIEC1508 provide guidelines to evaluate the performance of these systems. Markov modeling, a technique appropriate for such evaluation, also able to handle the uncertainty in data, is an approach which allows the instrumentation practitioner to model all the aspects that influence the performance of safety systems. A significant disadvantage for the instrumentation practitioner is the complexity of the development of Markov models. Usually this is a manual process, not only time consuming, but also tedious and difficult. Therefore, analysts very often choose other less flexible techniques. This paper presents a computer Program that generates Markov models, in an automated way, for safety systems consistent with the ISA draft Technical Report 84.0.02. The requirements and assumptions of the Program are demonstrated through a specific case study, the so called, 2-out-of-2 Programmable Electronic System. The results of the Program are used to document quantitatively the performance of the 2-out-of-2 system