Interval Reliability for Initiating and Enabling Events

This paper describes generation and evaluation of logic models such as fault trees for interval reliability. Interval reliability assesses the ability of a system to operate over a specific time interval without failure. The analysis requires that the sequence of events leading to system failure be identified. Two types of events are described: 1) initiating events (cause disturbances or perturbations in system variables) that cause system failure and 2) enabling events (permit initiating events to cause system failure). Control-system failures are treated. The engineering and mathematical concepts are described in terms of a simplified example of a pressure-tank system. Later these same concepts are used in an actual industrial application in which an existing chlorine vaporizer system was modified to improve safety without compromising system availability. Computer codes that are capable of performing the calculations, and pitfalls in computing accident frequency in fault tree analysis, are discussed.