A novel framework to model the chronology of incidents is presented—depicting the relationship of initiating events with the various regulating and protection systems of the process—eventually leading to consequences, varying from zero to high severities. The key premise is that the departures and subsequent returns of process and product quality variables, from and to their normal operating ranges, are recognized as near-misses, which could have propagated to incidents. This leads to the availability of vast near-miss information recorded in distributed control and emergency shutdown systems databases that monitor the dynamics of the process. New performance indices, which utilize this abundant information, are introduced to conduct quantitative and qualitative (absolute and relative) assessment of the real-time safety and operability performances of an industrial fluidized-catalytic-cracking unit (FCCU) at a petroleum refinery. Also, new techniques for abnormal event tracking and recovery-time analysis are presented, which help to identify the variables that experience operational difficulties. It is shown how this information can be used to suggest improvements in the alarm-system structures for the FCCU.
Keywords: incident investigation, alarm databases, fluid catalytic cracking unit, refinery, safety improvement, dynamic risk analysis