(47az) Identifying Early Indicators of Incidents through Near-Misses

Post-incident investigations show that there are several near-misses or precursors that occur before undesirable events in different industries (chemical, manufacturing, aerospace, IT, etc.).  Near-misses can be categorized into two categories – observable and non-observable near-misses – both of which indicate conditions which had the potential to cause incidents but did not actually result in much impact – due to sheer luck or existence of other protection systems.   Observable near-misses refer to events that are readily visible to operating personnel, such as equipment failure, whereas non-observable near-misses refer to events that are not casually visible to the operating team, such as special-causes related to control loops.  To improve reliability of systems and processes, it is of utmost importance for the risk management and reliability teams to not only identify and prioritize these near-misses, but also to learn from them.  Particularly, cross-industry lessons on near-misses can provide new insights in their identification and management. 

The presentation will demonstrate that having a strong near-miss management philosophy, system as well as culture, addressing both observable as well as non-observable events, help companies predict risks at their budding stages and prevent catastrophic events.  It will include case studies from different industries on the importance of such events and their timely identification for improvement of reliability, and process safety.  New quantitative techniques and methodologies to identify them systematically will also be discussed.