Over the past six to eight weeks we have seen a multitude of articles, interviews and other sources of information telling us how safe 2012 was for commercial aviation. According to each source and specific safety metrics, operators of large commercial jets enjoyed one of the safest years on record, if not the safest. In fact, the 2012 metrics show about a 50 percent improvement over 2011, in itself a record year. Everyone in aviation should be proud to achieve such performance. But, as aviation professionals, we know that challenges continue, especially because all the commercial jet accidents we saw in 2012 were preventable.
The accident rate is the ultimate safety performance indicator, and may in fact be an indication of a declining latent risk in operations, especially when analyzed over long periods. However, it is important to understand that precursors to accidents (safety risk) continue to exist in aviation. System failures and human errors sometimes produce “near misses,” which fly below the public radar, often just one or two variables away from being classified as accidents.
In other words, the absence of an accident does not indicate a lack of possibility. For the purposes of safety management systems, “safety” has been defined as “the reduction of risk to a level that is as low as reasonably practicable [ALARP].” Understanding that accident metrics are a measure of how the system performed during a specific period, but not necessarily of how much safety risk exists in current or future operations, safety professionals continue to focus on managing latent risk to the ALARP level.
In effect, the safety level is a measure of the risk posed by hazards in the aviation system, not necessarily a measure of past consequences of those hazards. As we have seen in the past, latent conditions may not manifest as an accident for a long time, and may not become readily visible until the accident happens. If we are truly moving to proactive and predictive safety management, we will need to enhance the ability to identify and manage low-probability, high-severity hazards, focusing on the overall impact and context of the hazards as they relate to other deficiencies and safety barriers, not just focus on the probability of the hazard causing an event.
The consequence of a hazard, although considered “low probability,” also may be more severe as a result of interaction with other system failures. For example, in the case of Air France Flight 447, each failure was arguably a low-probability hazard, which drives down the overall risk (using the current established process). But if you could somehow identify the possible interfaces with other hazards, you could develop safety barriers to prevent the interaction of hazards. With large amounts of data gathered through maturing safety programs (state safety programs, safety management systems), we often can see precursors to accidents when performing post-accident investigations and analysis. Our challenge is to enhance the tools and methodology for identifying the systemic hazards that cause these issues and unwanted states, and mitigate risk posed by those hazards before they actually cause an event.
Aviation is an extremely safe mode of transportation, but for aviation professionals, the reduction of risk remains the number one goal. Enhancement of strategies to identify causes of accidents before they happen must continue to be a priority. The Flight Safety Foundation is committed to this goal of preventing the preventable.