Chemicals & Materials Now!
From basic to specialty, and everything in between
Finding What You’re Looking For: Safety Investigations
Posted on December 11th, 2015 by Mike Schmidt in Chemical Manufacturing Excellence
On December 1, 2015 investigators studying the crash of an AirAsia flight the previous December released their report, in which they concluded that the crash stemmed from an upset caused by pilot error. This, despite the fact that a control system had malfunctioned because of a cracked solder joint and that records showed the system had malfunctioned 23 times in the year before the crash. This, despite the fact that upset recovery training had been deemed unnecessary because the manual from Airbus, manufacturer of the aircraft, said that the aircraft was designed to prevent it from becoming upset. Yet as reported by Tiffany Ap for CNN, the investigators said, “Flight crew action resulted in inability to control the aircraft.”
Not surprisingly, we find what we’re looking for. If someone is an aircraft crash investigator and believes that pilot error is the cause of aircraft crashes, they will find pilot errors. Likewise, if someone is a product liability attorney and believes that poor product design or manufacture is the cause of accidents, they will find poor product design or manufacture. Or if someone is a regulatory inspector and believes that accidents are the result of noncompliance with regulations, they will find regulatory noncompliance. It is in our nature. We find what we’re looking for.
Not All Accidents Are Preventable
There is an insidious safety slogan that is often repeated as though it is holy writ: “All accidents are preventable.” As a slogan, it reminds us to avoid complacency. As a truth, it presupposes that if an accident occurs, then someone failed to prevent it—that someone is to blame. But not all accidents are preventable. Certainly there is no empirical evidence to support the notion. As a matter of thoughtful consideration, it must be apparent that preventing accidents requires resources, so preventing all accidents requires infinite resources. Have any of us encountered an organization with infinite resources?
The pursuit of discovery and innovation in the chemical enterprise is fraught with opportunities for accidents. The mantra of “All accidents are preventable” notwithstanding, accidents happen, and they happen for a number of reasons. That is not to suggest that fatalism is the answer. Que sera, sera is not a safety policy. However, any reasonable person, while not accepting the accident, accepts that they must be prepared to deal with accidents. Part of dealing with accidents is safety investigations.
The Myth of Objectivity
Most people would quickly, and incorrectly, agree that a good investigation must be conducted objectively by someone with no personal interest in the accident. This immediately eliminates anyone involved in the accident, especially a victim of the accident. Rationally, this also eliminates anyone responsible for supervising anyone involved in the accident. After all, who is most responsible for creating the conditions in which the accident occurred? Frequently, and unfortunately, the supervisor of the area or department where the accident occurred is the first choice for safety investigator.
So where does the objective investigator come from?
From nowhere. As Sidney Dekker rightly pointed out in his book, Just Culture, an objective viewpoint is a “view from nowhere”, which no one has. Our training, our experience, and our beliefs are lenses through which we view everything. Although it is unproductive to put someone in charge of a safety investigation who has a vested interest in the outcome, we need to accept that all investigations will be subjective. Because no one is truly objective, the best we can hope for is an investigator that is aware of their own viewpoint and takes it into account.
Why Safety Investigations Are Done
It also helps to keep in mind why safety investigations are done. Unlike criminal, regulatory, or insurance investigations, a safety investigation is not about assigning blame. A safety investigation has the singular goal of understanding why an accident happened for the sole purpose of reducing the likelihood of a similar accident happening again. Assigning blame rarely, if ever, contributes to reducing the likelihood of a similar accident happening again.
To prevent any accident, the first step is to recognize the hazard. A previously unrecognized hazard reveals itself during a near miss (at best) or during an accident (at worst). The second step is to acknowledge the seriousness of the hazard amidst the many, many other hazards. The third step is to adequately address the hazard in the face of competing demands for time and resources. Addressing a hazard necessarily means that something must be changed.
Before an accident occurs, hazards are speculative by their very nature. A hazard may result in harm. Then again, it may not. Deciding which speculative harm deserves time and attention and which does not may not be at all obvious. After an accident occurs, the hazard is no longer speculative, but manifest. In hindsight, the seriousness of the hazard suddenly becomes evident. Flukes and stumbles that were easily dismissed at the time suddenly become obvious near misses and warning signals.
So before an accident occurs, we should seek to understand the hazards and prepare for them, understanding that despite this, accidents will still occur. After an accident occurs, we should seek to understand the cause of the accident. We want this understanding, not to blame someone for the past, but to reduce the likelihood in the future. For the future to differ from the past, something has to be changed. If the cause is not properly understood, then proposed changes are unlikely to affect safety in the future.
In addition to concluding pilot error as the cause of the crash, Indonesia’s National Transport Safety Committee recommended that AirAsia train their pilots flying the Airbus 320 on how to make an upset recovery. This recommendation does not fix the basic cause, a control system that breaks and malfunctions, but it is something within the NTSC’s power to do.
When (not if) you are faced with the responsibility of investigating an accident, challenge yourself to carefully consider three questions:
- In a world of finite resources, do my conclusions and recommendations make for a safer future?
- What are my interests in this accident?
- What are my beliefs about accident causation?
Do not be satisfied with pat answers and glib responses. If they were enough, there would not have been an accident in the first place. Likewise, remember that you are not objective, that you do have a viewpoint; make sure you know what it is. Be sure you are looking for the right thing, because you will find what you’re looking for.
All opinions shared in this post are the author’s own.
 Ap, Tiffany, “Pilot response led to AirAsia crash into Java Sea”, CNN, Updated 5:39 PM ET, 01-Dec-2015, accessed at http://www.cnn.com/2015/12/01/asia/air-asia-crash-report/ on 2-Dec-2015.
 Dekker, Sidney, Just Culture: Balancing Safety and Accountability, 2nd Ed., Ashgate Publishing, Burlington, VT. 2012. p.116.
R&D Solutions for Chemicals & MaterialsWe're happy to discuss your needs and show you how Elsevier's Solution can help.
Principal, Bluefield Process Safety, LLC
- Setting Limits: Boundaries of a PSM-Covered Process
- One Company, Multiple Information Solutions
- Knovel-Wiley Partnership Empowers Engineers to Navigate Chemical Manufacturing Challenges
- Inconceivable: Unrecognized Hazards
- Meteors and Black Swans: Worst Case Scenarios