A World without Reason

Recently, I have felt like I'm in danger of becoming complacent with the bedrock of my chosen field. I'll admit that in the past, I've been fairly vocal about this bedrock's limitations and mantra-like recitation by aviation safety professionals the world over. But the recent apparent abandonment of this concept by one of the first Australian organisations to go "all-in" on it, gave me cause for reflection. I am, if you haven't guessed it, talking about the "Reason Model" or "Swiss Cheese Model".

But it wasn't a critical review of "Reason" that was on my mind. Instead, I started to think about whether we had embraced it enough to allow us to move on.

For me, being a "cheese-head" has just been part and parcel of being in the aviation safety game. Human factors was mother's milk during my first year of uni with CRM and organisational accidents the solids of second and third years. From there, I've continued along the modern system safety trajectory of culture, SMS and so on. I've never known it any other way.

But how has the rest of the world taken to it? The general public, I mean. The great-unwashed ;).

To look examine this, I thought I'd look at MSM coverage of aviation accident investigations in Australia. So, I took to Google and searched for pages related to three accidents in the days following the release of the related accident investigation report. I was looking at how the news reported the "causes" of the accident.

The three accidents I chose were:

  • Lockhart River - Australia's worst air disaster in the last 40 years or so and an investigation I knew did follow the accident causation chain right up to the regulator.
  • Pel-air - The trigger for all the current controversy and, in opposition to the above, a report that is generally said not to follow the causation chain beyond the frontline operators
  • R44 @ Jaspers Brush - The most recent investigation report to be issued which would have received media coverage and also a relatively small accident in which organisational factors might be hard to identify.

In Lockhart River's case, I could really only identify three stories in the immediate aftermath of the investigation report's release. One from the SMH, one from ABC News and one from Lateline (ABC as well). Overall, I thought the reporting was quite good. All three pieces discussed multiple contributory factors and generally shied away from the word "cause" - except for Tony Jones' intro to the Lateline piece which was actually more concerned with the regulator's role. However, the headlines for the SMH and ABC News stories were old school all the way - "Pilot error blamed for Lockhart River (plane) crash" - I guess we can blame the sub-editors for these ones.

For Pel-air, Google yielded only one real MSM link with a couple of other stemming more from the 4Corners story shown a couple of days after the report's release and many coming from aviation industry outlets. The Australian's story was fairly consistent with the characterisation of the ATSB report in that it focussed on the crew's actions but it did mention briefly more upstream factors. The other stories were quite critical of the ATSB report in its perceived lack of analysis beyond the Unsafe Acts and Local Workplace Factors levels.

In the final accident, the two MSM stories I found (The Australian & Fairfax Media) put a real emphasis on the "what happened" aspects and ventured little beyond that. In this case the operation was private and, I'm sure, some would argue that "Reason" doesn't apply. The fleet grounding and safety recommendation for a change to the fuel tank were mentioned.

At the very least, I sure more could have been said about the human factors aspects related to the event. And more could definitely be said about the aircraft and crashability standards for aircraft. As I said a couple of weeks ago, no man is an island. Even, private pilots and even, aircraft designers and manufacturers. Imagine the impact that this investigation could have had if its analysis showed that aircraft structural certification processes showed deficiencies in post-crash fire considerations.

I don't know if it does but if this is not the case, why then do we need to change the R44's tanks? This is not a high-level systemic fix. What is there to stop another aircraft type from having this problem in the future?

Okay, I'll admit that we can'y go on a mass analysis expedition with every accident investigation and we have to select those investigations that have the potential to yield the greatest safety benefit. As an idealist, I do have trouble with the finiteness of the real world even though I do have to deal with this in my day job.

But where does this leave "Reason"?

Well, the Lockhart River articles were (save for the subs) quite heartening and even the Pel-air coverage (overall) tried to encapsulate the complexity of an aviation system breakdown. I guess the disappointment is more the ATSB report which, as we saw with Lockhart River, can drive the media coverage.

I'd like to see the organisational accident or system failure approach to remain as fundamental for all aviation safety analysis and investigation. In fact, it should be extended to try to capture the non-linear, close-coupled nature of  complex socio-technical systems like aviation. The "Post-Reason" world may be upon us but I don't think it is based on the approach offered by the ATSB's chief:

If we want to go to Professor Reason’s model of investigation—though we think we have come a long way since Professor Reason’s initial work in the 1990s—there is error and there is violation. While the focus of our investigations is on error and understanding error—how to prevent it, how to detect it and how to deal with its consequences—there was also in this case an element of what, in Professor Reason’s model, would be viewed as violation; and that is principally the responsibility of the regulator.

Reason's error types fit well within his larger model and, to be honest, I don't see the ATSB-error/CASA-violation distinction. There's a whole other blog post on that one!

I still quite like the distinction I made in my other post on this subject where I considered the very high-level intent of the operator. If the intent of the operator was to get people safety from A to B on their aircraft, it falls within both CASA and the ATSB's courts. While the operator may intend on breaking a specific regulation or company policy, their overall intent remains getting their pax on the ground. If the intent of the operator is anything else, then it actually becomes a criminal matter for the police and OTS.

To analogise where I think we are at, "Reason" was a mud hut for safety professionals. It gave us a basic structure and shelter to develop the field a little more. Unfortunately, we've out-grown the hut and we need something more. Maybe a hard floor, doors, windows, who nows? There are quite a few options on the table to take us to the next level, its only a matter of time before someone puts it together in a package as neat as the "Reason Model" was. It's an exciting time to be a safety professional.