Safety Culture & Culpability

[WindSheer]

Hi,

I am after some feedback from both pilots and management on how safety critical 'incidents' are handled in your airline.

I will paint two scenarios, in particular, do you think there would be any difference in the development outcomes for the associated pilot/s.

1. A heavy landing as a result of an error during flare.
2. A heavy landing due to an unstable approach, i.e breaching company procedures on stabilised criteria and continuing with the landing.

Note. I work within management in UK rail and am always looking to transfer best practice. Any help greatly appreciated!!

:ok:

[Piltdown Man]

WS - I'm not sure what a "development outcome" is. It sounds like nasty HR speak for disciplinary action. But assuming it means "how do we fix the problem that caused this event" then yes, the two would be handled differently. Heavy landings are part of operations. They shouldn't happen but they do. In the first scenario, the Chief Pilot might have a friendly word and ask if the guy is happy to continue flying or whether they might like additional training. But this would "free" training and not appear on their records. In the second event there would almost certainly be a replay. The crew will be expected to justify why they continued. Depending on that justification additional training, gardening leave etc. may be allocated. To date, I know of no instances where stronger sanctions, if that is the correct term, have been taken.



ps. More vigorous action is unlikely due to the presence of a very strong pilot's Union. But even if stronger action was taken, would it work? Our history (>100,000 flights per year) shows that occasions like the second event are rare enough for stronger action not to be required.

[WindSheer]

By development I meant the company taking responsibility for any identified deficiencies and ensuring they are addressed through coaching. This would be fully transparent and 'on the records'.
Your point about the unions is interesting, as what you describe is more of a no blame culture rather than a just one. There are differences between the two.....

Oh and I will add, Uk rail still works with an extremely influential union.

[Denti]

In my company the process is pretty long, and only the final step of an event evaluation would be a culpability check, done by an event review group that consists of the safety department, works council (ombudsman), pilot of confidence (a pilot trusted by most pilots on the fleet), and fleet management. They follow a pretty strict flow chart and their decision has to be unanimous. After they have decided what kind of misconduct, if any, has happened they do decide on corrective action, which could be either briefing, coaching or training. The aim is to enhance the companies safety culture in general, and therefore none of the mentioned incidents, without further information available, would be a reason for disciplinary action.

Disciplinary action is the same as non disciplinary, either briefing, coaching or training, however it will go into the personal file (personal or flying file) of the pilot as disciplinary action, whereas non disciplinary action (corrective action) will not be entered into the personal files.

To be facing disciplinary action is not that easy, only for very few events with clear intent to harm others or oneselve can one expect to be held culpable in that manner.

We do have union representation and therefore both a works council and a company council. The company council is a union thing, the works council is based on a union contract but allows non unionized members to vote or be voted into the works council as well. The works council CLA is based on a german law (works constitution act (http://www.gesetze-im-internet.de/englisch_betrvg/)) and the whole structure with the division between union company council and non union works council is pretty unique over here.

[WindSheer]

Thank you Denti for such a well written and informative reply.
I like the idea of the committee, especially whereby reaching a unanimous decision is required. In my area this is more of a local manager thing. I.e. once an investigation of an incident is completed, any violation is picked out by the individuals manager and a local decision made on whether discipliniary action is required. The massive issue we have is the huge inconsistencies around one managers opinion on what consitutes a true violation against anothers. Even when using a typical James Reason style culpability chart, its amazing how individuals can end with differing outcomes. These vary from informal counselling to written warning or even dismissal.
These decisions should never be taken lightly and the consistent approach of a committee would definately help this. It's amazing how many presidents have been set over the years by managers who have absolutely no idea around the difference between errors and violations. Even when a violation is culture based (influenced by the organisation),they have no idea how to deal with this.
What part of the world are you reporting from - it obviously makes a huge difference.

Thanks.

[megan]

I am after some feedback from both pilots and management on how safety critical 'incidents' are handled in your airlineThe question I would have is, what sort of safety reporting system do you have in place? Are employees free and able to submit reports of the concerns they have? And if they are, what is the management response to those reports? Do they give an answer to those reporting? If not, it tells the employee "why bother". Both of the scenarios you paint could well be due to factors resident outside the cockpit.

Number one scenario may have absolutely nothing to do with the crew itself, but due to training, or lack there of.

Number two for example could be because of the manner in which management handle a crews decision to make a go around eg is it punitive?

At the heart in both cases it would then be a result of management decisions.

I just give those as examples, because you do state "do you think there would be any difference in the development outcomes for the associated pilot/s". To me your statement seems to imply that the "fault" lies within the cockpit. It is often the case that the gun is loaded by management, and the crew are the hapless souls who accidentally pull the trigger.

All too often the crew are the scapegoats for failures that lie elsewhere.

[ZFT]

megan,

Isn't this all covered (within EASAland) under AMC1 ORA.GEN.200(a)(3)?

This includes effective feedback processes else, as you rightly say, there is no point in reporting anything.

Also, the composition of a Safety Action Group and Safety Review Board are thus quite critical in determining whether the SMS will ever operate as desired or will just be a 'tick in the box' exercise.

Additionally, this process should not be restricted to perceived critical items but to anything that the reporter has safety concerns over .

I could not agree more with your last sentence.

[Denti]

What part of the world are you reporting from - it obviously makes a huge difference.


There might be a clue in the last part of my text, but anyway, from germany.

We do have a pretty nice explanation of the whole concept, but as it is company material i cannot hand it out.

[megan]

Isn't this all covered (within EASAland) under AMC1 ORA.GEN.200(a)(3)?ZFT, I come from the other side of the world, so am not au fait with your world, but my understanding from a search is that the document you mention refers to aviation, not railways as the OP is engaged in. Might be a good document, by a brief look, for the OP to wade through though.

https://www.easa.europa.eu/system/files/dfu/04%20Part-ORO%20%28AMC-GM%29_Amdt2-Supplementary%20document%20to%20ED%20Decision%202013-019-R.pdf

[WindSheer]

Thanks for all the replies.

I am a former aviator, I draw fond experience from this fantastic industry.
Our UK rail safety board also work closely with aviation to identify transferable practices....any help you provide is relevant and of use.

I will take a little look at the document and get back.

:ok:

[Dan Winterland]

A good safety culture comes from one that is perceived to be just. How to get that just culture and maintain it is enough work for a PhD thesis, which also indicates how many people know what an effective just culture is.

But it's safe to say that you have to have everyone on board singing from the same song sheet. That's the management, the pilot's union (representing the workforce) and the safety system, assuming if it's separate from the management of the company - which it needs to be if it's going to be effective. You need a clearly defined process which is adhered to if there is an event to investigate. And the safety reporting system needs to be open, transparent and free from disciplinary action, unless there is a wilful disregard for the rules - the response to which will be defined by the process I have just mentioned. There is plenty of information in aviation which has had a lot of success in this field. Please feel free to PM me if you want to discuss further.

But take this on baord. It can take years to build a just culture and only one day to destroy it if you handle an incident incorrectly. You have to get this right and be seen to get it right.

[alf5071h]

"If you think that you have a good safety culture then you are almost certainly mistaken," James Reason.

With that in mind look again at your beliefs about safety culture.
What is 'safety'? It's not just the regulatory or text book definition; what does it really mean to the operator and individual, what is actually being done?
''Culture"? 'It's what you do when no one is looking', work as done vs believed to be done? Go look; not after the incident, but at all times.

Reconsider the assumptions embedded in the incidents; particularly the effects of hindsight bias in posing the question after the event with a known outcome, even within this forum; question the effects of time and outcome knowledge with that available at the time of the event.
Question cause and effect; does an error in the flare always result in a heavy landing, what is the degree of 'heavy', who judges. What is error anyway?

Similarly for an unstable approach; what is the definition, why, who decides. Do unstable approaches always result in a heavy landing (there may be a greater risk of an overrun incident).

Reinvestigate the incident with a view that the human is an asset, trying to do their best. Check what happens on a daily basis - the operational behavioural norm; how many near heavy landings, how many unstable approaches without incident across the fleet? Context.
Why, why, why, why, why.

Finally, with all of the natural human biases, remember that the next action is a judgement (but without a jury), similar to decision-making with all of the variability of human behaviour.
Could this have happened to you; ..... see line #1.

"A just culture is not where the line is drawn, but who draws it," S Dekker.

[Piltdown Man]

Of the processes described above, the two most enlightened are alf & Denti's. The latter due to the process of asking who judges (Dekker - his YouTube channel is worth watching if you haven't already done so). Denti's process is to be commended in so much that at least one person involved knows about the nitty-gritty at the sharp end and that there has to be a unanimous decision. It might be better if there were more from the sharp end. The reason I say this is that the more developed our systems become, the more complicated they become. But far too regularly our managers fail to appreciate that the systems they put in place are unworkable or unmanageable. But if they are the ones doing the assessment on an employee who has been caught up in an incident I can not see them taking the wrap. Managers rarely accept responsibility for their own actions.

While we are here, it think it fair to state that I intensely dislike Reason's "Just Culture Decision Tree". Even the starting premise is wrong - the victim (because that is what they are) enters "Guilty" and only after five steps can you get absolution. But don't foul up again or step number five will get you. It is the modern day equivalent of the medieval ducking stool.

I'll tell you how we start an investigation. It starts with a single question. Did this person/crew go to work that day to do a bad job? If yes, we stop right there. On the other hand, we can cut to the real task of an investigation if the answer is no. The employees under investigation know they are "safe" and the management know they will get to hear what needs fixing, because everybody feels safe in the knowledge that they can speak openly. It is that background that sets the context of my first post.

The next question is, does it work? Yes, I think it does. Through the reporting system, regular "own goals" as well as every sort if other incident are reported and these, together with the more major events are analysed. The major threats are then addressed through training and the lesser ones through feedback to the crews. Individual weakness are addressed confidentially to the satisfaction of most of us.

PM

ps. The strong Union reference referred to the action (not) taken against some of our more flamboyant employees. Certain people have been able to get away with blue murder for a whole variety of non-safety related events.