Air Traffic Management

As I have explored in previous posts; Innovation in training, the future of ATC, etc, ATC in the future will change.

A lot of the current instructors, especially those preparing students during  intial training have left active operations for 100% training provision.

A good number of these instructors will never go back to operations, and although a number of schemes, within the training provision certification process require them to do familiarisation or observation visits, these schemes remain marginal and only provide tasters of how the operational world is evolving.

Now with further autmoatisation and dyinamicity of the ATC system, this category of staff runs the risk of being left behind.

This will impact training institutions three fold:

First, a lot of the changes in the ATC systems will be radical enough to change initial training, especially the rating part.  This will mean that training institutions who deliver this training will need to find further schemes to have personnel able to understand [let alone teach] the implications of the new systems, procedures and ways of working.

Second, the training staff I am referring to, are often also those involved in writing the training objectives and syllabi [for example the review of the Common Core Content syllabus for ATCOS].  How will they be able to evaluate the training gap and training needs if they do not understand the implications enough?

Third, training institutions risk of having some of their staff made redundant, or at least off-side and this we all know is neither effective nor efficient.

I suggest that training institutions who have full time ATC instructors working for them and who are not keeping a current license should think about these factors and about ways of how to enhance the instructors ability to stay on the train.

If the train leaves without them it might then be too late for them, and quite expensive for their employers.

Share on Facebook

“When a pilot reports a manoeuvre induced by an ACAS RA the controller shall not attempt to modify the aircraft flight path until the pilot reports returning to the terms of the current air traffic control instruction or clearance but shall provide traffic information as appropriate.” (ICAO Doc 4444: PANS-ATM, Chapter 15, para 15.7.3.2)”Once an aircraft departs from its clearance in compliance with an RA, the controller ceases to be responsible for providing separation between that aircraft and any other aircraft affected as a direct consequence of the manoeuvre induced by the RA. The controller shall resume responsibility for providing separation for all the affected aircraft when:a) the controller acknowledges a report from the flight crew that the aircraft has resumed the current clearance; or b) the controller acknowledges a report from the flight crew that the aircraft is resuming its current clearance and issues an alternative clearance which is acknowledged by the flight crew.” (ICAO Doc 4444: PANS-ATM, Chapter 15, para 15.7.3.3)

“Note: the commander [the pilot] is required to climb or descend in response to an RA even if ATC has instructed him/her to take the opposite action.”Â

It is clear!

When the aircraft departs from its clearance in compliance with an ACAS RA, the controller ceases to be responsible of the separation between the aircraft and any other aircraft directly affected by the manoeuvre.

Or is it clear?

Before I try to explain what I mean, let us consider the present air traffic control paradigm* which is based on the controller being in control and striving to achieve, amongst other things, an orderly flow of traffic, and of pilots responding to and executing clearances.  And then there is an exception: ACAS RA.

As incidents have shown, it seems not to be so easy to exit this paradigm when an RA occurs.  Occasionally, controllers still issue clearances to aircraft executing an RA, at times in the direction of the RA manoeuvre but at others in totally conflicting directions.  And pilots still sometimes listen to controllers instead of their RAs.

So what could be wrong?

One may point at the lack of preparedness [see training] of the people making these mistakes.  But is it only a question of training?  Can adequate training solve all the problems?  Does the solution of the problem go a longer way than training?

Working in the training development field, I sometimes observe that we try to solve most of the complex problems by supplying [more] training, while in fact the most effective solution[s] would be better taken elsewhere.

So let us get back to the problem at hand: Controllers and pilots occasionally do not follow the rules concerning which manoeuvres should be executed when there is an ACAS RA.  I am not a human factors expert, but I think that even if training can go a long way, there are some other things that relate directly to human behaviour that would need to be dealt with at a more fundamental level.  Let us start by thinking of why the pilot would listen to a controller instead of the ACAS RA? Is it maybe due to a combination of habit, pressure and a degree of trust in the person issuing the clearance? How many times do we, even as assertive human beings, subconsciously make a subjective judgement of what we are being told based on who is telling us to do it?

Can we train to get over this?  To what extent can I train to make an objective judgement when I face people with whom I have a relationship based on a high degree of trust? Especially in situations where I am under pressure…In summary, apart from the objective judgement of which instruction to execute, what about the subjective judgment based on the trust that is built between the controller and the pilot?  Can we train to discard this?  Should we find other means to deal with the exceptions? Or should we find ways to include this kind of intervention within this trust relationship?  Food for thought…Then, what about the controllers still issuing clearances to aircraft even when knowing that these are executing an RA.

It is easier to say that “the controller ceases to be responsible…” than to acknowledge that the way we work as controllers [at least the way I did!] is to make a complex plan, and to execute it with an extended margin of flexibility to adapt it in the medium term, but with a significant reduction in the margin when situations become imminent. So in the example of an RA which produces a domino effect over a complex situation, is it that simple for the controller to cease to be responsible….?

The paradigm on which the controller’s work is based is that he or she has control over the situation.  It is also based on establishing and maintaining an orderly flow of traffic.  So how does this fit in with the chaos and loss of control that a [or multiple] RA manoeuvre[s] often cause?

Again, is reactive training composed of presenting the controller with RA situations where they should not intervene enough, or should we also train to deal with loss of control and chaotic situations?  Could we provide tools other than training, such as supplementary technology or alternative ways of working, that could help a controller who loses control to make sense of the situation?  Food for thought…[Yes, I know, down-linking the RA to the controllers is work in progress, but even this solution has been disputed by controllers and safety experts.]

Finally, what about the future?

In the future it is foreseen that there will be more situations where the controller will lose full or partial control of the situation. [I am referring for example, to the delegation of separation to aircraft] Can we use the lessons we are learning from the ACAS RA situation to encompass the new solutions, with enough consideration of the effect they have on the way controllers and pilots execute their responsibilities, and specifically on the way humans organise their work and relate to each other?

Technology, even when highly useful and when it has a proven record of saving catastrophic events from happening, (as in the case of ACAS) should not be introduced in semi isolation.  How the rest of the system, including the humans, need to adapt in order to accommodate it should be considered at a very early stage.  And training should not be seen as the only solution to achieve the necessary adaptation.

By way of a conclusion

The established procedure describing the responsibilities when an ACAS RA takes place, seems to be clear; however there needs to be much more than a description when it comes to complex relationships between the controller and the pilot and the way that each one of us works. Presently, controllers feel comfortable when they are in control and when the situation is well ordered.  Pilots are also conditioned to listening to controllers, and a trust relationship exists between the two parties.

Training to adapt to situations which are different from the norm will help; however, on its own, it is not enough.  Other things, such as considering softer issues such as the relationships between people and between people and systems, need to be considered.As regards to the future, those designing new systems which interfere with the paradigm of ‘control and order’ should probably consider more things than just training and written procedures…

---

[1] * Paradigm = A set of assumptions, concepts, values, and practices that constitutes a way of viewing reality for the community that shares them, especially in an intellectual discipline. Â

Share on Facebook

In a previous post I wrote about how ATC is likely going to change in the future – horizon bracket 2012-2020.

Automation, increased traffic and the quest for efficiency and respect to the environment while increasing safety levels will drive ATC to change.

In fact change is already happening. Dynamic flight data processors,  further airborne surveillance systems, dynamic sectorisation and different communication mechanisms than simple R/T are seeing the light as I write.

But is training for these new functions changing?

I think not, or at least not yet. I think that where it comes to training Ab-Initio controllers, we are still stuck in post WWII processes using late eighties technology [with some hybrid solutions here and there] and nineties methodologies.

Let us start with the processes:

ATC used to be [and still is to a certain extent] a predominantly linear activity where a controller has a linear cause to effect relationship. For example, a controller gives a clearance [cause] gets a readback, [effect 1a] and monitors initial execution, [effect 1b]. Then moves to problem 2 analyses, decides whether to take action, takes action, gets feedback, etc. From what I see, ATC could become a multi-tasking process [I know that many ATCOs will tell me that we are actually multi tasking now, but for me in fact what we do is to do many linear tasks rapidly.]. So, what do I mean by multi-tasking: Problem A is identified using a what if scenario, information is sent to pilot flying. Problem B is identified a solution is proposed by the system. In the mean time Pilot from problem A acknowledges and the system assists by prompting the controller, etc.  Then back to Problem B,…

If this is the case how will we train our future controllers to be competent with multi-tasking in ATC? The systems need to help us, but our processes also need to change. In basic radar courses, for example we advice students to upon a turn clearance wait until a turn is monitored. Will this be advisable in the future?

There are other processes that need to be addressed and where in we need to innovate such as the relationship between an executive controller and a multi sector planner or between a controller and a pilot who has been delegated separation. The scope of this article is not to discuss in detail how we can innovate on these and other matters but to highlight the need to innovate.

Then let us move to technology.

For skill acquisition we are extensively using simulators – which are high fidelity replicas of the online environment with record and replay features. Nowadays we are starting to use pc based part task trainers [PTT] for some parts of the training.

Can skill be acquired by other means than simulator practice? What we are doing nowadays to face the increase in complexity is that we increase the number of simulation hours. Can we not innovate here?

Can a number of functions such as training to spot problems, training to multi task, training the eye to spot details and system prompts, etc. not be trained using other systems and technology?Â

Do we always need hi fidelity replicas?

Could our systems give a degree of automatised feedback to the learner?

Could our systems not dynamically adapt to the level of the learner and perhaps either become increasingly or decreasingly complex, or take over certain functions off the overloaded learner?

Some of these concepts are present in PTTs, but should they not evolve further?

Can simulation practice not be done between 2 or more networked students over the internet?

And what about knowledge acquisition?  I have written about this previously. In brief is 75% of knowledge acquisition time classroom based necessary? Do students need always to be on location? Do they always exclusively need an instructor?  Can they not collaborate to learn? Can they not be offered interesting training prior to the formal commencement of their course?

These are all areas where our systems, both for skill and knowledge acquisition could be innovated.

Lastly our methodology:

Currently for safety reasons during simulator 1 instructor monitors 1 student. It is up to the instructor to give feedback to the student. The student will learn based upon the experience and the feedback given.

But can the student not learn alone? Could he not compare his performance against an ideal? Can he not be taught in bits and then shown how they make a whole?

Can student not during a facilitated discussion following a practice exercise [i.e. 1 instructor to many students] not learn from the experiences and the hindsight of other students? Is simulation time really safety critical? At all phases of training?

In parallel for knowledge acquisition, do students always need guidance from a tutor, at all time? Can they not collaborate to learn? Can they not go [under a different type of guidance] and research for the information themselves?Â

Should we not better teach them how to learn and where to look for the information, rather than teaching them the information ourselves? They will need to keep current and learn new procedures and how to operate new systems throughout their careers, so is it not more effective to teach them how to learn?

How can we innovate our teaching methods? How can we leverage on the fact that the generation being trained are used to go an look for information for themselves. That they are comfortable with technology. That they are better multi-taskers than most of us are.

That they will be working with the future system?

Indeed, the ATC system and the role of the human within the system is moving on. The drivers of automation, increased traffic and the quest for efficiency and respect to the environment while increasing safety levels are inherited onto training. It is time to innovate training unless we want to become the weakest link and/or produce a generation of misifittingly skilled controllers.

It is time to launch a research and development programme with the aim of innovating the three axes I mention above in line with the ATC system evolution.

Share on Facebook