December 8, 2011

Multi-core Cranium

The past few months have been filled with the trials and tribulations of training.  My trainee (TQ) has spent the last few months training on the HNK/DNY sectors. This is my first main project (it seems strange to refer to my friend/coworker/trainee as a "project", but it is hard to categorize anything that takes over two years to develop as anything different - I would never mean to dehumanize....) as a primary instructor where the trainee was not already a controller somewhere else before transferring to my area.  It has been a learning experience for everyone.

After spending the obligatory two months in the classroom and simulators, honing phraseology and so forth, TQ hit the ground running, so to speak, and left me with a false sense that this might be easier than I thought.  This was right after Labor Day, and the weather in our area was really quite horrendous that first week he was back on the floor.  There were epic levels of traffic (since we were working all of our normal traffic, plus all of New York Center's traffic, and some of Washington's traffic, and they were all deviating).  I took over the frequency for a few extended periods of time, but only because I couldn't keep up, let alone would I expect a brand new radar trainee to keep up.  "Just watch for a little while" is pretty much all I could muster as words of wisdom. 

And then the worst of the summer traffic was behind us.  It still gets busy, but not for hours and hours at at time at freakish levels.  Mentally, we all settled in for a long winter.  I took a few days off from work in the early autumn, and the secondary instructor on our training team was involved in a union project of his own, leaving TQ to train with different instructors. 

As thick as our regulations manual appears, the actual performance of the job depends on strategy and actual execution of these imposing regulations.  We call it "technique."  There is more than one way to get planes A, B, C, D, E and F across the sector to points Q, R, S, T, U, V in a safe, orderly and efficient manner. 

So, TQ was exposed to many new and different techniques from other trainers in the area.  Hardly a day would go by without an accusing "who on earth taught you to do it THAT way!?"

As long as the rules are being followed appropriately, no one is seriously going to judge your technique.  But technique is everything.  So while I, or someone else, may not agree that TQ's way is the best way, as long as the job is getting done, we'll simply discuss legal alternatives that would work in that same situation.   Or theoretically if that one plane was a prop?  Or if there was icing at 17000?  Or if you gave that clearance and the pilot said "unable"?  Or if these planes had to deviate around storms? Or if you had 30 miles in trail to Kennedy.  Oh, wait, you do.  Get on that!

So after a while, TQ's brain started turning into a confused mushy sponge.  The basics got lost in there somewhere.  Life wasn't very fun for a while.  It made me start thinking about what was going through his mind.  Which made me start thinking about what goes through MY mind when I'm working traffic.  After sifting through some weird stuff that I'm sure you don't care about, I may have actually found something useful to dole out as worthwhile instructor-ey type insight. A few weeks later and I have an even weirder thought that maybe some of you non-ATC folks might find it interesting. 

Compared to your math teacher in High School or even the flight instructor at your local airport, Air Traffic Controllers don't receive much of any instruction on how to be instructors.  They send us up to the classroom for a few days of team building type exercises, but that's about it.   We learn how to control airplanes and how to be future instructors all from the same people:  The instructors in our area.  And they, of course, learned in the same fashion.  Training is supposedly tamer now.  No more "power training" with rulers held at threatening angles or random headset unplugging to draw your map for the millionth time with a golf pencil.  Heck, they can't even yell at the trainee's for bad strip marking.  Strips are gone.  The older generation has almost all retired.  These things are all myths to me. 

One question always arises in any discussion about ATC training, however.  Can you teach someone to "see traffic?  This is a two part question.  (1)What exactly would be the process of teaching someone how to "see traffic" and (2) do we currently do that or know how to do that?

Ok.  I know some of you may have expected one of those questions to be "what exactly is 'seeing traffic' anyway?"  It all revolves around a controller's "scan". 

Let me start by saying that I enjoyed radar classes when I was in college more than tower classes because I preferred having all of the planes right in front of me on the (simulated) radar scope as opposed to being completely surrounded by airplanes in the control tower (simulator).  My scan in the radar was better.  I rarely have to turn my head around.  I like that.  I belong in a dark room.  Yes, I accept that.  Luckily, the FAA called me up 7 years ago and said "you are going to work in a dark room."  Yay for me.

Scanning a radar scope starts with one airplane.  Pick one.  Generally, you would start your scan on the part of the scope that has a tendency to contain airplanes that have a high chance of being in need of your attention.  If nothing pops out at you as extremely urgent (planes about to enter another sector without a handoff, aircraft not at their assigned altitude, two airplanes at the same altitude and very close together, etc) then a quick scan in that area commences.

When I say quick, I mean spend about one second per plane.  That one second of time would contain the follow mental process: Is there anything that I told myself I'd take care of next time I scanned this plane?  What altitude is this plane currently at, what did I assign, does it look like they'll make the restriction I issued if I issued one, are they on their cleared route of flight?  Where are they going, what do I have to do with this plane based on its type of aircraft and destination, how much do I have left to accomplish? If everything seems normal, and the answer to the last question is "nothing", then I take another second to flash the handoff to the next sector if it isn't already and I move on to the next plane nearby.  If there are other planes going to the same place on the same route, I'll check the speeds.  This may take a few seconds as speed is not always permanently displayed.  If the answer to the tasks accomplished question is not "nothing", then what is there left to do?  Here comes the seeing traffic part: What do I have left to do with this plane and what other planes are going to interfere with my goals.  And so the eyes will take over again.  Where could potential issues come from?

The extensive training we receive at each sector teaches us that each common traffic flow has a certain number of confliction points with other common flows of traffic.  These common danger zones are where we look first.  My brain then tends to look in an arc shaped pattern, searching for planes that are a similar distance from where this plane would cross with them.   If a plane is  25 miles west of DNY, and other planes tend to cross at DNY, then I'm looking for planes that are about 25 miles away from DNY in all directions, with my arc changing distance depending on the wind.  If the wind is strong out of the west, planes traveling south and north will fly a little slower than the eastbound traffic, and the westbound traffic will fly the slowest, so I need to search closer to DNY in those cases.  My short term memory should remind me if I have any other planes that are flying abnormally fast or slow, and I should do an extra search for those planes, expecting to find the unexpected.  Another common traffic flow is for BDL arrivals. Common conflictions include LGA arrivals along a similar path, but often at a slower speed since they are lower in altitude already (and because Air Canada has been flying REALLY slow lately),  PHL traffic coming from the north at FL240 or FL260 most often, and traffic climbing slow out of BDL westbound.  ALB arrivals tend to conflict with EWR arrivals first, watch out for HPN and SWF traffic, then LGA props lower, and then ALB departures. There are over a dozen of these common flows in each sector and they all cross in a 3D pattern. 

Once you spend a second or two doing this scan for other conflicting planes, and you've scanned most of the other planes around your original plane, the overall scan order then moves around the scope to areas of lesser priority until we're back scanning the plane where we started.  

When we get really busy, these are the default settings our brains use to increase the speed of our scan.  A good scan forces the brain to keep looking for conflicts it doesn't expect.  When all the airplanes are deviating around storms and others are on reroutes, this extra scan is required for every airplane every time.  So while you're used to a scan taking a second or two for each plane, now its taking maybe five times longer, yet your brain is stressed out and wishes it could scan in half the time.  After a while, this gets frustrating.        

Can we teach someone how to go about "seeing traffic" in a way I just explained?  Sure.  We can.  We do.  But with limited traffic.  It takes time to go through and talk about each and every plane, just from a scanning perspective.  Then you have to talk about what to do with that plane, technique wise, and then that changes how that one plane interacts with all the others, of course.

The issue isn't that we can't teach someone how to scan airplanes on a radar scope.  The issue is how do they react to the really busy sessions with a frustrated brain.  Do they dwell on little lapses in their scan when they discover their overlooked traffic a few moments later (perhaps when their scan finally gets around to seeing that previously overlooked target)?  If they've made it to Radar training, hopefully we've established that they'll never just freak out and give up.  That's rule number 1: Never give up.  Ever.

As I list all of these mental questions, keep in mind that while many of them are "yes/no" type questions, or perhaps there is a "well, if this happens, then that will happen" question, the key to being a great controller is taking a busy, overwhelming situation, and generating creative solutions to urgent problems that pop up on a second by second basis.  Letting your brain operate like a computer program works great most of the time.  But most of the time isn't ALL the time.  Most of the time doesn't cut it in ATC, either.  And so, in order to create a sense of confidence and creativity, trainees need to start understanding why I use my techniques the way I do, when I use them and when I don't use them, and how to decide when to abandon your first plan and create a better one, quick.

I don't always have a good answer for why and how I do things.  I wish I did.   On the other hand,  changing plans is a basic ATC function.  As a controller scans their airplanes, they are thinking of all of that stuff from a few paragraphs above, but there is always one or more back-up plans in place, and thoughts about how that back up plan will effect everyone else.   TQ has reached that point in the last few weeks.  He's not afraid to change a plan.  He knows he'll find all the new conflicts quickly.  He knows he'll figure out a way to solve any issues that come up.  It might not be the way I'd do it, or how anyone else in my area would do it, and I might have to pester him for a few minutes to figure out why he thinks that's a good idea......  And why wouldn't I do that!?


With all of this maximized use of our mental capacity, I think the hardest and most physically damaging aspect of ATC is that we are constantly changing mental speeds.  My brain could be thinking at 100 miles a second, my eyes darting all around the scope, contemplating a plan of action for all of my airplanes, and then forecasting the future position of all of my planes to see if my strategy works out if everything else goes according to plan.........but then I need to TALK to one of my airplanes.  I suddenly have to think only as fast as my mouth can clearly issue a clearance. 

I need to precisely READ the aircraft's callsign in the datablock on the scope, then issue the clearance I intend (distinguishing it from all the other ideas I had about my other planes I was scanning) at a speech rate that can be easily understood and that exudes confidence.  Then I need to LISTEN to the pilot read the clearance back to me, and decided if that matches what I just said.  Often times, while listening, my brain can go back to scanning.  In fact, after reading the callsign, I will usually make a visual scan ahead of the plane that I am talking to, as one last check that what I am in the midst of saying is safe, all the while typing away at the keyboard and making glances down in the corner of the scope to make sure my inputs are being accepted.  How do I do that?  How do I teach someone else to do that?  No one ever taught me this last part.  I just do it.  It seemed necessary to learn how to do all of these things all at once in order to survive training;  to feel confident that I could handle anything.  I noticed TQ started doing this recently.  I'll move my chair so I can see what he's looking at.  After he reads the callsign, I see his eyes move ahead of the plane as he's issuing the actual clearance.  I couldn't tell him how to get his mind and body to do that, but he taught himself.  A few shaky weeks in ATC will motivate a human to do amazing things.     

Till next time....



Steve said...

My brain hurts just reading all that... but excellent post. It's really interesting to read about how you're thinking in that dark room!

Frank Ch. Eigler said...

Great writeup. It's not often that one hears such introspection about a technical task.

LRod said...

Two words came to mind in your discussion about teaching scan: shrimp boats.

It never occurred to me, but I have a strong sense that those of us who pushed shrimp boats had a whole lot easier time learning how to scan than the kids that came along PSB (post shrimp boats).

Interesting thought problem.

ZJX, ORD, ZAU retired

"La Vida de Perro" said...

LRod, I agree about the boats. It involved a "hand and eye" coordination that hepled us keep track of things. Another tool we had was target trail. We could "feel" aircraft speeds and motion in a sort of analog way rather than just read a digital display.

I wonder if the "old" ways weren't better designed for the human brain the the new digital displays. Ultimately, the computers will be making most of the decisions anyway, so we'll just have to learn to think their way 'til that happens.

Anonymous said...

Way too long and technical.

Kinda like telling a fellow good golfer about your recent "career" 100 yd wedge shot. You can do it in a few words, i.e., "...aiming at the right side of the green letting the wind work it right to left. I nailed it; after it landed the ball spun back toward the hole." Or, to a golf ignorant person(s) you could explain it technically; i.,e., Hit the ball with downwatd force. Now I create that force with establishing clubhead speed. I do that by turning my hips, then my shoulders clockwise, then slowly, blah, blah, blah. The non-golfer is put to sleep.

AC2usn said...

Great post!! Training is the hardest part of the job (for me).

Explaining your thought process and reaction to the developmental and the mind reading that goes along with this in the trenches takes much longer than the event simply because we write slower than we think plus the fact that we are checking on several things (close to) at once (scanning and planning).

Anonymous said...

Love it! I enjoy reading about what you guys in the Center do!!!


Cedarglen said...

Great post. Keep on posting. Yup, agree... It takes years to become comfortable with a full screen and to recognize/compute t he details quickly. Some can and some cannot, but they have yet to find a way to test that ability.

Don said...

Excellent blog, great insights. Am sharing your URL on my EAA Webinar on Communications this month. Please keep up the insights. This particular entry on teaching challenges and the true complexity of the art strikes many chords - flight instruction shares many of the same challenges. Pilots can bludgeon their way through the system or gracefully waltz in a dance with ATC, but the latter involves insights and skills that aren't usually part of a pilot training program.

Pete Templin said...

Alright, who's the wiseacre who scared off Delta Mike? You should know he's not the best at journalism. Or, who's the awesome matchmaker who distracted DM with a new ladyfriend who recognizes DM for who he is? Either way, DM, we miss you.