Recently, while conducting yet another tour of Atlanta Center, I noticed a pilot's fascination with the D-side. He just couldn't quite figure out what the D-side was doing. We had walked over to a different Area than mine to take a look at his home airport. It just happened to be quite busy.
As a matter of fact, it was so busy that the controller that had worked him out of the airport was still working the R-side (Radar or Radio controller). That turned out to be fortunate (for the pilot anyway) in that he was able to ask specific questions about how he was handled coming out of the airport while the situation was still fresh in his mind and the controller's. The sector also happens to include a significant portion of non-radar work. But I'm getting ahead of myself.
The D-side is the data controller. Some refer to it as the manual controller. Others refer to it as the non-radar controller. I need to point out (to spare myself the argumentative email) that the D-side position at Atlanta Center (ZTL) has evolved over the years somewhat differently than at other facilities. Controllers that transfer into ZTL are mystified by the differences, on occasion. Others actually learn to like it and appreciate the differences. There is one more caveat I'm going to throw in here: What follows is how I believe the D-side position should function. You can find plenty of debate about "technique" among controllers.
The primary job of the D-side is to separate airplanes, just like every other controller. The processes he uses to get the job done are very different from that of the radar controller (the guy you talk to on the radio). Atlanta Center still uses Flight Progress Strips (thankfully) and these are the primary tools of the D-side. The strips are printed next to the sector and the A-side (Assistant Controller) places them into plastic strip holders and delivers them to the appropriate sector. The D-side takes them, reviews the information printed on the strip and sequences the strips according to the time printed on the strip.
Time is everything to a D-side. It's everything in Air Traffic Control. Release time. Clearance Void Time. Expect further clearance time. Turn rates: three degrees per second. Climb/descent rates: 1,000 feet per minute. Speed: nautical miles per hour. Time, time and more time.
If there is a tie on the time (two times exactly the same) then the tiebreaker is altitude. The lower altitude is sequenced below the higher altitude. If there is a tie on both time and altitude, you'd better hope they are on routes that don't conflict. If they aren't, you've just discovered the first duty of the D-side: discovering flights in conflict.
As obvious as it sounds, the first step in keeping airplanes separated is to notice that they are not. This is the time that a D-side should put red "W"s on the two strips in question. Most controllers would scoff at that idea now. I'll let the fact that I still do it speak for itself (for now). It's nothing more than a visual "note" that you'll need to take some action to separate the two airplanes. I'll admit -- in this day and age of near universal radar coverage, GPS and computers -- that the idea does seem quaint. There's more to it than meets the eye, though.
Continuing on ... the D-side does this with every single strip that arrives at the sector. We typically have about 16 strips in a strip "bay." A strip bay is a piece of metal with guide rails on it that holds the individual strip holders in place so that we can stack them on top of each other. It keeps the strips in a position where they are easy to read and write upon. The number of strips a D-side has to keep up with changes constantly. Each sector at Atlanta Center has at least two full-size strip bays and two small strip bays. It's not uncommon to see all four bays full; that would be around 48 strips (one per airplane) for those that are curious.
As the D-side puts each strip into the bay, he is supposed to be building a mental three-dimensional model of what the traffic will look like. I don't know anyone who can do that with 48 airplanes scheduled to be in their airspace in a 30-minute time span, which is one of the reasons strips have fallen into such disfavor with controllers these days. That and the fact that so many airplanes are on direct routings. Conflicting random routes are much more difficult to "see" in your mind's "eye" than two aircraft on airways. During the peak traffic periods, you just can't use strips like you're supposed to use them.
That doesn't mean they aren't still useful. And it certainly doesn't mean you can't use them as intended when traffic is a little more manageable. It just means that strips -- like any other system -- have their limitations. As a matter of fact, in that I'm a safety representative, I actually view this "limitation" as a benefit. Strips let you know you're going to get busy (or that you'll reach your limitations) long before you can see it on the radar. They let you know in time to do something about it before you're overwhelmed.
Think of it this way. The D-side (using strips) can think and plan in more strategic terms than the R-side (using radar). Using radar alone, you aren't aware of an aircraft until it shows up at the edge of the scope. The D-side has known about that aircraft anywhere from 10 to 30 minutes. Just as important, a datablock on a radar doesn't really tell you much. You see the callsign, the altitude and the ground speed. Actually, you're seeing what was the altitude and ground speed a few seconds ago.
If you want to be able to see into the future -- if you want to know what the flight wants to do and what it is going to do -- the data you need is on the strips. For instance: Is it an arrival? Will the aircraft need to descend? Has the STAR been issued? Is it a turboprop or a jet? That will determine what kind of performance you can expect and what altitude you assign it before handing the flight off to Approach Control. All the information you need is on the Flight Progress Strip.
Along these lines, one of the most common "gotcha's" for controllers is the flight that makes a big turn. The vast majority of these flights are military training flights. Most other flights travel in relatively straight lines. When controllers let themselves get too wrapped up in the radar they quickly get into the mindset that everybody is traveling in a straight line. Then along comes a military training flight that makes a 90-degree (or greater) turn. Needless to say, if you aren't expecting it, that can ruin your day. A good D-side can save the R-side in a situation like that.
When a trainer is trying to explain the D-side to a trainee in simplistic terms, that is often the description they will use: Your job as a D-side is to save the R-side. At Atlanta Center, the D-sides have been taught to take as much workload off of the R-side as humanly possible. The D-side moves datablocks around on the radar scope to keep them from overlapping. They update the datablocks as they R-side issues clearances. All of the telephone calls are routed to the D-side. Yes, my fellow controllers, all of the calls. Even the ones involving radar coordination. I told you Atlanta Center was different. The D-side sequences the strips, makes and takes point-outs, makes and takes handoffs, makes and takes all the coordination calls and pretty much stays busy as a beaver. He does everything except talk on the radio.
The R-side only has to do three things if he is blessed with a good D-side: Talk on the radio, mark strips and think. That is another thing that is different at Atlanta Center. When it comes to marking strips the philosophy is: If you say it, you write it. It seems backwards to some people (and is backwards from a number of facilities) but it works. And it works very well, I might add.
I'll try not to take off on too much of a tangent here but other facilities use their D-sides like stenographers. They write down what they hear (or think they hear) the R-side say. For example: If the R-side says, "Airliner one twenty three descend and maintain one two thousand" the D-side will write 120 (for 12,000) in the altitude box on the Flight Progress Strip of Airliner123. The R-side will type in 12,000 into the computer to update the data block of Airliner123. In Atlanta Center it's reversed. The R-side will write on the strips as he talks and the D-side will type the update into the computer.
There are a whole host of issues involved as to why Atlanta Center works this way but I'm only going to take time to mention two. First, it's easier to write as you talk than it is to type. At least it was when I became a controller. That may be changing, as even preschool kids are learning to type on a computer. The second issue is something as simple as what your mother used to tell you: "If you want to remember something, write it down." In other words, writing reinforces memory.
This leads me to an area that I think suffers from an unconscionable amount of neglect in aviation: human factors. I hope the neglect isn't as bad on the pilot side of aviation as it is on the controller side, but I fear it is. There is an incredible amount of research available to aviation regarding human factors. More research is being done every day. But somehow, it never seems to filter down to the operational level. Take the above example about writing and memory. It's not that hard to understand. Anyone who has ever been to college and taken notes knows that half the battle is just writing it down. The physical act of writing reinforces memory. Even if you forgot everything said, just a glance at your notes "jogs" your memory enough to fill in the details. The guys with the PhDs can tell you how it works with words that will put you to sleep. What we need are people who can tell it in a way that will get through to the 20-something-year-old Type A personalities who need to use it and wake them up.
Another important function of the D-side is reviewing proposed flight plans (proposals). This is mostly a function of low-altitude sectors. Centers receive a proposed flight plan (supposedly) 30 minutes prior to the departure time. This gives the D-side enough time to review it for any errors and to see if the route and altitude requested will fit in with traffic volume and flow. If there is a problem, the D-side can fix it. That change must be coordinated with the controller who will issue the clearance and with the pilot. The coordination with the pilot will usually take the form of the dreaded full-route clearance.
When you call FSS to get your clearance off of an uncontrolled airport in Center airspace, the controller the FSS specialist calls is the D-side. We've had this conversation about requesting a clearance while still on the ground before, but I hope you'll see the obvious safety advantages with a new perspective. The D-side isn't talking to other aircraft on the radio. That eliminates the time pressure the R-side feels any time he has to give a long-winded clearance on the radio. It eliminates any chance of being interrupted by another aircraft's transmission.
Here's another advantage you may not have thought about. Remember way back when I said the D-side can look into the future using the strips? Let's say you want to come out of Statesville, N.C. (SVH). SVH has the distinction of being only five miles outside of Charlotte (CLT) Approach's airspace. Let's also say you need a 15-minute void time. The D-side, looking at the strips, can see that he has an arrival from the south (in the direction of CLT) due at SVH in 13 minutes. The R-side, looking at only the radar display, can only see a maze of airplanes close to the CLT airport. You may not like the decision the D-side makes if you're the guy on the ground and you only get a 10-minute void time. But I'll guarantee you'll like it better if you're the guy inbound and the D-side just saved you a three-minute hold while waiting on the departure to get out of the way. The sector will run better too. The D-side, with the better data the strips offer, can make a better decision. He also has a better chance of noticing the data available because he isn't under the same mental pressures that make the R-side so challenging.
I can't stress this point too strongly. Everybody and their brother can look over the R-side's shoulder, glance at the traffic and say, "That doesn't look so tough." The difficulty in working a radar position decreases exponentially for every step you take away from the scope. Even controllers are terrible about succumbing to this trait. A controller sitting in front of a radar scope will feel mentally maxed out. He can't handle another thing. A controller across the isle thinks the guy is just moderately busy. The supervisor will look at his scope and think, "That doesn't look so tough. I saw somebody work twice that many yesterday."
Unlike the guy who is actually sitting in front of the scope, they don't have to come up with a plan. They don't have to deal with the distractions. They don't have to sift through all the visual and verbal clutter to find the pieces of information that are needed right this second. But a good D-side can see it all. He's as absorbed as the R-side with all the details and all the nuances, but with one vital difference. He's following the plan but he doesn't have to formulate it. It's just enough mental wiggle room to allow him to step back and find the one piece -- the one move -- that will make everything fall into place. If you ever get to see it done, you'll see a thing of beauty: the teamwork they've developed over the years, the trust they display in each other. I wish you could see it just once. But you never will.
Virtually everything I've written about in this article has changed -- for the worse, in my humble opinion. The most obvious change is that we don't even have a D-side most of the time. It follows logically that if you spend most of your time working alone, you spend most of your time developing habits that aren't necessarily conducive to teamwork.
All the functions that a D-side does get transferred to the R-side when the R-side is working alone. On paper, anyway. In reality, the controller sitting at one of the departure sectors for ATL or CLT isn't reviewing the 15 to 40 proposal strips. He's not about to take his eyes off the 10 airplanes he's working on the radar scope to review the ones on paper that aren't even airborne yet.
Any errors in the proposed flight plans that could have been fixed while the aircraft was on the ground will now have to be fixed once the flight plan is active. That will not only mess up the flight data processing (usually) for other controllers; but for you, the pilot, it means you'll be looking for charts or reprogramming your GPS while you're trying to fly your airplane.
Remember back in the example I gave you on the arrival and departure at SVH? When you're working by yourself, it's much more difficult to switch mental gears and "project" into the future like a dedicated D-side can. I failed to do that just five minutes ago while working the sector by myself. And sure enough, I had an arrival show up for SVH after I'd given a clearance (yes, on the radio instead of through FSS) to an aircraft on the ground, departing SVH.
Which brings me to my final point. A D-side provides redundancy. It's simple to understand the "two heads are better than one" principle, but the redundancy goes much deeper than that. Scanning a bay full of strips -- looking for traffic and projecting into the future -- is a very different mental process than scanning a radar scope. Scanning a radar scope is much easier, faster and more efficient. Using a radar scope is very much like playing a video game. The video game analogy is a convenient way to explain air traffic control in a minimum amount of time, but I cringe every time I hear it. ATC is so much more complex. Besides, people who play video games usually wind up losing. Controllers can't afford to "lose." A D-side, worked properly, provides a redundancy in the mental process. Although the process is slower, it is much more methodical than the scanning technique used to work the radar position.
Regardless of how sentimental I get about it, the D-side -- as I've known it -- is on its last legs. It's been dying for years. It started at least a decade ago when the FAA changed a few internal rules and has been going downhill ever since. The FAA has a new system that is being installed called User Requested Evaluation Tool (URET). URET will automate many of the functions on the D-side. It is also being used to replace strips. Just the rumor that it was coming to ZTL caused many controllers to lay down their pencils and cheer. Marking strips has always been considered a tedious duty of dubious benefit by most controllers. Anything that will eliminate that duty would be popular.
As usual, I find myself on the unpopular side of the argument. Nothing new there. But let me provide you with a quote that caught my attention. It will probably never get any notice in the aviation media. It probably won't be long remembered by folks who seriously study air traffic control accidents. But it got my attention.
It's on page 77 of the German Federal Bureau of Aircraft Accidents Investigation's report on the Ueberlingen midair collision. For those who aren't familiar with the accident, a controller was working the position alone with some important systems down for maintenance. You'll have to overlook the odd use of the English language in the following quote, because the original report was written in German:
For all flights the control strips were available in time. From the control strips it was not apparent that the two airplanes (B757-200 and TU154M) were in a conflict situation. A common reporting point had not been provided. Both airplanes had been cleared for a direct approach to Tango VOR (B757-200) and Trasadingen VOR (TU154M) and thus the control strips did no longer correspond to the actual flight paths. Only in connection with a radar image, the conflict would have been detectable.
(The report is a 3.5 Mb Adobe Acrobat PDF file, and the two appendix files are, respectively 2 Mb and 2.4 Mb PDF files. You can read the report, Appendices 1 & 3, and Appendices 2 and 4 through 10.)
I can't help but wonder if this could happen to my fellow controllers or me. As we automate more and more to meet the demands of efficiency, I wonder if we will have the wisdom to recognize our dependency on automation and the discipline as a profession to retain the skills we need to survive without the technology. An air traffic control center isn't your local burger stand. The operators had better know how to count change if somebody pulls the plug. But those skills won't do you any good if the cash register won't open.
The need to meet the safety challenges while increasing efficiency is obvious. The will to meet those challenges -- the political, institutional and professional will -- isn't quite so obvious.
Have a safe flight.
Want to read more from air traffic controller Don Brown? Check out the rest of his columns.