Holding Improves Efficiency? Yup, Here's the Logic ...
A special supplement to Paul Berge's article "Who Needs Holds?"
The very idea of going around in circles in a hold at 150 knots is pretty ridiculous, particularly when it goes on so long that you end up diverting to an alternate. Makes you wish you'd taken the bus. Nonetheless, the feds really do want to minimize airborne holding and they use something called Flow Control to do it.
Flow Control was originally implemented as a temporary measure to regulate traffic during the 1981 PATCO strike. It has survived to become a permanent part of the system. The program is managed by the ATCSCC (Air Traffic Control System Command Center), formerly located in FAA HQ in Washington, and recently relocated to hightech quarters in Herndon, Virginia.
The program is complex and while it covers stuff that's not always important to GA pilots, you can still be affected by Flow Control. What they're trying to do, essentially, is to balance the arrival supply with the arrival demand.
Because any airport has the potential of receiving more instrument arrivals than it can handle (as happens during special events such as the Indy 500 or the Flying Farmer's Convention), even small airports are candidates for the dubious benefit of traffic management. Here's a thumbnail synopsis of how it works.
Step one is to determine the arrival rate, which is the number of airplanes per hour that the airport can handle. The arrival rate is determined by on-airport ATC, since it's influenced by almost anything: runway availability, wind, weather, staffing, the number of departures expected sharing runways with arrivals, even the phase of the moon (no kidding-a bright moon can make it easier for pilots to see the airport and other airplanes, allowing more efficient visuals).
Every airport is different, but under ideal conditions, O'Hare's arrival rate will be as high as 105 airplanes per hour. Under worst case conditions (say, one slippery runway available for all operations), it might be as low as 18.
ATCSCC takes a look at the anticipated demand for the period in question and compares it to the arrival rate. They actually work in 15 minute increments, but for our discussion, we'll use an hour. "Demand" consists of all airplanes that expect to land at the airport in the hour under consideration, including those airborne and those on the ground awaiting departure.
If the demand is less than the supply, no problem. Airplanes en route keep coming, departures go on schedule, and the Flow Control guys relax and drink coffee. But if 100 airplanes are expected to arrive during the hour, and the arrival rate is 80, 20 airplanes will enter the terminal airspace with no place to go, except around in circles at some holding fix. This is not considered to be a good thing.
The solution is to lower the demand for the hour by 20 airplanes. This could be accomplished with en route holding, but because airborne holding wastes fuel and increases controller workload, this isn't a preferred method. Instead, as soon as the ATCSCC sees excess demand developing, they'll start issuing EDCTs (Expect Departure Clearance Times). In effect, they artificially lower demand by ground-holding 20 airplanes that haven't left yet. These are sometimes called "gate holds," since the airplane doesn't push back from the gate on schedule, but waits until it's close to the ATC-dictated departure time. Airports without enough gate space may park the airplanes in a run-up pad or other holding area.
The problem with this scenario is that the airplanes hit with EDCTs of, say, one hour from the original proposed departure time, are now going to be arriving in the next hour. If the next hour is as busy as the last, the airport will have 20 more arrivals to add to the 20 excess arrivals it already had, for a total of 40 airplanes it can't handle. So, the same scenario repeats, and while it might seem logical for the original 20 airplanes to receive priority this time around, it doesn't always work that way. Due to the pecking order imposed, some (or all) of the original 20 that took the brunt of the blow last hour may get hit again.
So, what is the pecking order? At the top of the heap are airplanes that are already airborne, since en route holding is a last resort and rarely used. Airplanes inbound from Alaska and Hawaii usually get a free pass, too. Their long en route times make it difficult to judge how busy things will be when they arrive. Airplanes departing foreign points for arrival in the U.S. are exempt for similar reasons, which is just as well. Can you imagine an independent-minded French controller heeding the flow-control demands of some FAA bureaucrat?
Which leaves the domestic flights from within the conterminous United States. And among those domestic flights, the shorter the flight, the better the chances it will be hit with some sort of ground delay. After all, if the airport is saturated, airplanes that are closest to the destination present the most immediate problem. Airplanes that are close by, but not yet airborne, are the easiest to hold back. Unfortunately, they're often the last to be released when things improve. That's because their very proximity makes them useful pawns in the traffic-management game.
The slightest change in any single factor in the equation can affect the entire plan. Let's say that out of those 80 airplanes that the ATCSCC thought would arrive during a particular hour, 10 are arrivals from New York airports. But due to a stuck mic on one of New York tracon's frequencies, all of the airplanes take an unexpected 30-minute delay.
Now there are 10 extra arrival slots, but airplanes that are holding on the ground at distant locations can't possibly get to the arrival airport in time to fill the gaps. Suddenly, airplanes being held on the ground at relatively close-in airports become a hot commodity. Nearby Centers and terminals start scrambling to find airplanes to fill the vacant slots.
This being the case, the closer the departure airport is to the arrival airport, the more erratic the accuracy of the ground-delay times. If you're given an EDCT delay from an airport that's a short distance from the impacted arrival airport, keep your handheld tuned to clearance delivery while you're cooling your heels. You may be called to serve on short notice.
So, is the point of Flow Control to avoid airborne-holding altogether? Surprisingly, the answer is no. In fact, airborne holds of 5 to 15 minutes are considered to be ideal, in that they keep pressure on the airport, assuring that the final-approach courses are always full, with no gaps caused by a lack of arrivals to fill them.
For Traffic Management folks, success is measured only by a constant flow of airplanes on the final approach courses, at minimum spacing, with airborne-holding being kept to an acceptable level. The fuming of pilots on the tarmac at sunny locations hundreds of miles away is of little consequence. Lucky for them, flow controllers only have to look at the big picture, not the annoyance on the faces of passengers who thought they had no time to spare, so they went by air.