I think it’s safe to say that most of the audience here has endured the plight of a flight delay. Notwithstanding the frustration of not getting where you need to be, when you need to be there, the plethora of reasons as to why your flight could be behind schedule is at best overwhelming to comprehend. I’m bombarded by questions about delays from passengers, and it has dawned on me that there is a serious lack of information available to provide answers. That is until now. It’s my mission over the course of the next few articles to shine some much-needed light on this subject.
The airline route-structure model has for the most part remained untested since the dawn of the legacy airline age. A few airlines such as Southwest tend to stray from it, but the hub-and-spoke system is the name of the game. An airlines hub is a strategically placed central point where flights arrive from all over into one place, and from there passengers in mass connect onward. For this model to work best, flights are coordinated to arrive and depart in banks. Because of this, there are peaks and lulls in the daily schedule as aircraft arrive and depart at the same time, resulting in sensitivity to traffic backups as well as a need for this article. The hub locations in today’s legacy airline business model are mainly geared to group passengers together to feed the international route network. With fierce competition in the domestic market since the advent of the low-cost carrier, legacy airlines have turned to the highly profitable international markets, as the premium on business travel and cargo is quite lucrative. Naturally, the like-mindedness of airline executives tends to yield similar locals as ideal placement for a hub. Hence the major hubs located in the Northeast megalopolis of the U.S., starting with IAD and moving north to include PHL, EWR, JFK, and BOS. Beyond that there’s also ATL, ORD, DTW, MIA, IAH, DFW, LAX, and SFO – all major international gateways. However, bear in mind that the brunt of commercial aircraft concentration focuses in on the northeastern U.S. This yields the busiest airspace in the world, and therefore the most commonly delayed as well.
Busy airspace leads to traffic congestion, just as a busy highway leads to bumper-to-bumper traffic. This congestion can arise for a multitude of reasons, mostly revolving around a variety of weather phenomenon, but can ensue simply from scheduled demand in finite airspace. Each airport across the country and around the world houses it’s own unique issues that can almost predictably cause traffic to back up. Whether it’s the high winds in ORD, morning fog banks in SFO, tropical disturbances in MIA, or convergence of weather systems in NYC, it’s part of a frequent fliers muscle memory to recognize and tense up when they transit one of these areas.
But to really grasp the scope of traffic congestion and all of the details I’m going to discuss, we cannot just focus on the airports themselves. Often, the problem behind airborne back-ups is caused by issues in-between you and your destination. En-route storms, occasional ATC radar or communication outages, or simply the convergence of aircraft onto departure and arrival corridors surrounding a busy metro area can also grind traffic to a halt. I can’t tell you how many times I’ve explained this to passengers during a delay. Fliers will tell me, for example on a flight to New York that has been delayed due to weather, that “I just got off the phone with a colleague in NY who tells me it’s a gorgeous day there, what are you guys trying to pull here?” In this case I might explain that the reason for the weather delay to NYC is rooted in a line of thunderstorms 300 miles west of the city over Pennsylvania that’s clogging the flow of arriving traffic. That’s the nutshell of the story. We’re going to dissect this further later on.
There are times when airspace isn’t congested at all, but the airport itself cannot handle the amount of traffic scheduled to arrive there. Maybe there’s a temporary runway closure due to construction or snow removal. Sometimes an aircraft lands with a malfunction that disables it right smack in the middle of the runway for hours at a time. Wind is another huge factor. Airplanes generally must take-off and land into the wind for performance reasons. If a particular airport has three runways, one facing north/south and two of them facing east/west, only one runway may be available for use if winds are sustained out of the north or south. Normally taking off or landing with a crosswind isn’t a problem, but when the winds really start howling, the limitations of the aircraft don’t allow for it. It’s also not uncommon for the direction of the wind to suddenly shift 180 degrees, causing the flow of traffic into and out of an airport to be reversed. When the tower decides to switch runways, intense coordination with aircraft and the air traffic controllers who handle arrivals and departures in the airspace surrounding the airport ensues. This can cause delays. Imagine rush hour traffic on your most demised local freeway all of a sudden being forced to turn around and drive the other way! This is the plight of such a situation. When prudent, these runway switches will occur during lulls in traffic, but that’s not always possible.
Low visibility can also limit available runways for use. When fog, low clouds, or heavy precipitation move in, pilots and air traffic control (ATC) really start earning their paychecks. Most major airports are equipped with an instrument landing system (ILS) that allows pilots to guide aircraft accurately to the runway with both horizontal and vertical cues. Under most circumstances, these systems allow pilots to use such guidance down to roughly one-half mile of visibility. However, when visibility drops to less than that, additional guidance such as enhanced runway lighting is required to land. Many airports may have multiple runways equipped with an ILS, but fewer runways with the equipment necessary for extreme low-visibility landings. This is yet another limiting factor that can cause delays revolving around the capabilities of the airport itself. I’ll also note that there are a handful of airlines that are not approved for these extreme low-visibility landings. In this case, a delay can be related to their limitation. Poor visibility also hinders the timely flow of aircraft due to the increased workload of ATC having to more closely monitor aircraft spacing.
To begin the discussion of complex, more difficult to understand delay reasoning’s, it is important to consider the structure of the ATC system, which is responsible for the safe separation of aircraft. Alongside the ATC tower at the airport, which handles takeoffs, landings, and ground movement, immediately outside of their sphere of influence lies the jurisdiction of the TRACON, or Terminal Radar Approach Control. These folks handle the flow of air traffic into and out of an airport area. For example, New York TRACON has many controllers who each control a given sector of aircraft that are arriving or departing the general New York metro area. This includes major airports such as EWR, LGA, and JFK, among others. Pilots refer to these controllers on the radio as “New York approach” if arriving, or “New York departure” if departing the area. Once aircraft leave the TRACON control area, they are handed off to the guidance of the ARTCC, or Air Route Traffic Control Center. Pilots address these controllers as “Center” on the radio. There are multiple centers across the country with much broader control areas compared to TRACON.
And then there is the FAA ATC System Command Center, or ATCSCC. This facility coordinates the balance of traffic demand with available capacity. Capacity is the key word here because if there were unlimited capacity, we wouldn’t really have much to talk about. The ATCSCC traffic management specialists will adjust traffic flow so as not to exceed capacity in a given local through the issuance of a delay program. Whether it’s a runway closure, weather issue, radar outage, or some sort of national emergency, a delay program is a dynamic means by which to regulate the flow of aircraft into given airspace. There’s a lot of automation involved in calculating the ideal balance of restriction vs. efficient flow of aircraft, but alongside this is close coordination of the command center with the various ATC facilities across the U.S. as well as the airlines themselves. As a matter of fact, every few hours, there is a teleconference between the command center, various ATC facilities, and the airlines themselves to hash out issues and tweak delays as needed. And here in lies the nitty-gritty details of one of the most commonly misconstrued facets of delays – which flights have priority during a delay? We need to go deeper yet to answer this question.
At peak travel hours, it’s not uncommon for more than 5,000 commercial flights to be airborne at once. It’s impossible for this many aircraft to fly randomly all over the sky without having separation issues. A coordinated and standardized system of both filtering aircraft into and out of the airport vicinity as well as on and off the en-route structure is necessary. To do this, aircraft follow published arrival and departure routes as well as airborne highways, termed airways. Although analogous to on-ramps, off-ramps, and highways, these published airborne routes are defined not by pavement markings but rather navigational facilities or latitude/longitude coordinates that are called intersections. Each facility and intersection is given a unique name. The navigational facilities are generally VOR (very high frequency omni-range) facilities that emit a radio signal in all directions for aircraft to navigate to with onboard navigation equipment. Many times aircraft navigate instead to the coordinates of these various VOR’s via GPS. In any event, coordinates are strategically placed to form standardized routes that allow for aircraft into and out of an airport area to stay safely separated from one another, as well as from threatening terrain. Most major airports have multiple arrival and departure procedures, each given a unique name that is assigned to flights depending on their inbound or outbound general direction. For example, aircraft headed to EWR from the south generally follow the PHLBO arrival procedure; while aircraft inbound from the west navigate on the Williamsport arrival.
During the en-route phase of a flight, aircraft follow airways that act as highways in the sky. Just like the highway system on the ground that doesn’t necessarily allow for a straight line to your destination, the same is true with airways. Often, numerous different airways are followed between destinations. Occasionally aircraft will be granted permission from ATC, depending on airspace congestion, to fly off of these airways to avoid weather or proceed more direct to a destination. You might hear the pilots make an announcement that an early arrival is expected because a “shortcut” has been given, which means they have been cleared off of the airway.
Although this standardized system allows for high levels of safety with regard to aircraft spacing, it is prone to backups similar to highways. Frequently, especially in the summer months, thunderstorms may block airways or departure/arrival routes. In this case aircraft must be moved away from these routes, similar to when a broken down car on a multi-lane highway causes traffic to merge into the remaining lanes. When this situation develops, the workload of ATC increases immensely and coordination of air traffic that was once nicely spread out must now be condensed while maintaining separation standards. In order for this to work, the capacity in the effected airspace is reduced and aircraft are often times held on the ground longer than expected awaiting a clearance to fit into the constricted airspace. Under these circumstances, traffic starts to backup, and a series of predictable events starts to unfold.
On a reactionary basis, when either airspace or an airport is overwhelmed, airborne aircraft are delayed in various ways. Unlike cars on a backed up highway that can simply stop to wait out congestion, airplanes are obviously not capable of that. ATC will sometimes start by issuing “delay vectors” or mandating a speed reduction. Delay vectors result in an aircraft flying off course to various assigned headings often resembling “s-turns”. Obviously reducing speed is another way to effectively delay entry into over-crowded airspace. When those fixes don’t suffice, the next step is for ATC to issue holding clearances. A specific point in space, altitude, and time expected to hold is assigned to affected aircraft until they can be squeezed into backed-up airspace or a closed airport. Clearly aircraft are limited by the amount of fuel onboard, and therefore cannot hold indefinitely. If necessary, a flight may divert to an alternate airport to get refueled and wait out the delay on the ground. Once a flight has diverted, the good news is that ATC prioritizes its to-be-continued journey ahead of most other delayed flights.
On a more proactive basis, the ATCSCC described previously, will begin the process of issuing a Ground Delay Program, or GDP. GDP’s are designed to delay a flight to its intended destination by keeping it on the ground rather than delaying it in flight. When a situation is anticipated to cause maximum capacity in a given ATC sector or airport to be exceeded, a computer generated average delay and maximum delay is formulated. This calculated delay time is then appended to the affected flights original departure time. The term given for the newly delayed departure time is what we call a “wheels-up time”. For example, the command center may issue a ground delay program for traffic into LaGuardia (LGA) airport that extends from noon – 4pm due to anticipated congestion. Flights that are scheduled to depart and arrive into LGA at noon or after are then given a delay. The closer to the beginning of the GDP a flight is, the more likely it will have the average delay time appended to it, say 45 minutes. But as flights begin to pile up, a snowball effect begins, and flights that are scheduled to arrive later into the GDP scope will trend towards getting the maximum delay added to their scheduled departure time, say two hours. It’s all a function of the snowball effect, an obvious result of the gradual piling up of flights into congested airspace.
If airspace or an airport becomes overwhelmed to the point that a GDP program doesn’t suffice, the dreaded Ground Stop will be issued. This halts aircraft on the ground with no estimated departure time given. Ground Stops are often issued for duration of an hour, with an update as to whether or not the stop will be extended at the end of that time. Often it’s quite predictable when severe weather is the culprit as to whether a stop will be extended for hours at a time or not. And therein lies one of the most frustrating aspects of delays for passengers - the fact that during severe congestion, no one really knows when a flight will depart; it’s a very dynamic situation. Sometimes a Ground Stop is cancelled before its scheduled expiration, which is good news. Other times however, it’s extended indefinitely - clearly bad news. The pilots, gate agents, and even air traffic controllers are often simply “along for the ride” and can only estimate when recovery is possible.
Once a ground stop has been cancelled, generally a GDP program follows. The pilots will make a radio call to ATC who will issue them an Estimated Departure Clearance Time (EDCT). This time is just that, an estimated time that ATC foresees an available slot for a flight to fit into the mix. About 15 minutes prior to the EDCT time (generally well into the boarding process), the pilots will make one more call to ATC to get a firm wheels-up time. This time may differ slightly in either direction from the EDCT time, and if it’s later, a slight tarmac delay can result. Although frustrating for passengers to wait for a takeoff clearance on the taxiway as opposed to in the terminal, it’s crucial to be in position to make the assigned time. If the wheels-up time is missed, there’s a good chance for a much longer delay, as someone else will be given your valuable slot! I might add that even this wheels-up time is subject to change at the last minute if ATC decides there is still too much congestion. Ground stops are always looming and can strike without warning. It’s really a gamble that we take as pilots and airline personnel – we can only act on the information at hand. The saying in our industry is always “hurry up and wait”.
Lest we not omit delays placed on flights due to departure constraints either. Weather, such as thunderstorms, or airspace congestion during peak hours can constrain the smooth flow of aircraft leaving an airport. The New York metro area and the northeast corridor as a whole are a great example of airports that are sensitive to departure related backups. Since all flights are headed to a destination that is either north, south, east, or west of their starting point, standardized departure routes are assigned to flights that orient them in the correct direction to join an airway that will be followed to the destination. Although EWR, LGA, and JFK each have their own unique departure routes, often times they all end with the same intersection that joins onto an airway. This means there is a chance that multiple aircraft from each airport could end up flying over the same intersection at the same time - clearly a problem. The fix in this case is not as simple as turning aircraft away from each other, as the busy NYC airspace in particular is highly constrained with both arrivals and departures flowing into many different airports. Available airspace to maneuver aircraft slightly off-course simply is not a luxury there, or at many busy airports across the U.S. In order to space flights departing from multiple airports that are all headed to the same general point in space, TRACON coordinates with the tower at each airport to launch aircraft in a staggered fashion. TRACON may say, for example, that due to high demand over a particular intersection, 15 miles of in-trail separation is needed between aircraft. The tower at each airport in the area will then limit the flights headed that way to depart such that 15 miles of space is cushioned between them. This requires immense coordination on the ground during peak times, as large amounts of flights headed in various directions are all taxiing to the runway at the same time. Ground control will sequence aircraft in line for takeoff in such a way that say, every other flight that departs is headed west. It’s a big puzzle that sometimes leads to gridlock on the taxiways. To get a sense of the gravity of this coordination nightmare, there are some great audio recordings available online of typical ATC/pilot communication during periods of immense gridlock.
Everything previously mentioned can occur even on the nicest of days in busy airspace. Throw fowl weather into the mix, and the problem is exasperated. Thunderstorms can render a crucial departure route useless. Aircraft assigned to that route now must be re-routed onto another procedure that often times is already saturated to begin with. It’s times like these that tarmac delays can get down right ridiculous and bring about such things as the passenger bill of rights. But that’s an article in-of-itself.
Even though we’ve covered a lot of ground when it comes to the complexities of delays, there are still other facets that could be discussed. I’ve laid out the most common issues in an effort to illustrate the big picture and hopefully give you a better understanding of what happens behind the scenes. But I want to answer the burning question of who gets priority in these situations? Big planes? Little planes? International flights? There’s really no concrete answer. In some ways, it differs between airlines, although some general guidelines do exist.
When there is a GDP in effect that limits available slots into an airport, no one airline or flight is given priority by ATC directly. The only flights given priority by ATC are those flights in the air that may be declaring a low-fuel emergency. The aircraft on the ground are all subject to the same GDP program and must wait out their delay time. However, the airline has the ability to swap slots if necessary. For example, there may be a flight with a wheels-up time at 1pm, and another with one at 2pm. If the flight slated for a 2pm takeoff is deemed “high priority” by the airline compared to the 1pm flight, they may slot-swap it. This would allow the flight delayed until 2pm to depart at 1pm. The folks on the 1pm flight that just found out they’ve been delayed another hour aren’t going to be too happy! Generally these decisions, made internally at airline operations, are done for the better of the masses. There are many parameters airlines use that automatically prioritize flights. Such factors may include the number of connecting passengers, flight crew mandatory rest issues, and aircraft positioning. If a delayed flight has for example 50 passengers that are all connecting to once-a-day international flights, the airline may want to ensure they make their connections, as re-booking them is a nightmare. If your flight crew is close to exceeding their limit on allowable duty time for a given day, your flight may get priority to prevent a cancellation. Also, crewmembers often are flown as passengers to be in position at another airport to operate a given flight. The airline may prioritize such a flight to ensure those crews are in position to work their next flight with minimal delay. It’s a complex and equally dynamic equation that is always being solved during times of high traffic congestion.
If you’ve made it through this lengthy read, my hope is that perhaps you’ll approach a future delay experience armed with knowledge. If I can alleviate some frustration through enlightenment, I’ve accomplished my goal. In a future article, I’ll provide insight on how best to avoid, or at the very least, adjust your itinerary when faced with flight delays. Safe travels and happy holidays.