Delays – three questions and many answers

8 March 2017

Last time you took a flight, was it delayed? By how much? And why? Simple questions to ask, but it’s much more complicated to answer them. First is the issue of whether you measure lateness on departure or arrival. This seems to be a question of personal preference and history – there is no right answer.

Measuring delays

Next, what does the scheduled time refer to – so for departures is it when the aircraft leaves the gate/stand or when it takes off? For arrivals, in addition to touch down and coming to a full stop, there’s another time to consider – the time that the door is open with an air bridge or stairs connected.

When does the flight start and end?

When does the flight start and end?

The standard airline measure is to use ‘gate-to-gate’ – the times that the aircraft starts and stops moving (however, for the purposes of EU compensation, under EC 261/2004, a court has ruled that the arrival time is when at least one door is open).

This use of ‘gate-to-gate’ helps to explain why your ‘flight time’ is almost always less than the time allowed in the schedule for the flight. It can take some time for an aircraft to taxi around the airport, particularly if the wind direction means that it has to go to the far end of the runway. And for some airports at some times of day, aircraft have to queue, each waiting until it is safe to take-off.

The gate-to-gate time for a flight includes taxi times. It will also depend on the wind (is there a head wind or a tail wind during the flight?) and on whether the aircraft can fly the most direct route between the departure and arrival airports. There are many reasons why this might not be the case such as military exercises, storms, strikes etc. There is some potential for the aircraft to fly a bit faster in order to make up time lost – but flying faster does cost more in terms of fuel.

Departures slightly more delayed than arrivals

On average, the calculated delay on arrival is slightly less than that on departure – 3.9 minutes less in 2016. Of course, this is an average and this figure will vary significantly between pairs of airports (and even more so for individual flights). For airport pairs (such as Frankfurt-Madrid flights or Amsterdam-Lisbon flights), actual average gate-to-gate times typically range from about 15 minutes less than the scheduled time to about 5 minutes more. The percentage of flights for a particular airport pair that exceed the scheduled time varies but is typically less than 40% and, for 2016, the average was 28%.

When does the flight start and end?

Each dot corresponds to all the flights by one airline on one airport pair. The graph shows the difference between arrival and departure delays (DDI-F), against proportion of block times longer than scheduled (BTO), by airport pair (analysis based on data from 10 major European airlines). So an airport pair in the bottom left has on average less arrival delay than departure delay and few flights that overrun the standard scheduled time for the flight.

What this means is that there is typically a bit of padding (called ‘buffering’) in the schedules to take account of all the above issues. This also helps the airlines arrive on time even if they depart a little late. How much buffering there is will vary between the different routes flown. Now this leads to the question as to whether the airlines are trying to avoid claims for compensation due to late arrivals by artificially padding their schedules.

The answer to this question appears in general to be “No”. The difference between departure and arrival delays has not increased significantly over recent years and is relatively small. The figures do not seem to support the theory that airlines are engaged in widespread manipulation of schedules in order to avoid paying compensation.

This is not so very surprising because if aircraft regularly arrived much earlier than scheduled then they would be spending much more time on the ground. This is because an aircraft arriving early would not, in general, be able to take off on its next flight any earlier than the scheduled time of departure (as the passengers would not be ready). More time on the ground is wasteful for the airline and it also makes life more difficult for the airport, which has a limited amount of gate and stand space available.

What causes delays?

By far the most significant cause of delay is ‘reactionary’ which refers to the late arrival of the incoming aircraft (or connecting flight/crew/cargo). This accounts for about 45% of the total amount of delays in Europe or five minutes per flight on average in 2016. This also explains why you are more likely to be delayed if you fly later in the day rather than in the morning. It also means that if we can minimise non-reactionary delays (called ‘primary delays’), particularly those early in the day, then that benefits not just that flight but subsequent flights as well. Eliminating a delay on the first one or two morning flights of a particular aircraft could benefit all the passengers using that aircraft over the next 15 hours.

Airlines could in theory reduce reactionary delays by allowing much more turnaround time between flights. However, this would significantly increase costs for the airlines, which have been striving to make the turnarounds quicker and more efficient. Aircraft and crew are only generating income when they are in the air and so having a lot of unnecessary slack in the system in order to guard against intermittent delays would mean more expensive fares. And as mentioned above, airports also want short turnarounds to maximise the use of their gates/stands.

When does the flight start and end?

The next largest group of causes for delay, accounting for slightly under 30% of all delays, is entitled “Airline”. This term covers a number of areas, including:

  • Delays in boarding
  • Baggage handling (including off-loading the baggage of passengers who are late at the gate)
  • Aircraft cleaning, fuelling, catering
  • Technical defects
  • Issues with documentation
  • Late crew boarding or crew shortage/illness

Any airline will confirm that it can be a real challenge to move hundreds of passengers (and their baggage) off and onto an aircraft and to prepare a jet for departure, all within a tight timetable and with strict security and safety regulations to follow.

Some of the departure delays may fall in the category of ‘Airport’ rather than ‘Airline’. For example, the entire baggage handling system might be out of operation, or a runway/taxiway might be closed for works. There could be problems with the fuelling system, or insufficient fire cover. Snow could make movement around the airport difficult, both for aircraft and for support vehicles. Airport and weather delays (except for those as a result of flow management – see below) account for about 7% of total delays.

Even when an aircraft is ready to depart, it may still be delayed by ‘flow management’ (ATFM). You may well hear that your flight is ‘waiting for its Air Traffic Control slot’ and this is explained in another article. In short, if there is a problem that will delay an aircraft either en-route or on the approach to the destination airport, then it’s safer and cheaper to hold the aircraft on the ground, not burning fuel, and give it a delayed take-off time.

EUROCONTROL as Network Manager looks at all the flight plans and identifies where and when there are too many flights/too little capacity. This might be as a result of a peak in demand (aviation has rush hours too!), weather problems, limited capacity at the destination airport or many other possibilities. In total, flow management delays account for about 15% of total delays, mostly relating to issues at the destination airport.

There are a few other causes for delays (such as those caused by delays in immigration, emigration or security) but these are relatively minor.

Looking at all the possible causes for delay, then the average per flight is about 11 minutes on departure and slightly less on arrivals. Naturally, this average is derived from a wide range of delays – from flights which are ahead of schedule to those which are several hours late. However, over 90% of flights arrive within 30 minutes of the scheduled time.

So what are we at EUROCONTROL doing to help? Well, reducing delays is one of ATM’s (Air Traffic Management’s) big four targets (the others being safety, cost efficiency and the environment) and so most of what we do has reducing delays as an objective. For example, we are working with airports to introduce Airport Collaborative Decision Making; this saves money and reduces delays – by improving the flow of aircraft around the airport and also by enhancing the predictability of traffic across the network as a whole.

When does the flight start and end?

Implementation of Airport Collaborative Decision Making at European airports

As Network Manager, we help match capacity to demand by working with the air traffic organisations in each country so they can assess how many controllers they need at any particular time. It is part of the process of Flow and Capacity Management that is essential for improving delay performance. This process covers everything from strategic planning to tactical actions on the day of operations. For example, if there is an expected disruption to the network (such as a strike or a major sporting event), then we will hold operational teleconferences to make sure that everyone can make informed decisions to help the traffic flow as smoothly as possible.

When does the flight start and end?

Network Manager Operations Centre

Where there are particular difficulties, we will work with our partners to address them. For example, the physical capacity (gates and stands) at some Greek island airports is limited. We have worked with the Greek ATC organisation, with the airports and with the airlines to optimise the use of the available capacity. This has included working with the airlines on their scheduling, in order to make sure that the stand availability at the Greek airports is used as efficiently as possible.

More generally, scheduling is an area on which we regularly help airlines by, for example, providing detailed historical information on taxi times. We have an extensive database on what has actually happened across Europe and this is invaluable in enabling airlines to make decisions based not just on the average flight or taxi times but on their statistical distribution, using data relating to all flights, not just those of that airline.

One area where EUROCONTROL, as a civil-military organisation, provides particular added value is by improving the coordination between the military, which has a vital mission to perform, and civilian flights, which have to avoid airspace while it is being used by the military. It’s vital to make sure that such airspace is made available for civilian use when possible so that flights aren’t unnecessarily delayed. Similarly, if there is a particular problem expected in the network (such as an ATC strike) then the military may be able to postpone an exercise in order to help out – even if the problem is in a neighbouring country.

Last year Europe had a record number of flights – over 10 million. That equates to an aircraft taking off or entering our airspace every three seconds (on average). Europe’s airspace is busy and its airports even busier, with building new airports or runways becoming increasingly difficult.

When does the flight start and end?

Busy skies

By historical standards, delays are actually quite low (although it may not seem like that if you are the one delayed!). However, with traffic is set to increase to over 14 million flights a year by 2035, there is a real risk of much longer delays becoming the norm once again.

All the players in aviation, including airlines, airports, air navigation service (ATC) providers, the military, national regulators, the European Commission (and its agencies) and EUROCONTROL are working hard to make sure that this doesn’t happen and that flights are safe, efficient, environmentally responsible and on time. Just as there is often no simple answer to the question posed at the beginning of this article, there is no simple solution to the challenges ahead. However, it is clear that working together collaboratively is the way forward.

If you are interested to see the delays on the busiest day in 2016, check out our animation below:

Related news