There are plenty of people out there who think the airlines are trying to “cheat” when they pad their flight schedules to account for delays, but I don’t get it. That is not what’s going on here. What they’re really doing is trying to delicately balance operational efficiency and customer satisfaction, believe it or not.
Scott McCartney at the Wall Street Journal took this on recently and started digging around. Evan at FlightCaster took it one further and analyzed all flight times. So what’s the result. Since 1996, schedule times have bumped up 8 minutes on average.
Of course, that number isn’t very relevant. You need to look at the biggest pain points to see where the most egregious increases are. JFK saw an average 27 minute increase on departures while Continental’s late afternoon departures from Chicago to Newark bumped up a whopping 45 minutes. My question is . . . so what?
Some passengers think this is an evil plot to make it look like they have a stellar on time record, but that makes absolutely no sense. First of all, think about it from a passenger standpoint. If you book a flight leaving at 8a and arriving at 4p, isn’t that what you expect? Does it matter if you arrive a couple minutes early?
Sure, it’s annoying when you arrive 45 minutes early and your ride is nowhere to be found, but that’s an anomaly. Think about it from the airline side. Do they want to pad their schedule? Absolutely not. If they could schedule their flights to take less time, they could cram more flights into a day and bring their costs down. That would be fantastic. So why do they pad?
Well think about it. Gate-to-gate flight times have actually gone up in the last 15 years. As airports have become more crowded, they’ve been forced to spend more time taxiing or holding in the air. And it’s not just blanket changes like that. Weather patterns matter as well. If you have particularly stormy time of year, you might build in some extra time for circling. During the winter, the airlines have longer block times on westbound flights across the country because the headwinds are stronger.
This is a science. The airlines have to delicately balance the desire to cram as many flights in a day as possible with the need to present passengers with an on time experience. People and computers are constantly working on trying to find the optimal block times, but that’s easier said than done. Things change often, and that means they can never be perfectly right. But they certainly aren’t trying to cheat.
37 comments on “Airline Schedule Padding Isn’t Cheating”
Airlines probably would’ve saved themselves some grief years and years ago when all of this was devised to instead of of naming specific takeoff and landing times, use windows: “Must be at gate by 3. Pending no weather, FAA, or other delays, flight will be airborne between 3:15 and 3:45. Pending no delays or diversions, plane will land between 5:30 an 5:45.” Or whatever. Offering a precise time (“Takeoff: 3:17 p.m.”) sets up unreasonable expectations in an uninformed public.
Now perhaps as far as the FAA, airport, airline, and pilots are concerned, the “real” targets are 3:17 and 5:41, for the sake of airport and flight plan operations, but passing that information along to the general public just seems like a bad idea. I’m instantly reminded of a scene in MKE, waiting on the very short MKE to MSN flight (20 minutes of flight time in an RJ), where a barely cogent senior citizen went off for 15 minutes on the poor gate agent about “why does my ticket say the flight is 45 minutes, then?” when she indicated the aircraft would be slightly delayed, but no worries since it was such a short flight.
Yes, the airlines should always defer to giving the public more information rather than less, but…
I agree that the padding is not really cheating. In January we were on a US Airways flight from Chicago to Phoenix and it snowed heavily the night before into the early morning hours and the flight was scheduled for 4 hours. Once we deiced which took about 20-30 minutes our flight time to Phoenix was only 3 hours in the air giving us an extra 30 minutes in Phoenix. The captain even said they add the extra time just incase of these situations.
Its actually pretty smart on the airlines part to do this in case of those situations because people get anxious/fustrated because they feel they will miss any connections on their flight which would give them some comfort. I just remember there was a woman sitting near us kept bothering the flight attendant when deicing about if she missed her connection they would be in a world of hurt….silly flyers
In the old days airlines wanted to show they were faster then the other guy at getting you from point A to point B. You know, 36 hours instead of 48 hours. Now I would rather know a ‘real’ time then some on paper time.
If they were padding the time so it looked like their flights were never late, then why do we still have late flights. So that should show they are just trying to give a more accurate time. It helps their planning also to know when flights really will depart and arrive.
I remember when the flight time between SFO and LAX was under an hour. Now the airlines show their flights at anywhere between 1hr 20mins to 1hr 35mins. The two cities haven’t moved farther away from each other, it’s just the sky has gotten more crowded, so why say it’s under an hour when it never will be.
What’s interesting is when a flight leaves well behind schedule and arrives “on time.” Had a NW captain tell me once that our late departure time (45 minutes) was made up by 1) getting clearance to fly the route as the crow would fly it, and 2) going faster and burning more fuel.
Flying way back in the 80’s (wheels up to touchdown) sure seemed to be faster, even on a present day flight where we aren’t put in a hold pattern to wait for a runway. God knows I was much more impatient back then so noticing a difference is remarkable IMO. So all things being equal why would the same route “feel” longer? My sneaking suspicion is they fly the aircraft slower these days to save fuel burn, which has been built into the “padding.”
As a side note, remember those “cowboy” captains that would start winding up the jets while the aircraft was still turning the corner to line up with the runway? Time seemed a higher priority in the past.
I don’t mind at all.
Also, you don’t mention it, but this becomes much more important when connecting.
I rarely connect and fly a lot of transcons. I enjoy getting home early.
Would much rather have a realistic time schedule in order to make my connection than have a “hoped for” flight time that leaves me holding the bag at my connecting airport!
Actually, I’ve had flights arriving 25-45 minutes early (the longer the flight, the more egregious – trasnsoceanic flights are the worst) on a fairly regular basis, and it is obnoxious, both as a passenger and as someone picking up people at the airport. I find it, surprisingly, to be more annoying than arriving late. I’d rather be 5 minutes late because it took us longer than expected to taxi, than 40 minutes early because they padded their schedule.
If they need to pad their internal schedules (turn-around times) then fine, but there’s no reason to take us along with them.
Someone was recently complaining to me about this “plot” of the airlines to fool everyone. The example I shot back was this:
How long does it take to travel from downtown LA to Santa Monica by car? Well, technically, you can do it in 15-20 minutes if you’re the only car on the road. But, we all know that’s not the case. Depending on the time of day, it can actually take you an hour or more. So, if you were unfamiliar with LA and asked someone that question, would you want them to tell you 20 minutes? Of course not. You’d want a real world time estimate that reflects actual conditions so you could plan your trip appropriately.
The problem is with the FAA, and the public paying too much attention to the ‘on time’ statistic. If the FAA classified a flight as late if it departed OR arrived more than 15 mins past schedule, the statistics would be much more telling, and would catch the artificial padding, problem solved.
It’s just a way of cheating the numbers in a world where it’s very easy to compare. Claiming we’re 93% on time, is really like advertising a cheap fare with a ton of surcharges. We all know it isn’t true, but put up with it.
Cranky, I’m sure you have a better insight into this stuff than most of us, well, me anyway. We all have experienced schedule issues, and often we think it’s some manipulative scheme. Perhaps we’ve experienced some unbelievable early arrivals and long waits to get a gate, or else a terrible delay and no reason given. We chalk it up to the poor customer-service we’ve come to expect from this industry, yet operationally, things may have gone the best as was possible. Still, it comes down to whether or not you did the best you could for your customers.
I take these stats with a grain of salt, as I do with airline web sites explaining why a particular flight may have left or arrived late. {United, for example, should be ashamed of its assignment of “reasons” for why a flight was off-schedule. Forget it. They mean nothing.]
Personally, I believe these stats are primarily for deciding staff bonuses and for company advertising bragging rights. For the customer.., meaningless.
What really does a customer want to know? How about simply the percentage of time, by flight, an airline closes the door at origin at the listed departure time, and the percentage of time it opens the door at destination at the listed arrival time. Best if it showed both early, and late. Both can create problems for us.
The fact that an airline may push back and then have you sit somewhere for an extended period of time before you take off, and the reverse on arrival at destination, may be operatational, beyond my control, and sometimes beyond the airline’s control, or maybe it is arbitrary, maybe “penny-wise, pound-foolish,”, or maybe simply a sloppy way of operating.
Just be honest with us about what’s going on. Maybe even ask us what we want. I know, that’s customer-service, something woefully missing in this industry.
The people that complain about this are the same ones who wonder why it takes 3 hours to fly from New York to LA and 9 Hours to get back. Clueless.
But I recently experienced three flights, two trans-Pacific and one trans-Atlantic, where the in flight display showed us traveling at a blistering 425 MPH. All three flights were at least an hour late. Same thing with a Tokyo to Singapore flight – 90 minutes late and 420 MPH. I don’t think that was headwinds…
Some honesty would, indeed be nice.
You’ve been putting out some great information – makes me want to fly Air New Zealand tomorrow!
So, in a world where manufacturers manage JIT supply, retailers reduce their costs and supply time through using efficient fulfilment, international enterprises speed their processes through technology and lean process management, the airline industry does, well, nothing. OK I’ll grant that they are at the mercy of an inefficient ATC system, that airport operators and airlines themselves cram their schedules with small aircraft at busy hours, that there’s an almost total lack of sense in the industry … etc. Oh and yes, snow, wind and rain really do upset things all year round. But this article strikes me as one big floppy excuse for inefficiency.
It is a fact that padding is happening. As a pilot for a major domestic carrier, virtually every single leg I fly is padded to the extent that we arrive consistently early by 10 -15 minutes. From my vantage point, the on-time statistics are becoming a race of which airlines can pad the most but still remain within some ill-defined, time boundary without being too outrageous about it.
The padding is coincident with reduced block hour scheduling, which results in lower aircraft utilization rates. Instead of an aircraft flying around 11 hours a day, it’s down to 9 or 10.
At my carrier, our on-time performance is significantly improved, but it is not due to improvements in operational methods or management. It is strictly due to padded times. I would venture to say that when our utilizations rate return to higher levels (and rest assured that they will) the on-time numbers will reflect it.
Mantini wrote:
There’s a factor at work on those long trans-continental flights that the airlines really have little control over – wind speeds in the jet stream. I’m assuming the block times for a flight from, say, DFW to LHR are computed based on some kind of average, using expectations of headwinds, tailwinds, etc. If those end up stronger or weaker than whatever average was used, you’re going to end up with a difference between you scheduled block time and actual flight time; the longer the flight, the more magnified the effect. It’s not some evil plot by the airlines in this case.
A wrote:
Funny, I had this experience on a flight from ORD to DFW just on Sunday. I thought it was pretty cool, personally.
@ Paul:
It’s absolutely true that the long haul flights are much more susceptible to en-route winds. But as other have stated, the padding is happening in addition to jet stream behavior.
More often than not, an aircraft is not cleared for takeoff immediately, especially at busy airports. Crossing runways, taxiways and airspace ahead have to be clear. Additionally, takeoff performance can be a factor when leaving from shorter runways. You won’t see a rolling take off from Orange-County John Wayne airport, for example.
Mantini wrote:
Okay, you can sit in your cramped airline seat for 45 minutes after the plane has arrived. The rest of us can sit in a slightly more comfortable seat at the airport.
Scott wrote:
Why would you care if it left late but got there on time? Seriously it would be sweet if back in the days of the Concorde if one of the 747s got delayed they just dropped in two or three concords two or three hours late, and got the customers there on time.
JayB wrote:
I bet you they have. People want to know when they’re arriving, and when they need to be at the airport…
Hunter wrote:
Good point since I used to live in L.A. and know the traffic has gotten worse since I moved and the last time I was there it took longer to drive the same routes I used to. So why would people think air travel would be any different.
How about boarding time padding? On two recent flights (LAX–BOS on United, PHL–LAX on US Airways) I arrived at the gate at the time printed on my boarding pass (30 minutes prior to departure), to find that the flight was almost completely boarded and I was one of the last people to get on the plane. Is this a new trend? And why does the printed boarding pass not reflect the earlier start of actual boarding? I assume they won’t close the door on someone who arrives at the printed time, but that person might end up having to gate-check their bag even if they have boarding priority.
Rob Lipman wrote:
That’s funny but so true. Many moons ago when I worked at TWA we got that question alot. Or one of my favorites was ….’why does it only take 2hrs to fly from Paris to New York, but 3hrs from NY to Los Angeles’. Like you said, clueless.
Mantini wrote:
Arriving early would be fine with me if the gate was empty when we landed. But the pilot bragging about being 30 minutes early when the wheels hit the runway, only to be told 5 minutes later that we have to wait 20 minutes for the plane at our gate to leave is infuriating.
@ David SFeastbay:
Do these same people wonder if there is a time machine that allows you to arrive in Los Angeles before you left Sydney?
There are a few stones left unturned in this discussion:
a) Travel Agency Displays.
As mentioned earlier, in the days of travel agency dominance each airline made it’s bread and butter by trying to be “Line One, Screen One” of an availability display. That meant scheduling “blue sky” for the shortest possible elapsed time in order to get the most favorable screen position in front of travel agencies.
Add in to that the 15-minute window for DOT ontime statistics and virtually every flight in the system operated late. Not good for business. In a sense, THAT was really cheating because the airlines never took in to account the realities of weather and traffic delays when publishing their schedules: they over promised and under-delivered and have finally decided to be realistic with their schedules.
b) Fuel Economy
Particularly when Jet-A kerosene was in the $3/gallon range, airlines began flying slower as well as boarding reduced reserves to save weight and expense. Even in “blue skies,” flying at 510 instead of 525MPH means getting there a few minutes later. Build that in to the schedule as well.
c) Directional Headwinds, Seasonal Weather and ATC
Some headwinds can slow a flight to a virtual crawl but they don’t always show up every day of the year. One flight at 425mph one day might fly like a bullet two days later as the jetstream changes its pattern over the same section of air space. Still, better to build in some kind of aggregate between best and worst cases according to year-before data.
I used to work in Baltimore where the flying time between BWI and JFK is 25 minutes. The flight, however, was scheduled for 90 minutes to allow for ground traffic at JFK which , in the winter, is oftentimes fairly optimistic. New Yorkers will sniff out a lie real quick if some fool schedules a 4PM departure to BWI or DCA with an arrival of 4:45! They WILL accept a published 5:30PM arrival, however cuz it’s still shorter than four hours on the train or driving, especially in bad weather.
d) Gate Management
Plane arrives “early” and the gate is occupied. Been there? Stretch out that schedule so hopefully when you get there you have a gate to go to. Not by any stretch an economical or desirable way to do business but there it is. Other measures have included “pin-wheeling” the hub instead of scheduling tons of arrivals at the same time. Some connections may be longer than others but usually there’s a gate for everyone and the most popular connections haven’t really been affected.
The FAA and airlines have long pointed fingers at each other regarding the state of the air traffic control system so let’s not even go there. The reality is it stinks and no one is willing or able to pay for the upgrades it needs so current service has to find a way to work under antiquated controls. Read: flight delays, ground stops, holding patterns and other bugaboos that also contribute to “left late, arrived on-time or early.”
e) Crew Salaries
The longer it takes to get somewhere the more the crews get paid, also not a desirable way to do business but, again, there it is. Longer flights mean less overall service to offer (can’t offer five flights in the same space that now can only support four) but crew costs stay the same? Something’s gotta give….hello parked aircraft and furloughs.
What needs to happen to fix all this? Take your pick….a better economy, cheaper energy and 21st century air traffic control systems. Your guess is as good as anyone else’s which one will happen first.
The best recent inventions to deal with the problem?
Text alerts from the airlines and travel agencies, services like Cranky Concierge and the cell phone parking lot to let your meeting party know when you’ve finally hit the gate so they can meet you at baggage claim!
“Airline Schedule Padding Isn’t Cheating”
Right! It’s honesty.
There’s some cause and effect being lost here. They have to pad the schedule because they’re overscheduling with respect to the airfield or airspace capacity. So the gate-to-gate time goes up. And then one little glitch, and it all goes down the tubes.
A solution would be to use larger planes at a lower frequency, but people want the flexibility that comes from having an RJ every 45 minutes. And then they complain about flying on a RJ. But they’d rather have that, and padded times and the whole system balanced on knife’s edge, rather than less frequency.
Nicholas Barnard wrote:
You care if an airline leaves late, and arrives on time, because then you know that it’s contributing to a statistic that is meaningless. If the FAA want the on-time statistics to mean anything, they need to take into account departure and arrival times.
I realize that trans-oceanic winds can make a huge difference, but let’s look at the three scenarios
Scenario 1: Planes leaves on time, arrives on time. This is good, you get 100% in the FAA “on time sweepstakes”
Scenario 2: Plane leaves late and arrives late. Because you arrived late you lose a point in the FAA “on time sweepstakes”
Scenario 3: Plane leaves on-time, and arrives late. ATC, weather and other things can cause this, but in terms of airline operations, they were doing their best although they lose a point in the “on time sweepstakes”.
Scenario 4: Plane leaves late and arrives on time. Because the airline fudged the enroute time numbers, they get a point from the FAA. Basically they lied in order to pass the “on time” test.
Airlines are in the business of ‘looking good’ with respect to the ontime numbers, and #4 makes them look like they scored, when in reality all they did was move the goalposts.
If you’re going to measure something, you should measure something where the participants can’t change the rules.
“Think about it” three times in seven paragraphs. Annoying; think about it.
I remember taking Jetblue from Buffalo to JFK back in 2000 and it took 50-55 minutes. Now it takes at minimum 1:20 or even 1:45 on Delta Express. At one point, Delta scheduled one of their JFK-BUF flights as taking 2:00 (although that may have been on a DH2).
@ Scott:
So basically you’re saying the FAA shouldn’t publish ontime stats?
If the FAA isn’t going to own their mistakes, then they shouldn’t expect the airlines to own theirs n
David SFeastbay wrote:
Yeah, and they even marketed airplanes that were faster. The Convair 880 and 990 were marketed because of how fast they were. They died a painful but quick death when fuel prices skyrocketed in the 70s. Nobody wanted a fast airplane that burned fuel like that machine.
A wrote:
Beats me! Sometimes they do slow down a bit to save fuel but we’re usually talking about a difference of minutes. If the flight is particularly turbulent, they slow down as well to the optimal speed for reducing the impact of turbulence.
Mantini wrote:
I’m pretty sure that puts you in the minority. Most people would rather be early than late. Also, being even just 5 minutes late starts to impact the next flight as well.
JayB wrote:
I dunno – it seems to me that the customer really just wants to know if he’ll get to his destination when the airline says he will. I don’t know that the average flier cares about much else. I mean, sure, you’d rather have your flight leave on time than late, but as long as you get there on time, that’s what matters most.
JayB wrote:
If you’re pushing back and then not going anywhere, that’s almost always going to be ATC’s fault. You must push back and get in line to depart – you can’t reserve a spot. So you might see that there are 40 planes ahead of you, but you can’t wait until the line goes down. You have to get out there. If it’s the airline’s doing, then it may be to free up the gate for the incoming flight. For example, you might be in Raleigh heading toward Chicago and there’s an ATC delay in Chicago. You could wait at the gate, but if another airplane comes in and needs that gate, they’ll just push you back if they don’t anticipate the wait being that long.
Stephen wrote:
If you have a solution, I’d love to hear it. The problem is that much of this is out of their control. If a supplier fails to deliver on time to a manufacturing facility, the company can go elsewhere. If an airline isn’t happy because their supplier can’t provide enough capacity, they don’t have an option. Too much is out of their control.
Paul wrote:
I’m not sure what carrier you’re at, but yes, airlines will definitely do this if they’re running a poor operation. When I was at America West, we ran a terrible operation around 2000. Jeff McClelland came in as COO and added block time to give some breathing room. The operation bounced back and it gave them time to figure out what was going wrong. They ended up cutting back down again once they were able to make improvements. These moves aren’t taken lightly, because it costs money not being able to schedule these airplanes in the air for more time on more flights.
Ron wrote:
That’s a new one for me. I have no clue why they’d do that unless they thought it was going to take a particularly long time to board that airplane on that day. (I remember a flight to Phoenix from Omaha that was met with 18 wheelchairs – yeah, that took awhile.)
AirlineWONK wrote:
Thanks for your YouTube-like comment. It really added to the discussion tremendously.
Well think about it. Gate-to-gate flight times have actually gone up in the last 15 years. As airports have become more crowded, they’ve been forced to spend more time taxiing or holding in the air. And it’s not just blanket changes like that. Weather patterns matter as well. If you have particularly stormy time of year, you might build in some extra time for circling. During the winter, the airlines have longer block times on westbound flights across the country because the headwinds are stronger.
================================================
BINGO…………….Awesome article, Cranky! You explained things very well.
Furthermore, I INVITE your readership to “google” the amount of departures per HOUR at airports like JFK, EWR, DCA, LGA. Then add in some weather into those figures and see WHY airlines “pad” their schedules AT GREAT EXPENSE TO THEIR LABOR COSTS. Crews are paid by scheduled flight times, NOT ACTUAL.
I agree padding makes sense in that it takes into consideration all the eventualities that are part of every day travel. We need to address these issues like ATC, gate availability, etc. Heck, even the weather: we need to start flying on bio fuel…
One of my biggest gripes however is the absolute nonsense of all planes trying to get out of and into an airport in an unrealistic time-frame like between 6:30 and 9:00 AM. How many flights are scheduled to leave and arrive at for instance La Guardia at that time?
Someone mentioned “boarding padding” – and yes, this is a new phenomenon, especially on flights from outside the US to the US. First they build all those shops and bars/restaurants and there are all the new TSA/DOT rules so now we need to show up 3 – 4 hours in advance. Otherwise there is no time to shop…
Although I see your point in ‘padding’ I have to disagree.
If I am late to work 15 minutes every morning, I am still late. No matter how you slice it, it’s still late
Scott wrote:
Your statement make an erroneus assumption: that the aircraft is going to operated the same way regardless of the time it’s leaving the gate. This doesn’t always hold true in practice.
Before a flight, the aircrew programs the plane’s Flight Management Computer with something called a Cost Index. The CI is basically a ratio of time cost (wages for aircrew, etc) to fuel cost. The FMC uses the CI to compute the mach number and cruise altitude for the aircraft continuously throughout the flight in order to minimize the total cost of the flight. In general, going faster reduces time costs, but means burning more gas, so the CI is the airlines’ way of telling the aircraft how to balance the two costs.
On any given route, the airline will have computed a nominal CI that minimizes the cost of the flight for a normal flight day. However, some airlines also have a second, faster-speed CI they use when an aircraft departs at least a certain number of minutes late. This may mean that they’re not running the aircraft as cost-efficiently as they were before, but that they perceive that they make up the loss by having happy customers that appreciate getting to the destination by the published arrival time and won’t take their business elsewhere in the future, as well as by minimizing the impact of a late arrival on the rest of the airline’s schedule.
So, in short, the airline didn’t necessarily “lie”. It’s quite possible that they changed how they were operating the aircraft in response to the fact that they were running late.
CI also addresses A’s thoughts that flights “feel” longer, and his hunch, I believe, is correct. Recent rises in fuel costs mean that CIs are lower, and, therefore, so are airspeeds.
Boeing has a good article up that explains CI in a bit more detail: http://www.boeing.com/commercial/aeromagazine/articles/qtr_2_07/AERO_Q207_article5.pdf
A wrote:
I had this happen this past weekend and couldn’t help but smile. The captain took his little CRJ-900 and just whipped it around the corner and into a smooth take off. Smaller plane + long runway = do it as you will. I actually commented to the captain as I was getting off that I noticed it. He laughed a little to himself and then responded, “What’s the point if you can’t have fun?”
@ malbarda:
New York La Guardia Airport today is 78-85 flights per hour (arrivals and departures) in Optimum weather
Greg Edwards wrote:
That maybe depends on how one defines “realistic” if accounting for other factors like weather, traffic, etc.