Is there some source or research on crash statistics for each type of airplane? For example I want to know the number of crashes, fatalities, and the dates. I want to know which type of airplane has the safest record. Is there a correlation between the price of the ticket and the safety record?

  • 9
    There are a great many variables that go into defining what makes an airplane "safe" or not, and I think too many to be answerable here in a single post. The DC-10 had many early crashes, but after several design flaws were corrected, proved to be very reliable. The Concorde, on the other hand, never got the chance. Besides, I'd be more concerned about unsafe airlines as opposed to unsafe aircraft; poor maintenance or training can turn even the "safest" plane into an accident waiting to happen.
    – choster
    Commented Jun 3, 2014 at 23:10
  • 1
    Arguably one of the first - some of the 'Wright' planes NEVER crashed.
    – Mark Mayo
    Commented Jun 3, 2014 at 23:13
  • 10
    There is no such a thing as "safest plane". There are Russian planes and other planes, always take the non Russian. Commented Jun 4, 2014 at 0:53
  • 1
    @MeNoTalk Nothing against the Tu-154 ! It was not guilty for terrorist plots, cargo overloading or drunken air traffic control. For the only plane capable landing on unpaved (!) airfields in artic conditions it has a pretty good record. Commented Jun 4, 2014 at 2:16
  • 7
    "Too broad" makes no sense at all for this question. It asks for statistics. It's fair to discuss the problems with statistics but stats exist and there's no way this question is off topic or too broad. It doesn't make the internet better by closing this question. Commented Jun 4, 2014 at 3:02

4 Answers 4


All modern aircraft have to go through incredibly stringent safety checks and are essentially equally safe. Accidents are so rare that any apparent differences in aircraft safety are mostly meaningless statistical anomalies.

Airfleets.net has a chart of accidents per aircraft type, but it's not easy to make any practical conclusions about it. For example, the Boeing 737 has way more accidents than any other, but that's not because it's unsafe, it's because there are more 737s than any other jet plane. It's also been around since the 1960s, with many of the old planes still flying in places like Africa with poor maintenance and minimal oversight, which explains why every single 737 accident since 2000 (except one in northern Canada) has occurred in third-world countries. By some measures, in fact, the revised 737 "NG" (next generation) is among the world's safest aircraft, with one accident per 16 047 900 flight hours, meaning that, statistically, you'd need to sit in one for nearly two thousand years on average before it crashed.

There are also planes like the Airbus 340, the Airbus 380 "superjumbo" and the Boeing 787 Dreamliner that have never had a fatal accident, but they're also relatively new and comparatively rare. How do you even measure their safety?

All in all, the old saw about the taxi to the airport being far more dangerous than the flight still applies, but if you wish to minimize the odds of being in an accident, looking at the track record of the airline, the size of the plane (basically, the smaller the worse) and the conditions it has to fly in (mountains, bad weather, etc) will be more useful.

  • 4
    Your link to 737 accidents since 2000 only includes crashes that killed people, wrote off the aircraft or were otherwise unusual (see Wikipedia's guidelinesn air crash lists). The Aviation Safety Network has a more comprehensive list. E.g., a 737 cargo plane had a landing gear collapse in the UK in April 2014; a UK passenger 737 had a tail strike landing in Portugal in February; another was damaged in a heavy landing in the Azores, also in February. Commented Jun 4, 2014 at 7:31
  • 5
    But even the more comprehensive statistics I just quoted don't change the main message of your post. Commented Jun 4, 2014 at 7:32
  • I think the number of crashes per one million of flight could be a good measure after considering the safety of a airline.
    – MOON
    Commented Jun 4, 2014 at 9:34
  • 2
    Pegasus has never had a significant accident, the closest was a landing gear problem resolved with no casualties. en.wikipedia.org/wiki/Pegasus_Airlines#Incidents_and_accidents Commented Jun 4, 2014 at 23:51
  • 4
    "I think the number of crashes per one million of flight could be a good measure after considering the safety of a airline." Not really. It would bias in favor of long-haul craft and against short-haul craft. The vast majority of accidents happen during takeoff or landing, so comparing per-flight-hour statistics for a plane with an average flight duration of an hour to those with average flight durations of 8-10 hours wouldn't be very meaningful.
    – reirab
    Commented Jun 5, 2014 at 14:19

Rather than viewing safety as being a function of the aircraft, it's much more accurate to say safety is a function of the airline. To provide perspective, AirDisaster provides a ranked statistical analysis of selected aircraft by fatal accidents (accurate to 2004, so it omits the more recent models). Even the Concorde, the worst ranked plane, had a fatal accident only 0.001% of its flights.

Model                      Rate Events No. Flights Rank 

Saab 340                   0.33   3   9.0 Million   1 
McDonnell Douglas MD-80    0.45   9   20 Million    2 
Boeing 767                 0.46   3   6.5 Million   3 
Boeing 757                 0.56   4   7.2 Million   4 
Boeing 737                 0.62  47   76.0 Million  5 
Boeing 727                 0.66  46   70.0 Million  6 
Airbus A319/320/321        0.67   4   6.0 Million   7 
Fokker F-70/F-100          0.67   3   4.5 Million   7 
Embraer 120 Brasilia       0.71   5   7.0 Million   8 
McDonnell Douglas DC-9     0.76  42   55.5 Million  9 
Aerospatiale Concorde      12.5   1   0.08 Million 19 

If you compare that to its statistics on airline accidents, a number of airlines have higher rates than the worst aircraft. If you're concerned about safety, it's an irrational fear but you can try to mitigate that fear by avoiding airlines with a recent history of fatal accidents rather than trying to avoid specific aircraft.

  • 17
    I wondered when somebody would mention the Concorde. It went from "safest" to "least safe by a significant margin" in an instant. Commented Jun 4, 2014 at 1:05
  • 8
    And that's a perfect example of how useless these kind of statistics are. 30 years of incident-free service, one hull loss due to FOD on the runway. How does that make it an unsafe aircraft?
    – Paul
    Commented Jun 4, 2014 at 12:02
  • 2
    All this talk of the Concorde is making me incredibly nostalgic, poor Aviation SE enthusiast speaking here :( Commented Jun 4, 2014 at 15:57
  • 3
    @Paul Well, to be fair, it was the aircraft design that allowed said FOD to destroy the aircraft. IIRC, the FOD only caused a tire to blow. An aircraft should be able to survive blowing a tire without catching on fire and crashing.
    – reirab
    Commented Jun 5, 2014 at 14:25
  • Regardless, from a statistics standpoint you can't draw any conclusions about Concorde's overall safety from one incident. Commented Jun 5, 2014 at 19:38

The question is for a specific reason senseless, it is like asking what kind of radio is safest in an automobile.

The reason is simply that most fatalities are not caused by the type of plane, but by pilot errors, bad care or environmental influences (downburst etc.). There is no reason to believe some crashes would not have occured with exactly the same pilot, but a different plane. As a crash is always involved with very high speeds, the kind of body protection ("internal safety") a plane offers is also identical...that is to say none.

Naturally there are differences in safety: A small plane like a Cessna or Piper or military jets are much more unsafe than civil jets. But I presume you are asking for different types of civil jets.

What does exist are "safe" airlines: Airlines who have good pilots, new planes with good care and a "better safe than sorry" attitude.

Even if there would be numbers of crashes by plane type, they do not say anything. Because the importance of the type is neglible and crashes are very seldom, the numbers have a poission distribution: One airline X could have 15 crashes and airline Y 0 crashes without any difference in quality.

  • @pnuts Naturally a simplification, if I continue the analogy if you have a very powerful radio and increase the volume you are risking your safety. Back to the examples: The Comet problem was unknown until the reason was found, the Starfighter is a military craft. The entertainment system caused problems in the A320, a Boeing 757 and a Boeing 767. The ADIRU failure not only occured in the A330, but also a Boeing 777-200. And what was that with the battery problems in the Boeing machines ? Really, it is not something which really impacts the safety in comparison to piloting errors Commented Jun 4, 2014 at 1:48
  • 1
    @pnuts What was the incident with the Airbus A320's entertainment system? I'm aware of Swiss Air 111 but that was an MD-11. Commented Jun 4, 2014 at 7:39
  • 1
    @pnuts No worries -- I followed your point, I was just unsure about the specific example. Commented Jun 4, 2014 at 11:02
  • The Comet had rivets punched through the skin for the top navigator's window. A punched (as opposed to drilled) hole has a multitude of tiny cracks around just waiting for a few hundred pressurization cycles to start spreading.
    – Phil Perry
    Commented Jun 4, 2014 at 13:45

If you run the statistics, you will likely get different answers for:

  • Safety per trip
  • Safety per travel kilometer
  • Safety per travel hour

You need to think about what you really want to know. You might also think about the causes of air transport accidents that result in deaths. They tend to be more often lapses in judgement, communication or training by the flight crew. In private aircraft, deaths usually result from overconfidence in bad weather.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .