3
$\begingroup$

Statistically, are planes with advanced autopilot functionality less likely to be involved in fatal crashes? Answer only with actual statistics, not your personal opinion.

$\endgroup$
  • 2
    $\begingroup$ This is a very broad question to which there exists no single correct answer. I suggest you have a read here: eurocontrol.int/sites/default/files/publication/Hindsight/… $\endgroup$ – J. Hougaard Jun 29 '17 at 15:01
  • $\begingroup$ @J.Hougaard: Can you provide an overall idea of why the number of incidents/accidents could increase with some automation equipment? It seems the bottom line is positive, as automation continues to be implemented, and the number of accidents continues to decrease. Perhaps this is only a matter of appropriate training to take over automation when it fails? $\endgroup$ – mins Jun 29 '17 at 19:14
  • 2
    $\begingroup$ @mins You can train for taking over when automation fails, but staying proficient will be nearly impossible because you only get a fraction of the hands on experience you would if you were doing everything manually. Case in point: AF447 $\endgroup$ – J. Hougaard Jun 29 '17 at 19:21
  • 2
    $\begingroup$ The answer is definitively yes if well-understood and used properly. $\endgroup$ – Ben Jun 29 '17 at 20:44
  • 1
    $\begingroup$ The autopilot is as safe as we programmed it to be. The more factors the autopilot can factor in the safer it will be. But I think the real difference between safety occurs when something abnormal happens to the aircraft like broken control wires making the controls very unpredictable. A human can understand this or at least try. To make a computer capable of this is very very hard. In daily operation a computer does not make mistakes if properly programmed. A human is susceptible to skip steps by mistake in routine work. $\endgroup$ – Brilsmurfffje Jul 1 '17 at 8:03
2
$\begingroup$

Using e.g. Boeing's annual commercial aviation safety statistics, one can essentially show that with each new aircraft generation, the accident rate reduces. This answers your question about statistic involvement, but doesn't really confirm a causal relationship, since it's not only automation but e.g. also human machine interface (flight deck) design and system architecture that improves. However, I would assume there are quite probably at least some safety improvements due to automation in there. One would need to go about analysing the accident factors to prove this.

As an interesting aside touching on this topic, a former colleague of mine flying commercial jets recently offered his first-hand view on different aircraft automation levels in his blog.

While not a comprehensive answer as such, I hope the links allow you to pursue this a bit further.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.