2 ms·
Planes are hit by drones all the time especially while landing and taking off. Of course they are called birds. Suck a few geese into your engines and you have
by coldcode 10y ago
Planes are hit by drones all the time especially while landing and taking off. Of course they are called birds. Suck a few geese into your engines and you have to land in a river. I doubt a drone is as dangerous as a goose at 200 knots.
- jessriedel 10y agoYes, and the damage from birds is large: "Experts say bird strikes...cause an estimated $600 million a year in damage to planes." http://www.nydailynews.com/new-york/birds-plane-crashes-article-1.361189 http://www.nydailynews.com/new-york/birds-plane-crashes-arti... The cost of a single downed passenger plane is ~$1B, while the entire yearly value of the drone market is $50M. https://news.ycombinator.com/item?id=11425782 https://news.ycombinator.com/item?id=11425782 Even their usage/value of drones were kept fixed, then it could make sense to ban them even if drones only cause one airplane crash per decade. Obviously, this argument fails when you consider future value, but the point is that the numbers are not intuitive.
- dingaling 10y ago> while the entire yearly value of the drone market is $50M. That's purely a total-airframe valuation estimate, excluding revenue being generated by commercial use of drones. If you want to use that type of valuation then the loss of an A320 ( as reported in the original story ) would be around $98 million. http://www.airbus.com/presscentre/pressreleases/press-release-detail/detail/new-airbus-aircraft-list-prices-for-2016/ http://www.airbus.com/presscentre/pressreleases/press-releas...
- jessriedel 10y agoIt's the value of everything in the aircraft. For a drone, that's basically the cost of the drone. For the airliner, it's the $10M per human life. Apotential ban on drones would allow licenses and exceptions (thereby not destroying much commercial value), so the primary impact of the ban would be the amateur market. And for them, the cost of the device gives a good estimate of the value. Yes, we could expand our calculation outward and try to capture all the externalities created by those dead people, or the drone, or the irrational fear of drones. But this would have quickly diminishing returns.