« CLC: Two Milestones!! | Main | Restoration Inspiration »
Friday, April 18, 2008
Ideal Recreation for Academics: Learn to Fly
Yesterday, six of us in San Diego (four profs and two other friends) rented a VERY small plane and flew up for the day to Mammoth. The mountain had marvelous snow and was pretty much empty of other skiiers. In the evening, we were safely back at home. It occured to me that each faculty needs a few pilots. When we were at Yale, the need was even more pronounced as New Haven had a lousy airport. More than once, my favorite prof in the world flew himself and four or five of our colleauges to conferences in places like Cornell, to which it would take otherwise many wasted hours (sometimes a couple of transition days really in order to attend a full day of conferencing) of train/connections/driving. And the best part: it really doesn't cost more than other modes of transportation. In fact, the trip to Mammoth, split six ways, cost less than the fuel it would take to drive up there. AND, it is safer than getting on the highway!
Posted by Orly Lobel on April 18, 2008 at 02:33 PM | Permalink
TrackBack
TrackBack URL for this entry:
https://www.typepad.com/services/trackback/6a00d8341c6a7953ef00e551e7600b8833
Listed below are links to weblogs that reference Ideal Recreation for Academics: Learn to Fly:
Comments
OK, this discussion went further than I anticipated. But having once calculated the risk of dying by car crash in the DC area vs. the risk of dying by DC-area sniper (car crashes came out way ahead), I'm all for this level of wonkiness.
I think the right comparison is per-mile fatalities, given the original post, which talks about which mode of travel to use to get to a given destination. The problem is in comparing apples to apples. I don't think translations of per-hour statistics to per-mile fatalities based on cruising speed works well, because the travel time of both planes and cars involves a lot of relatively slow movement (at stop lights, in traffic, on approach, holding patterns, etc.). That drags down the average. We'd need to know how exactly much going-nowhere or going-slowly time each form of transportation has per hour in order to make the right conversion.
This site has figures for both automobiles and airplanes in vehicle-miles:
http://hazmat.dot.gov/riskmgmt/riskcompare.htm
That reports 1.3/100m vehicle-miles for automobiles and 1.9/100m vehicle-miles for planes. But the problem there is that planes average a lot more passengers; what we really want is deaths per passenger-mile. (See note c.) Also, the plane figures lump in large commercial carriers and small commercial carriers, and exclude amateurs entirely.
Using Orin's figures, small planes experience 2.305 fatalities per 100,000 flight-hours. We still don't know what the per-passenger rate is. A Cessna R172 carries only four people, like a car, but it's expensive to fly so perhaps the average Cessna flight has more people on it. But let's assume not. Cruising speed is 177kph or 110mph.
http://www.airliners.net/aircraft-data/stats.main?id=141
Just to pick a figure out of thin air, assume that a Cessna's average speed over a flight is 75% of the cruising speed. (The Cessna R172 has a maximum range of 662 miles, so there's not going to be more than a few hours of uninterrupted level flight.) That translates to 2.305 deaths/8,250,000 miles, or 27.93 deaths/100m miles, which is 20 times the rate for automobiles.
Naturally, as Orly says, you can reduce that risk by choosing your pilot carefully, perhaps even moreso than for cars, where there's probably a relatively greater risk of some other idiot plowing into you.
Posted by: Bruce Boyden | Apr 20, 2008 12:42:06 AM
Orly,
Yes, if you're using the plane to get from A to B and not pleasure flying, but I thought the speed of a typical Cessna was only about 3-4x that of a typical car. In that case, the fatality rate per mile of flying is still about 100x that of driving, right?
Posted by: Orin Kerr | Apr 19, 2008 5:48:44 PM
I looked into this a bit more and found some useful literature. it seems that Orin's stats comparing time on the road versus time on the air is flawed. These articles suggest that if you want to get from point A to point B, you should calculate the per mile risk, and cars are much worse, because you would spend three-ten times more hours in a car getting from point A to point B.
see:
http://www.observer.com/node/40329
http://www.newton.dep.anl.gov/askasci/gen99/gen99845.htm
http://www.soyouwanna.com/site/toptens/accidents/accidentsfull.html
Posted by: Orly Lobel | Apr 19, 2008 2:57:07 PM
I strongly recommend this article about flying...
http://www.theatlantic.com/unbound/langew/turn.htm
Posted by: keitht | Apr 19, 2008 11:14:14 AM
One of my friends works for the Navy, and did some training with their pilots for fun. Apparently, most pilots rely on visual cues rather than their instruments, so yes, pilot error, inexperience, and overconfidence contribute greatly to the number of amateur accidents. Apparently, that's what happened to JFK.
The best test is when they put the hood on you--a helmet that blocks your peripheral vision. Flying with only instruments takes lots of practice, and many pilots stop short of what they need to be rated safe.
I'm with Orin. I am all for personal liberty, but when my boyfriend expressed interest in taking flight lessons, I was silently freaking out in my head and wondering how I could dissuade him without actually saying something like "you can't do that. you could die, and I would be mad."
Posted by: Belle Lettre | Apr 18, 2008 5:44:27 PM
thanks Orin, for not sharing these stats with me yesterday morning...
One of the questions is whether there are numbers that distinguish between human error and technical problems as the cause. I think alot of it is attributed to human error with some pilots being unexperienced or over-confident.
Posted by: Orly Lobel | Apr 18, 2008 4:25:08 PM
There are lots of stats on this on the web. Let's see where they go. First, some interesting stats from http://www.planecrashinfo.com/cause.htm
Type of Flight Fatalities per 100,000 flight hours
Airliner (Scheduled and nonscheduled Part 121) 0.089
Commuter Airline (Scheduled Part 135) 0.670
Commuter Plane (Nonscheduled Part 135 - Air taxi on demand) 1.230
General Aviation (Private Part 91) 2.305
So it looks like the fatality rate per flight hour for general small planes about about 250 times that of commercial flights.
In contrast, there are 1.4 fatalities per 100 million miles of driving (both for cars and motorcycles, I think), according to this site:http://www-fars.nhtsa.dot.gov/Main/index.aspx
If you figure that the average driving speed is around 50 mph, that means that driving yields about 1.4 fatalities per 2 million hours, or about .7 per million hours or .07 per hundred thousand hours.
If my numbers are right, it means that flying in a small plane has a fatality rate per hour of travel somewhere around 350 times that of driving (including on a motorcycle, I think.)
Posted by: Orin Kerr | Apr 18, 2008 3:11:21 PM
I know enough people that were killed in small planes to be skeptical that it is safer to fly in a small plane than to drive. Maybe that is just random bad luck or poor skill, but knowing of two accidents in particular that each killed every one in the plane makes me a little cautious.
Posted by: Orin Kerr | Apr 18, 2008 2:59:42 PM
I believe that the statistics hold even with the small planes, but there is greater variance per pilot -- you need a number of hours of practice to reach the "safety zone" of good pilots.
Posted by: Orly Lobel | Apr 18, 2008 2:43:27 PM
I'm curious: Flying in general is safer, but is flying in a small plane with an amateur pilot still safer, mile for mile, than driving? I don't even have a good guess on that one.
Posted by: Bruce Boyden | Apr 18, 2008 2:40:54 PM
The comments to this entry are closed.