Just doln’t tell anybody about the plane crashes. The news media never mentions that in the USA, there are more than two fatalities per day in railway accidents. More than 700 railroad deaths a year and the number has been steadily rising in recent years.
There is a real point to be made. People are notoriously bad at judging risk. When a commercial airplane crashes, there are several hundred deaths, big newspaper headlines and people start to worry. Car crashes happen every day, every hour, maybe every minute, but they don’t make headlines and people don’t worry. Cars have been made much safer in my lifetime and many of the safety features (e.g. seat belts) were strongly resisted both by the manufacturers and the general public. Not so with airplanes.
But I think the main relevant statistic is deaths among rail passengers, which is a far lower number, only around 5% of total deaths according to the paper linked below. The majority of deaths associated with rail transport are from trespassers and suicides, and I don’t think those should be factored into considerations of the safety of rail travel. I’m not quite sure how to think about some other associated types of death such collisions at road crossings, that would not occur if trains didn’t exist - I have little doubt that the majority of accidents at crossings are caused by the actions of the car drivers.
https://www.transit.dot.gov/sites/fta.dot.gov/files/docs/Rail%20Safety%20Statistics%20Report.pdf
There are a number of things different between planes and automobiles. (also, notable, if these things were required in automobiles a *lot *less people would die)
-
The FAA requires that maintenance schedules be adhered to. That means at 3001 miles in your car, by these rules, you could only finish a trip and then immediately take it to a mechanic, if the manufacturer says to change the oil every 3k miles.
-
The FAA requires that the operators of the vehicle be trained, have verified hours operating, be trained on the specific model of the vehicle, be given an annual medical screening, and all “close calls” and other incidents be reported. Most car drivers would fail and have their license to drive stripped if the government were as strict.
-
When a crash happens, the FAA requires that the cause be found and corrective action taken.
-
There is a dollar amount - if it costs less than about 10 million dollars to save someone’s life for a new safety measure, it will become required.
At a bare minimum if these requirements applied to cars, all cars that don’t meet the maximum crash safety standards would be no longer road legal. SUVs, illegal. Probably 4 point restraints, built in fire suppression, and a built in glass hammer so you can escape from a flooding car would be required. Your uber driver might have to play a video explaining how to escape from a burning wreck or flooding car each trip. (since you wouldn’t be able to drive yourself)
So you can obviously flip things the other way. You could let airliners just declare anyone is a pilot they want to. “De-regulate” the industry. Let them run their planes as long as they want and only do maintenance when the airline feels like it. Let them carry liability coverage that is only a fraction of the damage done in a crash, and let the airliners split up into a separate independent subsidiary per aircraft, with no right to sue the parent company if you are injured from negligence.
Even then, planes would probably still be safer, since the airline wouldn’t want to lose an expensive aircraft, but the difference would be less.
I build airplanes for a living and have been doing it for over 30 years. All I have to do is look at some the people I work with and amount of rework I do every day so I can complete my work. Based on that, anything that reduced the safety of current airplanes would kill the aviation industry.
That’s actually not true.
https://www.ntsb.gov/investigations/data/Pages/AviationDataStats2015.aspx
For all US Civil Aviation in 2015 there were a total of 1282 accidents, of which 238 had fatalities for a total of 406 deaths that year.
I used to work in airplane certification. A commercial airliner is designed so that the probability of any combination of failures which causes a catastrophic event (hull loss and major number of fatalities) per flight hour is 1 / 10**9. A great deal of statistical and probability analysis is done to ensure that. I seriously doubt that any such analysis is done for automobiles–but I strongly believe that a similar criteria should be applied to self-driving cars.
Plane travel probably didn’t have to become as safe as it has. The significant reduction in fatality rate since say 50 yrs ago probably isn’t that big a factor in many more people flying, compared to changes in economics (lower inflation adjusted fares albeit more complaining passengers, and rising incomes especially considering the whole world not just US). It was safe enough decades ago to overcome the ‘feel’ factor of less safety (what a lot of people’s gut tells them on an airline flight that their mind has to overcome) and the lack of control factor, for most people.
But if it suddenly became only as very safe as 50 yrs ago, so a much larger absolute number of plane crashes than now or historically (since there’s lots more airliners now) that would be a huge PR problem. And if it now became as unsafe as driving in general it would be non-viable. Another footnote on plane travel is that accident stats don’t include terrorism and war. Those are also very small risks historically, but terrorists are a sentient force trying to outmaneuver safety measures and could suddenly become much better at it. That’s an ‘unknown unknown’ which doesn’t apply to ‘regular’ flight safety. I know terrorism risk discussion can become political very fast though.
But while airline travel is much safer than car travel it’s not at all irrational to reject general car safety stats as applying to an individual driving in a particular car, manner, location, time etc. IIHS data on driver death rate by car model has several luxury sedan and SUV models which have suffered no driver fatalities in multi-yr periods of driving in the US and a number of others in low single digits per million vehicle-years v. an average of 2014 model cars of 30, and 2002 model cars of 87. Further consider driver and condition differences within models: it’s a very skewed distribution and the illogical thing would be to assume the average stats must apply to you. They could greatly over or understate your risk.
If it made air travel a lot more cheap and convenient would that be a price worth paying?
Fair point. On the other hand, though, rail deaths would, if publicized as raw statistics without the caveats, the public would be demanding safer rails. Imagine if Fox and Friends decided go on an anti-rail campaign and raw statistics went viral, Amtrak would be forced to shut down. Similarly, air safety was triggered by the high visibility of plane crashes.
I acknowledge the point though that car travel is much safer if you’re doing it right, which is why I’ve always wondered why cities and states see traffic enforcement primarily as a revenue measure rather than making a real effort to make safety a priority. It also seems that making driving nearly universal is a higher priority than making sure the people that do drive do it well. We really should cull the herd, IMO. We’d probably reduce traffic deaths by a huge number if we just took away the licenses of the worst 10% of drivers.
I was thinking of airliners since that’s what the OP was talking about.
It was even publicized(and Trump tried to take credit for) that there have been no airline fatalities in several years.
Amtrak probably should be shut down except for a few well traveled routes like the Acela corridor. You might have just given Fox and Friends a good idea to get that done.
But seriously, bus service is getting better due to competition with Greyhound and Amtrak is slower and more expensive than flying for most routes. Those who can’t afford to fly take the bus. those who can afford to fly, fly. The train is for people who like trains or who live near a route where it’s actually superior to bussing or flying. I take the train to Orlando from Fort Lauderdale, but anywhere outside of Florida it’s not viable at all.
Well, air safety is heavily managed by the federal government through the FAA, NTSB, industrial regulations, and who knows what else, so it’s quite fair for the head of the federal government to take credit for it.
Of course, he didn’t do anything and the trend predates him by a LOT, but just not breaking stuff is an accomplishment of sorts. I’d rather see a President take credit for something the government does control(air safety) rather than something it doesn’t(gas prices, the economy).
I’m not going to try to dig out the stats, but Id very surprised if more than a tiny number of auto accidents are caused by vehicle hardware failure. Nearly all are human error. Even cars that have deplorable maintenance and outdated part are still rarely involved in crashes if the driver is reasonable competent, attentive and sober. Similarity, I think many if not most air crashes can be traced to human error. The fact that there are far fewer air crashes is attributable to being much more careful about whom we let fly them.
I agree, although there is one big difference. When a car fails, more often than not it just won’t go, or you come to a stop on the roadway. When a plane fails it would sure be nice if it just stayed up there until someone could come up and fix it.
Could you explain what you’re advocating or decrying relative to what here? Are you treating manually-driven car safety as a lost cause, and treating the advent of self-driving cars is an opportunity to bring car safety standards up to aviation safety standards? Or do you think a much higher standard for self-driving cars over manually driven cars is actually appropriate for some reason?
We probably wouldn’t, but mainly because we’ve structured our society so that those 10% will likely continue to drive even without their licenses (since they’ll have to in order to do almost anything.) In the medium term, technology may solve this for us with self-driving, but until that’s really in place, I wouldn’t expect stricter licensing to have much impact.
A self-driving car is going to have to work absolutely perfectly; the only way to ensure that is to have a sound regulatory framework to ensure that they’re built to a high standard of safety and reliability. The way that aircraft certification works (setting a goal for catastrophic failure rates) seems to me to be a good approach to use. Unfortunately, it doesn’t look to me as though the current regulations are nearly stringent enough, although I’m willing to be corrected.
It’s an interesting question, and I don’t think the answer is just to scale up the number of fatalities per mile driven to miles flown.
As some have pointed out, you can take certain steps to vastly increase your safety per miles driven over the national average, and in our hypothetical “loose rules” future I think we can, as a starting point, use the commercial bus fatality rate:
Bus drivers are subject to regulations, but they’re nowhere near the level that apply to pilots, and should therefore cost significantly less to meet.
From here it’s all speculation – on the one hand, there’s less stuff to hit in the sky, but on the other hand, what would the scene look like if planes had mechanical failures as often as commercial buses? Not every mechanical failure would result in a crash, as commercial airliners have a pretty good glide ratio, but certainly some would.
If we just take that bus fatality rate without any further speculation, would travelers accept a 65% increase in the risk of death for, say, half price airfare? I certainly would.
eta: I imagine most people wouldn’t be too keen on that trade-off, but I also imagine that most people wouldn’t be willing to pay twice as much per ticket for a 65% reduction in risk.