Hacker News Reader

Comment by Robin_Message | Hacker News Reader

I make a similar argument for train travel: since the death rate is ~10% that of other forms of transport, if trains could be made cheaper by compromising safety to say 50% of other modes, that would be a net positive as cheaper trains would move people off other, still more dangerous forms of transport.

View on HN

Safety Assurance is a big part of rail transit design but I wouldn't say it is a significant cost. Building and operating rail wouldn't cost half as much if we stripped some of the safety assurance process out of it. Maybe a couple of percentage points of cost for orders of magnitude more fatal accidents. The systems would mostly all still be there, unless you wanted to do something as drastic as removing all interlocking/train protection and rely entirely on drivers and timed signals - basically treat it like a road network - but that would have huge operational impacts - train headways have to be much larger if you can't know what sections have trains in them - as well as being extremely unsafe.

Frankly, these precautions exist for a reason, and it is because many many people died in the early days of rail. Many people who work in rail, particularly in ops/sigs/assurance are proud of how safe rail transport now is.

(I work in Engineering Assurance on rail infrastructure projects).

Fair enough, thanks for sharing. It's not really serious proposal but interesting to learn it wouldn't work. I guess the ongoing cost of rails and rolling stock are killer.

Much of the infrastructure cost is boring stuff like earthworks and retaining walls and moving existing services that are in the way. The shift to in-cab signalling saves/will save a bunch of money on lineside equipment maintenance (because you don't need it).

I was thinking about your comment and one way to save considerable infrastructure costs is more level crossing and less overpasses/bridges, but train vs pedestrian or train vs car are common and messy. Plus level crossings wreck traffic flow on the surrounding road network.

And you could certainly make the same argument about airliners, which have a death rate <10% that of trains (and <1% that of cars).

But if you try to make that argument on HN, a bunch of people will yell at you about how Boeing and the FAA are evil for putting cost savings over safety.

> And you could certainly make the same argument about airliners, which have a death rate <10% that of trains (and <1% that of cars).

You can, but it's not a fair comparison. 99.9% of deaths of railroad transport is people killed while crossing rails - e.g. pedestrians.

If you consider rail passengers only, it is far lower than airlines and cars, something on the scale of 2-3 people/year.

Airplanes are cheating in the sense that there are no pedestrians in the air to collide with.

I mean, not _zero_[0]... but close enough, really.

0: https://www.aopa.org/training-and-safety/air-safety-institut...

Wow.. And if you asked me to predict the outcome of that incident, I'd have said the fatalities would have been the other way around... go figure!

Haha, ok, I stand corrected.

Interesting stat.

The Safety cost and even the hull cost is simply not a major contributor to the cost of your ticket. It’s mostly fuel gate fees maintenance and overhead

Where do you think the maintenance fees come from? That's in part a safety cost.

Fuel as well!

"this isn't enough fuel to get more than 80% of the way there!"

"you don't need fuel for the descent though"

Keep in mind, if you compromise maintenance for costs, you'll make up the costs in crew wages as they obviously have 1-2 orders of magnitude more risk exposure than the average passenger and won't be willing to fly for anything less than massive hazard pay.

Scheduled maintenance is done with semi slave labor in the 3rd world. Routine is more expense local labor but the routine stuff can’t be deferred anyway

Airliners are much safer than other modes of transportation per mile traveled. The difference vanishes (mostly, or reverses) when interpreted per mile on the vehicle or per trip.

If airliners were only as safe as walking, nobody would use them, because you'd expect to have at least a minor accident (like a sprained ankle) when walking 2000 miles. A similar argument can be made for cars and trains (just a magnitude less distance per trip and consequently a smaller chance of injury).

It's a novel idea in terms of emissions avoidance, though: couple the externalities to personal risk.

People who currently are forced to commute daily with cars could be much safer doing so on a more robust train system. You wouldn’t get the same benefit from cheaper flights.

Did we both interpret the MCAS incident with Boeing differently? Because I saw this as regulatory capture putting it's metaphorical testicles where it felt comfortable.

Reply on news.ycombinator.com