Jump to content


Photo
- - - - -

does Society expect engineers to stop people dying?


  • Please log in to reply
189 replies to this topic

#1 mariner

mariner
  • Member

  • 2,325 posts
  • Joined: January 07

Posted 01 April 2015 - 15:06

and to finish , rather than just reduce the chance of an accident.

 

A somewhat large and off topic question in way but the several things have made me ponder on this - as a non engineer.

 

Firstly all the talk of driverless cars.They will prevent roads deaths has morphed intoa  statement by Ed Musk of Tesla " human drivers may have to be banned"- i.e engineers can stop accidents so the reponsiblity lies with them not individuals

 

Some peple are demanding pilotless planes immediately to stop pilot suicides costing lives. Even in this forum the question is " what engineering solution to the Nurburgring spectator fatality".

 

Don't get me wrong , Im not condemning such demands as stupid, silly or unaffordable but I was just pondering what the people here think that society now demands of engineering - including software design?

 

My sense is two things - Western Societies tolerance of risk has reduced substantailly over the last 30 years. Motor racing is no exception. It has ben made so much safer , thanks to engineers and sponsor/Bernie money. However if by some freak we had a year of F1 deaths like the 1970's would it now be banned or reduced drastically in speed/content?

 

Secondly safety has always demanded trade off's. As I understand it that is exactly how the FAA/CAA go about their work which seems to be less impacted by politics than say car  design.

 

As I say I am not an engineer but do recall one example 40 years ago of sitting in on a value of human life discussion. I worked in an oil company abut to commit to the production platforms in the UK North Sea. The number of platforms is determined by drilling technology but the size so cost of each one is basically driven by the weather.

 

Platform size equals cost. Size is deck area times height above mean sea level. If you make the deck small you have to re supply more and bad weather stops supply ship unloading hence you shut down operations. Safety dictates the deck is above the highest likely wave.

 

So we called in a weather expert who said basically " do you want to design for a 20 year, 50 year or 100 year wave?"

 

A one hundred year North Sea wave is 90ft plus or as he put it the height of a ten storey building. Having finally accepted that size was possible the discussion was " well, no not one hundred wave so how about 50 year wave as that is 2.5X the expected platform life.

 

 The discussion went on a long time as everybody did realise lives were involved. My sense is ,and I may be wrong, that such a conversation would not be allowed , or recorded today lest it was used to prove " profit before lives etc"

 

What to people think? Are such discusions stil allowed/commomplace in design?

 

I  suppose the ideal solution would be for goverments to state clearly their acecptable money value of a citizens life and let the enginers prove they designed compotently to that level.

 

 



Advertisement

#2 Bloggsworth

Bloggsworth
  • Member

  • 9,389 posts
  • Joined: April 07

Posted 01 April 2015 - 18:37

"Human drivers may have to be banned" is more about mixing the two than anything else. Autonomous cars will trundle around at 30MPH maximum, they will have accidents unless all roadsides everywhere are fenced off so that nothing can stray onto the roads - Meanwhile, in America, it is expected that this year, more people will die at the wrong end of a firearm than in road accidents...



#3 Greg Locock

Greg Locock
  • Member

  • 6,349 posts
  • Joined: March 03

Posted 01 April 2015 - 23:26

Both the FAA and NHTSA use a cost benefit analysis to determine whether a safety improvement is cost effective, although this is more politicised than it used to be, for instance the rear view cameras for SUVs failed the cost effectiveness test by a factor of 10 but still went ahead, from memory.Given the tendency of American courts to impose huge multiples of actual damages as punitive measures, this may be a wise precaution, albeit one the aircraft industry has managed to avoid.

 

Having said that the costs quoted by OEMs for adding the cameras also looked severely inflated to me, so maybe it all balances out. (I suggest the actual variable cost of adding a reversing camera to a new platform is 20 dollars, not 200, since the screen is already there, and it would be available as an option anyway)

 

Here's a discussion with some background

 

http://www.foreffect...lic-protections


Edited by Greg Locock, 01 April 2015 - 23:43.


#4 imaginesix

imaginesix
  • Member

  • 7,525 posts
  • Joined: March 01

Posted 01 April 2015 - 23:30

Nothing is certain except taxes.



#5 gruntguru

gruntguru
  • Member

  • 7,635 posts
  • Joined: January 09

Posted 02 April 2015 - 00:18

"Human drivers may have to be banned" is more about mixing the two than anything else. Autonomous cars will trundle around at 30MPH maximum, they will have accidents unless all roadsides everywhere are fenced off so that nothing can stray onto the roads - Meanwhile, in America, it is expected that this year, more people will die at the wrong end of a firearm than in road accidents...

 

""Human drivers may have to be banned" is more about mixing the two than anything else."

No it is about AV's being safer than human drivers. Google has plenty of data from millions of km of AV's interacting with uman piloted traffic to support that assertion.

 

"Autonomous cars will trundle around at 30MPH maximum,"

Source? Many experts believe AV's will safely operate at higher speeds and speed limits will eventually be raised as a result.

 

"they will have accidents unless all roadsides everywhere are fenced off so that nothing can stray onto the roads"

Source?

 

You clearly haven't researched the subject. Here is a link to get you started. https://www.google.c...s vehicle study


Edited by gruntguru, 02 April 2015 - 23:06.


#6 gruntguru

gruntguru
  • Member

  • 7,635 posts
  • Joined: January 09

Posted 02 April 2015 - 06:03

Mariner. Perhaps society (engineers included) is hoping engineers can continue to improve the safety of the death machines they designed in the past?



#7 gruntguru

gruntguru
  • Member

  • 7,635 posts
  • Joined: January 09

Posted 02 April 2015 - 06:17

Both the FAA and NHTSA use a cost benefit analysis to determine whether a safety improvement is cost effective, although this is more politicised than it used to be, for instance the rear view cameras for SUVs failed the cost effectiveness test by a factor of 10 but still went ahead, from memory.Given the tendency of American courts to impose huge multiples of actual damages as punitive measures, this may be a wise precaution, albeit one the aircraft industry has managed to avoid.

 

Having said that the costs quoted by OEMs for adding the cameras also looked severely inflated to me, so maybe it all balances out. (I suggest the actual variable cost of adding a reversing camera to a new platform is 20 dollars, not 200, since the screen is already there, and it would be available as an option anyway)

 

Here's a discussion with some background

 

http://www.foreffect...lic-protections

 

Interesting link. I gathered they only factored the reduction in fatalities (at $6.1 million per human life!! (What if someone reversed over Bill Gates?)) in their analysis. If the cost of injuries and damage avoided had been included, the numbers may have looked even better.



#8 munks

munks
  • Member

  • 428 posts
  • Joined: January 03

Posted 02 April 2015 - 14:40

IMHO, the rearview cameras are stupid, compared to the sensor/beeper that my wife's previous SUV had. With the previous SUV, I could look back and sideways like normal AND have the extra protection of the sensor/beeper for small objects (children, cats, etc.). Now my choice is to watch a tiny screen with no peripheral vision (but ability to see small objects right behind the SUV), or look back and sideways with no view whatsoever of small objects.

 

In any case, humans aren't worth $6 million. Or if they are, I'm curious what the price of other animals are.



#9 desmo

desmo
  • Tech Forum Host

  • 29,343 posts
  • Joined: January 00

Posted 02 April 2015 - 17:40

I read somewhere recently that the US government had calculated a figure of six point something million USD for human life.  As crass and philistine as it seems, you do need a number to plug in to run cost/benefit analyses in situations involving potential mortalities.



#10 Bloggsworth

Bloggsworth
  • Member

  • 9,389 posts
  • Joined: April 07

Posted 02 April 2015 - 19:46

Of course, it may be that autonomous firearms are the answer to gun related deaths in America...



#11 Fat Boy

Fat Boy
  • Member

  • 2,594 posts
  • Joined: January 04

Posted 02 April 2015 - 20:05

"Human drivers may have to be banned" is more about mixing the two than anything else. Autonomous cars will trundle around at 30MPH maximum, they will have accidents unless all roadsides everywhere are fenced off so that nothing can stray onto the roads - Meanwhile, in America, it is expected that this year, more people will die at the wrong end of a firearm than in road accidents...

 

Let's be clear on that, gun deaths are catching up to auto deaths. This may be the year that they surpass auto deaths, but it may not. The vast majority of gun deaths are by suicide.



#12 Greg Locock

Greg Locock
  • Member

  • 6,349 posts
  • Joined: March 03

Posted 02 April 2015 - 22:23

The article explains that the 6.1 million is not the exact cost of any one death, it is the marginal incremental (or whatever) cost. So society would pay 6.10 cents to avoid a 1 in 100 million chance of killing someone. 20 years ago the figure was 4 million dollars per life.



#13 Canuck

Canuck
  • Member

  • 2,384 posts
  • Joined: March 05

Posted 02 April 2015 - 22:57

Before you look at the societal expectations of engineers and harm reduction, I think you need to look at society's position on death. We could eliminate auto deaths instantly by simply banning them, full stop. Of course that's not reasonable or plausible - even less so than a self-driving car - but it would be 100% effective.

As an individual, I'm not keen on the idea and am willing to trade my safety in the (somewhat abstract) knowledge that by virtue of being in a mass of 4000ish pounds at speed, I am subject to a higher likelihood of both causing someone's death or being killed, both as a direct result of the use a vehicle. I suspect most people are willing to make that same trade-off via whatever rationale they come up with.

And it's interesting because I don't make those exceptions for much else around me. If a moment of inattention on the part of one of my offspring resulted in them being crushed to death by the piano while they were playing it, I wouldn't have it. Unacceptable risk. Or if the toaster threatened to electrocute you if you accidentally applied anything other than a purely vertical force to the button. So clearly there's a subconscious calculation and trade-off I'm making.

In quantifying the "danger level" of a thing, we ask what is the frequency of the activity or use, what is the likelihood of an accident and what is the likely result in terms of harm (minor, bodily damage, death). There's never a cost factor in that equation, but I've not yet run into a solution that costs so much as to be denied, though I'm certain there are plenty of scenarios where they exist.

Recently, a crew member on a film was killed when she was unable to outrun and ultimately run over by a train on the bridge the producer and director chose to film on without permission or notification. Pointless and without purpose.

Conversely, auto racing in its various guises often results in the severe injury or death of the participants, which in my own world I rationalize as the outcome of their personal choice (to race), one made willingly. Also pointless, but with a much greater degree of informed choice.

I get upset with deaths when I feel they are the result of neglect, hubris (particularly on the part of a company) or because a corporation somewhere (GM) doesn't want to risk an upset in the stock price, so they hide / ignore / otherwise continue to leave people in uninformed risk.

I expect engineers to design a product that won't explicitly try to kill me, and to inform me of remainong risk involved. Informed choice.

#14 gruntguru

gruntguru
  • Member

  • 7,635 posts
  • Joined: January 09

Posted 02 April 2015 - 23:26

Conversely, auto racing in its various guises often results in the severe injury or death of the participants, which in my own world I rationalize as the outcome of their personal choice (to race), one made willingly. Also pointless, but with a much greater degree of informed choice.
 

I expect engineers to design a product that won't explicitly try to kill me, and to inform me of remainong risk involved. Informed choice.

Rather than being "pointless" your motor racing example is in the same category as road driven automobiles. Participants are trading an acknowledged risk against a clear benefit.

 

I almost said "known risk" rather than "acknowledged risk" but I am sure the majority of road drivers greatly underestimate the risk level. We all set out on each journey with a high level of confidence that we will not be involved in a serious accident. More significantly, the fact that I am operating the car makes me even more confident that I will be able to avoid danger - I am in control of my own destiny!  It is loss of this level of control that creates fear and suspicion of AV technology. I believe that even the naysayers will develop a high level of confidence in AVs once they gain first hand experience.



#15 Bloggsworth

Bloggsworth
  • Member

  • 9,389 posts
  • Joined: April 07

Posted 03 April 2015 - 08:01

Let's be clear on that, gun deaths are catching up to auto deaths. This may be the year that they surpass auto deaths, but it may not. The vast majority of gun deaths are by suicide.

 

Ah, so that's all right then, as long as it is only the vast majority... Though I wouldn't have said that 60% constituted a "Vast majority."


Edited by Bloggsworth, 03 April 2015 - 08:03.


#16 Fat Boy

Fat Boy
  • Member

  • 2,594 posts
  • Joined: January 04

Posted 03 April 2015 - 16:16

Ah, so that's all right then, as long as it is only the vast majority... Though I wouldn't have said that 60% constituted a "Vast majority."

 

OK, you obviously have a soapbox you want to get on. Please do and get it over with. Take 5-10 pages and just get it all out in one big self-indulgent cum-shot.

 

 

 

For the record, I don't own gun and I'm not in the NRA. I don't even have a dog in the fight. I've just had an ass-full of people that want to say what I can and can't do.



#17 Canuck

Canuck
  • Member

  • 2,384 posts
  • Joined: March 05

Posted 03 April 2015 - 16:20

Rather than being "pointless" your motor racing example is in the same category as road driven automobiles. Participants are trading an acknowledged risk against a clear benefit.

 

 

Not the same category at all.  Racing is sport, entertainment, fun - all good things, but in the grand scheme of it all, racing doesn't appear to be necessary.  And in the strictest sense, given that they (with a handful of exceptions) start and end in the same location, they haven't even done any work  ;) , whereas I would guess the majority of road-driven automobiles are used for transportation - getting people and or items from point A to point B (cruise nights and aimless teenage driving notwithstanding).



#18 Canuck

Canuck
  • Member

  • 2,384 posts
  • Joined: March 05

Posted 03 April 2015 - 16:22

Though I wouldn't have said that 60% constituted a "Vast majority."

Unless the number was supporting your argument.



#19 Fat Boy

Fat Boy
  • Member

  • 2,594 posts
  • Joined: January 04

Posted 03 April 2015 - 18:37

I didn't do any exhaustive research, I was going off memory, I think the last numbers I remember was 3 or 4 to 1 (suicide/homicide). If it's 60%, then fine, that's the number. Drop the word 'vast' if you don't like it. As I think about it, I think that 3:1 or 4:1 ratio was suicide+accidental/homicide. Hell, I don't know.

 

If you want, you could point out that the vast (I think this works here?) majority of car deaths _are_ accidental/suicide. The number of car homicides has to be pretty small (again, I've got no idea what the statistics are, just an important counterpoint).



Advertisement

#20 Ross Stonefeld

Ross Stonefeld
  • Member

  • 70,106 posts
  • Joined: August 99

Posted 03 April 2015 - 19:51

I'd hope the numbers on vehicular suicide are astonishgly low...



#21 GreenMachine

GreenMachine
  • Member

  • 2,629 posts
  • Joined: March 04

Posted 04 April 2015 - 00:07

I'd hope the numbers on vehicular suicide are astonishgly low...

There may be research, but I don't know of it. In a previous incarnation I was a road funder, and the talk when this subject came up was that a 'large' proportion of single vehicle accidents were considered suicide. It would not be limited to single vehicle accidents though, I have in mind one case of a head-on, where the car driver looked at the oncoming truck, according to the truck driver he looked right into the truckie's eyes, and at the last moment steered across into the truck.

Unless there is a note somewhere, it can only be speculation, based in the lack of an apparent or probable cause for the accident, or maybe some personal details come out like divorce, debt etc.

 

There was another case a while back, involving a Ferrari, a country highway, and a large tree.  The driver's business was spectacularly collapsing, I think drugs may have been involved, and IIRC, the driver was also supposed to have been ex-CIA.


Edited by GreenMachine, 04 April 2015 - 00:12.


#22 Fat Boy

Fat Boy
  • Member

  • 2,594 posts
  • Joined: January 04

Posted 04 April 2015 - 02:23

There may be research, but I don't know of it. In a previous incarnation I was a road funder, and the talk when this subject came up was that a 'large' proportion of single vehicle accidents were considered suicide.

 

I remember reading something similar. I don't know if the proportion could be considered 'vast', but 'large' is probably good enough.



#23 mariner

mariner
  • Member

  • 2,325 posts
  • Joined: January 07

Posted 04 April 2015 - 10:05

One thing that always puzzles me about perceptions of road safety is the demand for "safer cars" and the risks accepted by drivers every day on , say, an expressway/motorway.

 

Nobody leaves enough distance in front at 70- 80 mph to stop safely if the vehicle in front literally stopped instantly. I doubt that AV's wil deviate that much from this habit as , if they did, road capacity would be drastically cut. Indeed the idea on AV's is to run them even closer than human drivers.

 

Contrast this with raillways. They have simple golden ruie - the train  mut be able to stop before the next red signal. The rail engineers rule is " your right to speed is your abilty to stop". That means stop to a dead halt before the next signal , not to assume the train ahead wil never stop instantly.

 

It's this contridiction that puzzles me on spending money on auto safety features. Why spend $200 per car  on a back up camera for 95 deaths in the USA when 250 million US drivers will drive , by logical safety analysis , far too close 365 days a year on highways. It make absolutely no logical or economic sense to spend money on things like reversing cameras without adressing tail gating. Of course making drivers, or AV's , keep  a gap based on the railway rules would cause huge delays and extra cost.

 

Think of it this way- assume  a person's time is worth $20/hour , thats 33 cents per minute. So, if people are wiling to pay 6.1 cents to avoid a 1 in 100 million chance of killing someone they should be be wiling to go 18% slower ( 6.1/33.3) on the motorway - i.e everybody would drive every day at 57 mph not 70 mph.

 

So if Congress accepts $6.1m as the price of human life it should, logically, re - impose the  55mph natioanl sped limit,, have a driving ban is the penalty for exceeding it and make 55 mph the legal controlled maximum of future AV's.


Edited by mariner, 04 April 2015 - 10:08.


#24 Canuck

Canuck
  • Member

  • 2,384 posts
  • Joined: March 05

Posted 04 April 2015 - 16:45

That presumes that speed alone is the single biggest drive of accidents.  It may well be - I don't know.  Has there been an increase in vehicle accidents per mile (or whatever the measure is) since the repeal of the 55 limit?  Has there been an increase is motor vehicle deaths specifically as a result?  Admittedly it's a pretty difficult puzzle to tease apart as vehicle safety increases may well be off-setting the increase in previously-fatal accident scenarios.

 

What about a government-enforced bumper buzzer that, much like the unbuckled seat belt annoyatron, sounds an annoying tone and lights up the dash idiot light at any point that the vehicle's speed and proximity to the vehicle in front puts it in an unsafe condition?  Much like the buzzer encourages all but the most determined seat belt non-users, safe-following-distance would soon become a learned behaviour.  The technology to implement it is likely cheaper than the $20 camera - a couple of proximity sensors that are already being installed in the rear coupled now also installed in the front, and a $2.00 circuit for the calculation and warning system.   Yes yes, lawyers and brake condition and wet roads and what have you make it more complicated obviously, but the general premise is there and the obstacles hardly insurmountable. Think of the children!



#25 Robin Fairservice

Robin Fairservice
  • Member

  • 599 posts
  • Joined: March 07

Posted 04 April 2015 - 20:50

I used to work for the British Columbia Ministry of Environment.  At a public hearing about air emissions from an industrial plant, evidence was given that at the plant fence line the chances of the emissions of a particular chemical causing a death were one in a million.  That apparently was not good enough for the residents of an adjacent small town.  From vehicle accident statistics I pointed out that of the present population of about 11,000, 2.4 present residents would have been killed in a vehicle accident within the next 12 months.  Nobody as much as batted an eyelid!  It is amazing what is acceptable and what is not. 



#26 Greg Locock

Greg Locock
  • Member

  • 6,349 posts
  • Joined: March 03

Posted 05 April 2015 - 01:57

"Nobody leaves enough distance in front at 70- 80 mph to stop safely if the vehicle in front literally stopped instantly"

 

The physical reason for that is that in the absence of enormous blocks of glass magically appearing on motorways it is impossible for the car in front to stop instantly without you being able to see what it has hit. Now, if there is a blind curve it may be that you can't stop before you hit a stationary vehicle, but the friction circle says if you can go round a bend then you can stop in 30 degrees of that bend (roughly)-and in practice people drive at around 0.2g on curves, and can brake at 0.8g, so blind curves aren't really a big deal at high speed. The correct interval is your reaction time*v+an allowance for the difference in braking of the two vehicles.



#27 gruntguru

gruntguru
  • Member

  • 7,635 posts
  • Joined: January 09

Posted 05 April 2015 - 04:52

Added to that - with reference to AV's.

 

 - What percentage of tail-enders are due to inattention rather than tailgating? If the vehicle ahead brakes hard and you don't notice for one second, you are now travelling 30 km/hr faster than it and will continue to do so, no matter how hard you brake.

 

 - AV's (including current auto braking systems) can follow much more closely with safety because they constantly maintain the following distance.



#28 J. Edlund

J. Edlund
  • Member

  • 1,323 posts
  • Joined: September 03

Posted 05 April 2015 - 10:29

That presumes that speed alone is the single biggest drive of accidents.  It may well be - I don't know.  Has there been an increase in vehicle accidents per mile (or whatever the measure is) since the repeal of the 55 limit?  Has there been an increase is motor vehicle deaths specifically as a result?  Admittedly it's a pretty difficult puzzle to tease apart as vehicle safety increases may well be off-setting the increase in previously-fatal accident scenarios.

 

What about a government-enforced bumper buzzer that, much like the unbuckled seat belt annoyatron, sounds an annoying tone and lights up the dash idiot light at any point that the vehicle's speed and proximity to the vehicle in front puts it in an unsafe condition?  Much like the buzzer encourages all but the most determined seat belt non-users, safe-following-distance would soon become a learned behaviour.  The technology to implement it is likely cheaper than the $20 camera - a couple of proximity sensors that are already being installed in the rear coupled now also installed in the front, and a $2.00 circuit for the calculation and warning system.   Yes yes, lawyers and brake condition and wet roads and what have you make it more complicated obviously, but the general premise is there and the obstacles hardly insurmountable. Think of the children!

 

Speed is generally not regarded as a major cause of accidents, so reduced speed limits does not means there will be fewer accidents. Actually, there are examples of reduced speed limits increasing the number of accidents, probably due to drivers feeling more safe at lower speed and putting less focus on their driving. Speed is however a major factor in the results of an accident; higher speed means higher forces are involved in the accidents that do occur and the likelihood of major injury or fatality is increased. So, lower speed limits should be seen as a way to reduce injury and death in the accidents that occur rather than a way to reduce the number of accidents.

 

Added to that - with reference to AV's.

 

 - What percentage of tail-enders are due to inattention rather than tailgating? If the vehicle ahead brakes hard and you don't notice for one second, you are now travelling 30 km/hr faster than it and will continue to do so, no matter how hard you brake.

 

 - AV's (including current auto braking systems) can follow much more closely with safety because they constantly maintain the following distance.

 

Automated vehicles and vehicles with automated braking systems can follow much more closely because they can react to the car in front braking much faster than a human. The perception and reaction time for a human driver is about 2.5 seconds, the automated system react much faster.



#29 Canuck

Canuck
  • Member

  • 2,384 posts
  • Joined: March 05

Posted 05 April 2015 - 15:23

2.5 seconds? Yeesh. Are the test subjects also on quaaludes? Seems awfully slow for an attentive driver at least. Perhaps far too few are actually attentive.

#30 Robin Fairservice

Robin Fairservice
  • Member

  • 599 posts
  • Joined: March 07

Posted 05 April 2015 - 16:12

I would like to know how these automated vehicles would cope with snow and ice on highways.  Also now that spring is here in northern BC many of the road lines have vanished due to snowploughing.  May be these poor little cars will just come to a full stop until some kind soul comes along an repaints the lines for them.  Then there are the GPS systems that have out of date GPS information.  Many SatNavs can't find my house, and it is 20 years old, but not in a municipality.



#31 MatsNorway

MatsNorway
  • Member

  • 2,822 posts
  • Joined: December 09

Posted 05 April 2015 - 22:16

Its not as black and white as Elon Musk wants it to be.

 

And im seeing more and more street racing in videos from USA. Massively powerful cars. 1000hp etc. I think racing will be a thing and a risk for a long long time ahead. If its safe people will up the risk until they find it fast enough..  If a racetrack does not support that or allow it people will street race.. sadly. Its just a massive amount of bike videos too. Difference is that they run from the cops often.


Edited by MatsNorway, 05 April 2015 - 22:16.


#32 gruntguru

gruntguru
  • Member

  • 7,635 posts
  • Joined: January 09

Posted 05 April 2015 - 22:24

Six things the Google car still can't handle (inclding snow & ice)

 

http://gizmodo.com/6...-han-1628040470



#33 Greg Locock

Greg Locock
  • Member

  • 6,349 posts
  • Joined: March 03

Posted 06 April 2015 - 09:00

Amazing how the author writes off a ~90% reduction in automotive deaths per year because he thinks buses are the way to go.



#34 Fat Boy

Fat Boy
  • Member

  • 2,594 posts
  • Joined: January 04

Posted 06 April 2015 - 16:33

2.5 seconds? Yeesh. Are the test subjects also on quaaludes? Seems awfully slow for an attentive driver at least. Perhaps far too few are actually attentive.

 

This seems like someone moved the decimal point over accidentally. Most of the time on the freeway I can see when someone is doing something stupid before they actually do it and I'll position myself accordingly.



#35 munks

munks
  • Member

  • 428 posts
  • Joined: January 03

Posted 06 April 2015 - 19:20

This seems like someone moved the decimal point over accidentally. Most of the time on the freeway I can see when someone is doing something stupid before they actually do it and I'll position myself accordingly.

I heard a very similar number a couple decades ago back in college: 3 or 4 seconds, as I recall. I had a hard time believing it then. But for somebody toodling along on the freeway, daydreaming away, with no expectation that anything could possibly go wrong, it might actually be accurate. Not everybody is as alert as a racecar driver.



#36 GreenMachine

GreenMachine
  • Member

  • 2,629 posts
  • Joined: March 04

Posted 06 April 2015 - 21:18

May be the difference between an alert driver, and a driver in autopilot mode, a/c on warm, music on, daydream on high ...



#37 Lee Nicolle

Lee Nicolle
  • Member

  • 11,036 posts
  • Joined: July 08

Posted 07 April 2015 - 05:27

Self driving cars are safe!! Hardly. Effectively self driving aeroplanes are supposed to be 110% safe. And how many hit hills, other aeroplanes. Quite a few. And that is with pilots to ensure a back up. And this is in the uncluttered [in comparison skies] And they have not perfected self landing planes yet either.

So expecting safe self driving cars,,,, dream on! Though far too many humans make mistakes.

 

As for the myriad of nanny and trendy devices in cars. A car with 6 air bags is safer than four!! Unlikely. Six airbags is a selling ploy. I would sooner have a well made car with none. Bloody things go off occasionally on a shopping centre hump, and do not go off in a fairly extensive crash. Or as I saw recently a front end crash [left front] where the l/f airbag did not go off, but the side airbags on the opposite side did!.

 

Reversing cameras have a place, though many still back into stationary objects. The oh so safe really are not. 



#38 Lee Nicolle

Lee Nicolle
  • Member

  • 11,036 posts
  • Joined: July 08

Posted 07 April 2015 - 06:06

This seems like someone moved the decimal point over accidentally. Most of the time on the freeway I can see when someone is doing something stupid before they actually do it and I'll position myself accordingly.

Often you can see it 5k before. Looking ahead and anticipating is what driving is all about. Road or racetrack.

Though nanny laws sometimes preclude commonsense. Recently I was pulled over for not keeping left [driving on the left] The reason ofcourse was that car and trailer was around 40 feet long. And too do the speed limit [freeway] I would have been weaving in and out the whole way. Not very safe! IF someone was coming up faster I would have gone left. Nanny laws cause accidents, road rage and just plain rage. Every second driver drives under the speedlimit so to do the limit yet alone god forbid above it means a LOT of looking ahead.



#39 275 GTB-4

275 GTB-4
  • Member

  • 8,274 posts
  • Joined: February 03

Posted 07 April 2015 - 11:08

and to finish , rather than just reduce the chance of an accident.
 
A somewhat large and off topic question in way but the several things have made me ponder on this - as a non engineer.
 
Firstly all the talk of driverless cars.They will prevent roads deaths has morphed intoa  statement by Ed Musk of Tesla " human drivers may have to be banned"- i.e engineers can stop accidents so the reponsiblity lies with them not individuals
 
Some peple are demanding pilotless planes immediately to stop pilot suicides costing lives. Even in this forum the question is " what engineering solution to the Nurburgring spectator fatality".
 
Don't get me wrong , Im not condemning such demands as stupid, silly or unaffordable but I was just pondering what the people here think that society now demands of engineering - including software design?
 
My sense is two things - Western Societies tolerance of risk has reduced substantailly over the last 30 years. Motor racing is no exception. It has ben made so much safer , thanks to engineers and sponsor/Bernie money. However if by some freak we had a year of F1 deaths like the 1970's would it now be banned or reduced drastically in speed/content?
 
Secondly safety has always demanded trade off's. As I understand it that is exactly how the FAA/CAA go about their work which seems to be less impacted by politics than say car  design.
 
As I say I am not an engineer but do recall one example 40 years ago of sitting in on a value of human life discussion. I worked in an oil company abut to commit to the production platforms in the UK North Sea. The number of platforms is determined by drilling technology but the size so cost of each one is basically driven by the weather.
 
Platform size equals cost. Size is deck area times height above mean sea level. If you make the deck small you have to re supply more and bad weather stops supply ship unloading hence you shut down operations. Safety dictates the deck is above the highest likely wave.
 
So we called in a weather expert who said basically " do you want to design for a 20 year, 50 year or 100 year wave?"
 
A one hundred year North Sea wave is 90ft plus or as he put it the height of a ten storey building. Having finally accepted that size was possible the discussion was " well, no not one hundred wave so how about 50 year wave as that is 2.5X the expected platform life.
 
 The discussion went on a long time as everybody did realise lives were involved. My sense is ,and I may be wrong, that such a conversation would not be allowed , or recorded today lest it was used to prove " profit before lives etc"
 
What to people think? Are such discusions stil allowed/commomplace in design?
 
I  suppose the ideal solution would be for goverments to state clearly their acecptable money value of a citizens life and let the enginers prove they designed compotently to that level.


Of course they are! Cost vs Engineering trade-off vs acceptable risk is commonly used in systems design...are you new to this planet?

Advertisement

#40 Magoo

Magoo
  • Member

  • 3,694 posts
  • Joined: October 10

Posted 10 April 2015 - 01:36

Since a court in Georgia last week awarded $150 million to the family of a child roasted to death in a Chrysler SUV fire, folks might want to revise their figures upward a bit. 

 

The fact is that a dollar value cannot be placed on a human life in any coherent moral system yet devised.


Edited by Magoo, 10 April 2015 - 01:42.


#41 Greg Locock

Greg Locock
  • Member

  • 6,349 posts
  • Joined: March 03

Posted 10 April 2015 - 03:13

I wonder what that'll get reduced to on appeal.  " the rear view cameras for SUVs failed the cost effectiveness test by a factor of 10 but still went ahead, from memory.Given the tendency of American courts to impose huge multiples of actual damages as punitive measures, this may be a wise precaution,", as I said earlier.



#42 GreenMachine

GreenMachine
  • Member

  • 2,629 posts
  • Joined: March 04

Posted 10 April 2015 - 10:05

Spot on Greg.

 

Magoo, morality is one thing, economics are another.  Economists are always putting values on things (the environment, human life, human misery), for the purposes of analysis, .  Nobody seriously suggests that if I pay $1m (say), I can morally go out and kill someone.



#43 275 GTB-4

275 GTB-4
  • Member

  • 8,274 posts
  • Joined: February 03

Posted 11 April 2015 - 06:58

and when will Corporations etc be brought to account...

Doesn't matter whether it is cars (autos), drugs or prosthetics or just about any other consumer goods you care to name, the appallingly criminal mindset and bad behaviour just keeps on surfacing...

The actions of some businesses make Gordon Gecko almost look like a saint...

The bottom line is one thing (and nobody wants to see unprofitable enterprises) but ethics or lack thereof is another matter.

What the hell do they teach em in Business School?

#44 BRG

BRG
  • Member

  • 25,883 posts
  • Joined: September 99

Posted 11 April 2015 - 19:27

What the hell do they teach em in Business School?

Not to exaggerate and make wild generalisations?



#45 Magoo

Magoo
  • Member

  • 3,694 posts
  • Joined: October 10

Posted 11 April 2015 - 20:40

Spot on Greg.

 

Magoo, morality is one thing, economics are another.  Economists are always putting values on things (the environment, human life, human misery), for the purposes of analysis, .  Nobody seriously suggests that if I pay $1m (say), I can morally go out and kill someone.

 

 

You cannot place a dollar value on a human life.  That, among other things, is the lesson of the Ford Pinto debacle, as well as any number of other disasters that purported to use cost/benefit analysis to perform this dubious feat. I don't mean that it's simply wrong. I mean also it can't be done. It's a lie, a fraud. When the geniuses at Ford calculated the cost of a defective fuel tank design in human lives, they presumed the cost to Ford, not the cost to their customers. You will find similar logical blunders in all such attempts. 

 

...every year a few hundred utility workers are killed on the job supplying electrical service across the USA. In the customary cost/benefit analysis as performed by the usual ****wits, this is determined to be a sound tradeoff: electric power for millions vs. the death of a few hundred workers. Great, super. Except that when we examine each one of these accidents individually, we find that not one was actually required in order to distribute electrical energy. Every death was needless and unnecessary and accomplished nothing at all. If there is any value at all in these tragic accidents, it is as teaching examples to avoid such disasters in the future. To claim these deaths are justified in cost/benefit analysis is a lie. There is no benefit. 



#46 275 GTB-4

275 GTB-4
  • Member

  • 8,274 posts
  • Joined: February 03

Posted 11 April 2015 - 21:44

Not to exaggerate and make wild generalisations?


Wha???? I studied electronics and computers and there was scads of exaggeration and wild generalisations taught in that world...

#47 Magoo

Magoo
  • Member

  • 3,694 posts
  • Joined: October 10

Posted 11 April 2015 - 23:06

In general, cost/benefit analysis in regard to human life is an exercise in the logical fallacy known as False Choice. 

 

Let's perform a cost/benefit analysis of drunk driving.

 

Sure, a few thousand people each year are killed, maimed, or dismembered by intoxicated drivers. How truly unfortunate. But on the other hand, many more thousands, by far the vast majority of drivers under the influence, usually manage to find their way home without fatally injuring or permanently disfiguring a soul. So in terms of the many, many thousands of passenger miles traveled each year in relative safety, not to mention all the wonderful hours of entertainment and relaxation in the process, isn't it totally fair and accurate to assert that overall, drunk driving provides a huge net benefit to society? 


Edited by Magoo, 11 April 2015 - 23:19.


#48 Greg Locock

Greg Locock
  • Member

  • 6,349 posts
  • Joined: March 03

Posted 11 April 2015 - 23:46

OK, so how do YOU evaluate whether a given thing should be deployed? You seem to be arguing that all preventable deaths irrespective of the cost of preventing them should be prevented if humanly possible. When you apply that argument to to end of life medical care you get ludicrous results, and when you apply that argument to aircraft design you will never be able to afford to fly again.

 

However I do agree that deaths in the workplace should be treated far more seriously than they are, I was disgusted to read that BHP were fined 75000 dollars for one death.

 

Cost benefit analysis is a pretty blunt instrument, but unless you can propose a serious alternative then your objections boil down to "I don't like it" which works with 5 year olds and food, but not so much with the grown up world.

 

You use CBA in your own life when you choose not to fit the best possible tires (not the ones supplied by the manufacturer which are built to a price, have low rolling resistance, and a comfortable ride, and don't generate too much noise) to your cars and maintain them in tip top condition, and not to buy your children large and new cars to drive around in, rather than hand me downs. 


Edited by Greg Locock, 12 April 2015 - 02:11.


#49 mariner

mariner
  • Member

  • 2,325 posts
  • Joined: January 07

Posted 12 April 2015 - 03:40

One question I would be interested in knowing about safety CBA is whether maintainace costs of safety systems are taken into account?

 

Its fine for people rich enough to buy a new SUV to be "happy to pay an extra $200 to avoid running over their kid" etc but the poorer people who run older cars which go wrong don't always have the income to fix a safety system failure. For them a car is often vital to hold down a job. In the Uk at least at the annual safety (MOT) inspection if an AIrbag,ABS, ASC or TPS light is on the car fails.

 

Obvioulsy not every safety system wil fail but if it does the costs can be more than a poor person can afford so no car, maybe no job, at MOT time.

 

My Fiat Panda at 6 years old may be a bad case, but a broken seat wire in the airbag harness was an MOT failure issue. Fiat wouldn't cover it as it was 6 years old but they also insist the whole airbag harness is replaced for one wire broken, The dealer quote? $3,500. Yes $3,500 to fix a wire. $1000 for the full harness and 12 hours of dash rip out labour at $200/hour.

 

The car was worth $4,000 max. I have a second car and enough knowledge to find an alternative fix but if I were some low wage Joe Public running this basic car I am not likely to have the odd $3,500 in the bank so no car, no resale value as no MOT and posibbly no job.

 

I recognize the cost to Society of car deaths but if the systems to reduce such risks means the poorre members of the same society can't run a car to get to work thats a cost to society too.


Edited by mariner, 12 April 2015 - 03:41.


#50 Magoo

Magoo
  • Member

  • 3,694 posts
  • Joined: October 10

Posted 12 April 2015 - 21:32

OK, so how do YOU evaluate whether a given thing should be deployed?

 

 

Through the exercise of sound, everyday moral judgement. Pretty corny, huh. 
 
Some here have even stated that morality and commerce are incompatible. I disagree. 
 
Reasonable moral judgement would have prevented the Pinto fuel tank, Corvair swing axle, GM ignition switch, and numerous other automotive disasters---and note, at a great economic benefit to the companies. The automakers did not profit from these ethical blunders, on the contrary. They were gravely wounded by them. Here, a little moral reflection would have saved big bucks. 
 
When we offer a product to the public, we accept a responsibility which compels to us ask: What can we reasonably do to make the product safer?  It is in our benefit to ask this question every day, not just the consumer's. What can we do to make the product safer as long as there's a buck in it is not the question. That's what cost-benefit analysis does. It doesn't consider the cost or benefit to the consumer but to the manufacturer. 
 
The concern here is comically misplaced. In the entire history of the automobile industry---spanning some 126 years and over 8000 manufacturers---how many companies have gone bankrupt by spending too much on the safety of their customers and workers?