Prev: Some poplular songs for 2010
Next: another hidden safety recall
From: Tegger on 26 Feb 2010 09:21
"C. E. White" <cewhite3remove(a)mindspring.com> wrote in
news:tKidnbYDkIcf0xrWnZ2dnUVZ_oOdnZ2d(a)earthlink.com:

> Toyota gets more runaway complaints than rivals, study says
> Neil Roland
> Automotive News -- February 25, 2010 - 2:35 pm ET
>
> WASHINGTON -- Toyota Motor Corp. vehicles have been the focus of more
> unintended acceleration complaints in the United States than any other
> major automaker in the past five years, a new study shows.
>
> Toyota was the target of 532 consumer complaints involving 2005 model
> year vehicles and later, according to National Highway Traffic Safety
> Administration data tallied by Edmunds.com. That's 57 percent more
> than the 339 complaints involving vehicles from Ford Motor Co., which
> was No. 2.



Let's put this in perspective so as to highlight the tiny numbers we're
discussing here: Total automobile and light-truck sales for 2005 were
almost 17,000,000 vehicles (actually 16.94 million).

Out of those 17 million, there were 539 "complaints" of SUA involving
Toyota, and 339 involving Ford. Both, not coincidentally, are the top-
selling manufacturers. Other makers had less market share and fewer
incidents.

This means that at least 99.99% of the 2005 fleet proceeded as their
drivers intended. One-thousandth of one percent of the fleet generated
SUA complaints.

How perfect do you want the world to be? How perfect CAN you make the
world? How perfect are human beings?

At some point the numbers get so tiny that effecting ANY change on them
becomes nearly impossible without drastic and unrealistic action, such
as removing all control from the driver. When you're down to 0.001%, I
think you're already in "nearly impossible" territory.

There's also the possibility of simple "statistical clustering", which
was Audi's primary problem in the '80s. With tiny numbers like those
above, the possibility of clustering becomes a serious problem. That's
why statistical analysis has long demanded sufficient sample sizes.
Those raw numbers (539 and 193) might seem relevant when raw, but what
about what's left after unfounded complaints are weeded out, and after
the pedal-misapplication ones are removed? What's left? A handful? None?

Like car sales, no figures will remain absolutely unchanged relative to
each other over the years. Toyota had 193 more complaints than Ford. The
fact that Ford was at 339 complaints and not zero hints at the strong
possibility that (like with Audi), the current SUA controversy has
little to do with the cars themselves, and lots to do with the drivers.


>


<snip>



>
> Congressional and federal investigations of Toyota's safety record are
> focused on unintended acceleration. Such incidents have been linked to
> at least five U.S. deaths in Toyota vehicles,



That's misleading. Four of those deaths can be fairly confidently
attributed to a specific dealer's failure to install the correct floor
mats, and failure to to secure those mats properly.



> with 29 other fatality reports being examined by NHTSA.



Some of those incidents being ten-years-old, and which predate
electronic throttles. Why hasn't the NHTSA been able to assign blame
even after ten years?


--
Tegger

From: dr_jeff on 26 Feb 2010 09:48
Tegger wrote:
> "C. E. White" <cewhite3remove(a)mindspring.com> wrote in
> news:tKidnbYDkIcf0xrWnZ2dnUVZ_oOdnZ2d(a)earthlink.com:
>
>> Toyota gets more runaway complaints than rivals, study says
>> Neil Roland
>> Automotive News -- February 25, 2010 - 2:35 pm ET
>>
>> WASHINGTON -- Toyota Motor Corp. vehicles have been the focus of more
>> unintended acceleration complaints in the United States than any other
>> major automaker in the past five years, a new study shows.
>>
>> Toyota was the target of 532 consumer complaints involving 2005 model
>> year vehicles and later, according to National Highway Traffic Safety
>> Administration data tallied by Edmunds.com. That's 57 percent more
>> than the 339 complaints involving vehicles from Ford Motor Co., which
>> was No. 2.
>
>
>
> Let's put this in perspective so as to highlight the tiny numbers we're
> discussing here: Total automobile and light-truck sales for 2005 were
> almost 17,000,000 vehicles (actually 16.94 million).
>
> Out of those 17 million, there were 539 "complaints" of SUA involving
> Toyota, and 339 involving Ford. Both, not coincidentally, are the top-
> selling manufacturers. Other makers had less market share and fewer
> incidents.
>
> This means that at least 99.99% of the 2005 fleet proceeded as their
> drivers intended. One-thousandth of one percent of the fleet generated
> SUA complaints.

Are you sure? What percentage of problems actually result in a
complaint? For vaccines, about 10% of the adverse events are reported.
Minor ones, like swelling are less likely to be reported than major
ones, like those requiring hospitalization or a coffin.

So, we know that 532 Toyota drivers filed complaints, but we don't know
how many of those complaints are from driver error, how of those
complaints are because of real problems, and how many real problems did
not result in complaints.

> How perfect do you want the world to be? How perfect CAN you make the
> world? How perfect are human beings?
>
> At some point the numbers get so tiny that effecting ANY change on them
> becomes nearly impossible without drastic and unrealistic action, such
> as removing all control from the driver. When you're down to 0.001%, I
> think you're already in "nearly impossible" territory.

Another option is to build in reporting on vehicles' performance that
can go to the car maker and NHSTA. When there is an unexpected
condition, the info automatically get sent to the car maker and NHSTA.
Software companies have been doing this for at least 15 years. GM does
it already with engine problems. There's no reason why this can't be
done after a crash or when there is an unusual brakign condition.

Sorting out the data is another big issue, though.

<http://www.leftlanenews.com/onstar-chevy-nearing-100-percent-engine-and-transmission-reliability.html>


Jeff
From: SMS on 26 Feb 2010 09:50
C. E. White wrote:
> Toyota gets more runaway complaints than rivals, study says
> Neil Roland
> Automotive News -- February 25, 2010 - 2:35 pm ET
>
> WASHINGTON -- Toyota Motor Corp. vehicles have been the focus of more
> unintended acceleration complaints in the United States than any other
> major automaker in the past five years, a new study shows.
>
> Toyota was the target of 532 consumer complaints involving 2005 model
> year vehicles and later, according to National Highway Traffic Safety
> Administration data tallied by Edmunds.com. That's 57 percent more than
> the 339 complaints involving vehicles from Ford Motor Co., which was No. 2.

The Ford owners thought it was normal for Fords to do stuff like this so
they didn't complain. Toyota owners expect perfection so they complain a
lot more.
From: dr_jeff on 26 Feb 2010 10:20
SMS wrote:
> C. E. White wrote:
>> Toyota gets more runaway complaints than rivals, study says
>> Neil Roland
>> Automotive News -- February 25, 2010 - 2:35 pm ET
>>
>> WASHINGTON -- Toyota Motor Corp. vehicles have been the focus of more
>> unintended acceleration complaints in the United States than any other
>> major automaker in the past five years, a new study shows.
>>
>> Toyota was the target of 532 consumer complaints involving 2005 model
>> year vehicles and later, according to National Highway Traffic Safety
>> Administration data tallied by Edmunds.com. That's 57 percent more
>> than the 339 complaints involving vehicles from Ford Motor Co., which
>> was No. 2.
>
> The Ford owners thought it was normal for Fords to do stuff like this so
> they didn't complain. Toyota owners expect perfection so they complain a
> lot more.

Just how do you know that? You comment is mere speculation. If that.
From: Clive on 26 Feb 2010 10:38
In message <Xns9D2B5F3C9F68Ftegger(a)208.90.168.18>, Tegger
<invalid(a)invalid.inv> writes
>There's also the possibility of simple "statistical clustering",
Which over here we would call noise, because the sample is so low that
it cannot genuinely be separated.
--
Clive

First  |  Prev  |  Next  |  Last
Pages: 1 2 3 4 5
Prev: Some poplular songs for 2010
Next: another hidden safety recall