Re: [sc] Bad Safety - the absence of science and engineering



Re: [sc] Bad Safety - the absence of science and engineering

From: tracyinoz_at_xxxxxx
Date: Tue, 18 Oct 2011 13:22:34 +1030
Message-id: <034ECACD-B52D-4648-B1A4-C1CC8966D8B6@xxxxxx>
Wilhard

I think your list points to, and supports,  what most people have been saying throughout this thread, which is that safety risk associated with road traffic is a complex problem. It should not be expected therefore that the addressing of that risk should be any less complex and Nancy quite rightly questioned whether: " ...arguments involving kinetic energy are going to answer the question".

We have seen from responses that the risk can both increase and decrease with variation in speed so to have a 1-dimensional approach is not an realistic  approach. Scientifically speaking, claiming reductions in risk via speed variations assumes that  "all things remain constant", which they clearly do not; it also assumes that controllability and exposability are not the more dominant factors, a claim to which is an essential element of spped-justification; it is a position that PeterL endorsed when he stated that: "You need the argument (that is, the proposed causal model) in order to let your evidence work."

The 'science and engineering' I referred to in my original posting was a reference to that complete argument-evidence justification, the existence of which is an essential element of any claim to be effectively assuring safety; the absence of that argument underpinning the proposed  speed-reduction' control' evidence relegates the claimed 'science' of that evidence to a mere  interesting factoid rather than anything constructive/effective.

Regards, Tracy

On 16/10/2011, at 12:58 AM, Wilhard von Wendorff wrote:

> Hi Nancy,
> The functional safety standard for electronics in cars and light trucks ISO/FDIS26262 names three major factors to be taken into account for a safety risk analysis:
> 1.) Controllability of safety incidences,so asking the question if a driver can control a safety related situation
> 2.) Exposability, so how often is a driver experiencing the safety relevant situation
> 3.) Severity, so how serious are the injuries due to a safety incidence
> 
> The major focus of the automotive industry in the past for my understanding was to increase controlability (e.g. ESP/VSC- Vehicel stability control) and decrease severity (airbag, structural crash absorbing), as these are topics which can be directly addressed by commercial systems and electroncs. Exposability is something heavely influenced by society. Therefore for my personal feeling this is not the focus of the private industry but has to be the focus of government and society. I see due to commercial reasons a huge development in increasing the controlability and decreasing the severiy, but less efforts to reduce the exposability. This leads to a bumerang effect. As systems and electronics improve the controlability, drivers tend to drive faster and more risky to stay in the same "comfort (controllability) zone" society has got used to due to the experience in the past. Studied showed that many ESP driver increased their average speed or decreased the distance to drivers ahead if using adaptive cruise control. On the other hand for my opinion the efforts of the private industry and society reducing severity (laws regarding seat belts, helmets on motor cycle, etc.) works out fine. As humans do not objectively use severity in their personal risk assessment (we tend to believe, an accident will not happen therefore do not take the severity of injury into our risk calculation) with the same weight as controllability (specially we men tend to believe ESP and a big SUV is great, so lets drive a little faster on this icy road).
> 
> But I see especially in Germany a need to decrease exposability. Decreasing exposability in general tends to decrease personal freedom and reduces quite often consume. Examples are speed limit, decreasing blood alcohol limits, etc. So decreasing exposability very often comes along with some type of abstinence. As abstinence is not an attribute having a high positive image within e.g. my German society, it is difficult to decrease exposability beyond the very obvious topics (drunken should not drive, driver need a license, helmets for motor cycles, helmets for children on bicycles, regularly inspection of vehicles).
> 
> So yes, we will have to change the personal experiences of every individual regarding safety. But this is not as fast as developing the next electronic gadget in the car. My children, not knowing any more the "good old times" driving a car not wearing seatbelts, issues with breaks not reliable working, children not sitting in special seats, cars as stiff as a shoe box, yes, they will have a different personal risk assessment and yes they may ask maybe for lower speeds. But the old generation like me. Will we change or will we die first?
> 
> Wilhard von Wendorff
> 
> Am 11.10.2011 14:56, schrieb michael.ellims@xxxxxx:
>> Good afternoon Nancy,
>> 
>> was it intend that this rather massive missive be addressed as a series of questions as written or as to comment on the final point (11) and the concluding paragraph? I could approach it either way...
>> 
>> Cheers.
>> 
>> ---- Nancy Leveson<leveson.nancy8@xxxxxx>  wrote:
>>> "
>>> Content-Type: multipart/mixed;
>>> 	boundary="000e0cd57004616d7304af03db09
>>> "
>>> X-YCS-Spam-Score: 0.0 (/)
>>> 
>>> --000e0cd57004616d7304af03db09
>>> 
>>> Content-Type: text/plain; charset=windows-1252
>>> Content-Transfer-Encoding: quoted-printable
>>> 
>>> I know nothing about road safety. But there seem to be a lot of issues not
>>> mentioned so far that might (or might not) be relevant:
>>> 
>>> 1. Controllability: Is a driver more likely to lose control at higher
>>> speeds? if a driver loses control at a higher speed or if something goes
>>> wrong in the car, is there more likely to be an accident? What about
>>> weather, e.g., wet and slippery roads?
>>> 2. Enforcement: posted speed limits may not have much impact on how fast
>>> people drive but instead the number of tickets that are given in that area
>>> or on that type of road may be more of a factor.
>>> 3. Design of automobiles: I think cars are designed to pass collision tests
>>> so that, for example, the force may be taken by the bumper instead of the
>>> passenger compartment (the bumper crumbles) up to a certain speed. What
>>> about at higher speeds? Are such designs actually more dangerous at  higher
>>> speeds?
>>> 4. Design of roads: the design of the roads will affect the number of and
>>> lethality of accidents.
>>> 5. Amount of traffic: On lonely, straight highways in very rural areas, is
>>> it safer to go fast? Or is there more likely to be a deer or farm equipment
>>> that suddenly crosses or pulls onto the highway?
>>> 6. Condition of the roads: Are they well maintained, pot-hole free, etc.?
>>> 7. Distance between cars when impact occurs: Is there more tailgating at
>>> higher speeds because those ignoring the speed limits come up close to try
>>> to intimidate the driver in front into letting him/her by? Or perhaps it is
>>> a matter of relative speeds.
>>> 8. Relative speeds: the German autobahns used to be the most dangerous road=
>>> s
>>> in the world (although I think I read the death rate has improved). One of
>>> the reasons was the very high speed limit on two lane highways where the
>>> trucks were going very slowly in the right hand lane while cars were
>>> whizzing along at extremely high speeds in the left lane. I once rented a
>>> car with very little power in Germany and greatly regretted it. I tried to
>>> pass the very slow trucks in the right lane, but had difficulty getting up
>>> enough speed to avoid cars on the left that were going so fast they were no=
>>> t
>>> visible before I started changing lanes before they were suddenly up on my
>>> tail. In addition, do lower limits make people drive at relatively equal
>>> speeds (i.e., do the slower drivers slow up or is it more likely that the
>>> faster ones slow down). Is it harder to pass safely (or to judge the
>>> distance required to pass safely)?
>>> 9. Psychology: Do people drive more recklessly at higher or lower speeds?
>>> (e.g., fantasies of being a race car driver)
>>> 10. Weather: what about areas where there is likely to be fog and other
>>> weather conditions that often lead to massive pileups?
>>> 11. Probably a lot more I forgot or don't know about. As I said, I know
>>> nothing about road safety. But I don't think that arguments involving
>>> kinetic energy are going to answer the question.
>>> 
>>> Simply using statistics, when the things being compared are very different
>>> and the sample size is not large enough to account for the differences (suc=
>>> h
>>> as terrain, straight versus curvy roads, weather, etc.), is not enough for =
>>> a
>>> safety analysis. In addition, there is a need to compare the actual speeds
>>> limits involved, for example, people might be more likely to be killed at
>>> higher speeds but have more accidents at lower speeds. Basing these
>>> decisions simply on statistics ignores too many other factors that need to
>>> be considered in a safety analysis.
>>> 
>>> Nancy
>>> 
>>> On Tue, Oct 11, 2011 at 6:17 AM,<michael.ellims@xxxxxx>  wrote:
>>> 
>>>> Part 1 of who knows how many... (apologies this has raised my interest)
>>>> 
>>>> The question on changing the speed limits in the USA is interesting,
>>>> because it is counter-intuitive AND the effects of the changes were
>>>> incorporated within the meta-analysis I mentioned before.
>>>> 
>>>> So far I have tracked down a single paper on-line, IMPLICATIONS OF THE
>>>> 65-MPH SPEED LIMIT FOR TRAFFIC SAFETY, DAVID J. HOUSTON
>>>> 
>>>> The headline is that on the roads effected, the rate of accidents increas=
>>> ed
>>>> i.e. "Based on the analysis reported above, there is evidence to suggest
>>>> that higher speed limits have had negative safety consequences on those
>>>> roads directly affected by the change in posted speed limits=94.
>>>> 
>>>> There is of course an =93however=94 i.e. =93states have experienced a red=
>>> uction
>>>> in fatality rates on other types of roads<snip>  reductions in fatality
>>>> rates are enough to offset the increase experienced on rural interstate
>>>> highways=94.
>>>> 
>>>> And why? The author suggest =93that riskier drivers will be diverted on t=
>>> o
>>>> the rural interstate highway away from other more dangerous roads, follow=
>>> ing
>>>> a change in posted maximum speed limits to 65 mph=94 and as the interstat=
>>> e
>>>> roads are safer be design... e.g. more Armco...
>>>> 
>>>> Cheers.
>>>> 
>>>> 
>>>> 
>>> 
>>> --=20
>>> Dr. Nancy Leveson
>>> Professor, Aeronautics and Astronautics
>>> Professor, Engineering Systems
>>> MIT
>>> 
>>> http://sunnyday.mit.edu
>>> 
>>> --000e0cd57004616d7304af03db09
>>> 
>>> Content-Type: text/plain
>>> X-Original-Content-Type: text/html; charset=windows-1252
>>> 
>>> 
>>> [The content of this part has been removed by the mailing list software]
>>> 
>>> --000e0cd57004616d7304af03db09
>>> --
>>> 
>> 
>> 
>> 
> 
> 
> -- 
> Wilhard von Wendorff Atterseestrasse 14 81241 München Tel: +49 89 8344566
> 
Received on Tue 18 Oct 2011 - 03:52:57 BST