[SystemSafety] Royal College of Paediatrics weighs in

[Peter Bernard Ladkin]

On 2014-11-19 11:44 , Peter Bernard Ladkin wrote:

I just came across the chapter on Transportation Safety in the Handbook of Transportation Science:
http://www.scienceservingsociety.com/p/134.pdf The author is Leonard Evans, who appears to have
grown up in Northern Ireland, did a Physics D.Phil. at Oxford, a post-doc in Canada, and worked at
GM Labs from 1967-2000. I presume he retired from GM in 2000. He has been running an organisation
called Science Serving Society since then. Does anyone here know him?

Evans uses the commendable style of writing third person about facts and first person about his
opinions. It is still rare to find people in science and engineering making this distinction, though
to my mind it's helpful to the reader.

Evans says inter alia that the reduction in fatalities on rural interstate roads concomitant with
the 55mph speed limit imposed in 1937 was 34%, so about double the overall reduction which I quoted.
And of course it is the rural long-distance roads on which the speed limit would have had most
effect in terms of altering behavior.

He divides driver phenomena (my term) into two: driver performance, by which he means the
capabilities of drivers, and driver behavior, which is what drivers actually do.

He suggests that changing driver phenomena makes the biggest contribution to road safety, as
contrasted with technology. He is particulary scathing about the claims made for airbags versus
their actual safety performance.

Alan wrote

> On 2014-11-19 11:09 , Dominey, Alan (UK) wrote:
> ..........
>> IMHO, the answer [to reducing road traffic unsafety] is BETTER driver training and
parents/schools bothering to educate properly on road use

In constrast, Evans claims that driver performance makes little difference to the road accident
statistics. He attributes the largest potential safety gains to modifying driver behavior,
mentioning the three phenomena of changing social norms, legislation and enforcement.

His example of changing social norms is drinking and driving. I think there is no need to explain
here how these norms have changed over the last (say) thirty or forty years, or even adduce the
measurable positive effect of this change on road safety. But this positive influence is limited.
In the same period, I would suggest that social norms concerning driving at a speed faster than
posted haven't changed much, if any, and may even have loosened (notice that "performance" figures
posted for cars usually include acceleration and high-speed achievement; they usually don't include
deceleration, or behavior such as braking on a curve).

Legislation and enforcement is a trickier issue than it appears, because what counts as legislation
is relative to the country. For example, the Transport Research Laboratory said to the UK
Parliament's Select Committee on Transport on 15 March 2006 that "Road traffic law is one of the
main tools available to society to reduce the numbers and severity of road accidents, by defining
behavior which is held to be unduly risky - such as drink-driving - to be illegal" (see the Minutes
of Evidence from 15 March 2006, submission by the TRL, at
http://www.publications.parliament.uk/pa/cm200506/cmselect/cmtran/975/97502.htm ). This is literally
not true in Germany, where road traffic law is hardly related to accident causes at all. German
traffic *law* is all about how vehicles get to be qualified to use public roads, how drivers become
enabled, and how permission to drive on public roads can be rescinded. It's not about "drive on the
right" or "traffic approaching from the right has priority unless the intersection is otherwise
signed" or "driving above posted speed limits is prohibited". That's all in the "regulations" or
"Verordnung" and those can be changed through different mechanisms than those required to change laws.

The TRL is talking about how behavior on the road is controlled, that is, about regulation. If I
were to phrase what Evans is referring to as "Rules of the Road" (RotR), I think that would come
close to what he intends by "legislation" and refers to Germany also.

I think it is inadvisable to distinguish technology from RotR so sharply nowadays, in light of the
possibilities of automatically enforcing RotR in-car. For example, as well as commending enhanced
surveillance technologies, the Transport Select Committee report cited above commends "Intelligent
Speed Adaptation [sic]", which is basically the automatic regulation of speed. For example, when a
car enters an urban area signposted at speed limit 30mph then its top speed is automatically
regulated in-car to be 30mph. Quite simple technology exists to do this, of course, leaving aside
questions of operational reliability. Since excess speed is a major contributor to injurious
accidents, and a significant portion of excess-speed accidents involve speed in excess of posted
limits, universal in-car speed limit enforcement could bring the numbers down quite a ways.
Technology is also available to inhibit "drink-driving", which is still a major phenomenon
associated with injurious accidents.

PBL

Prof. Peter Bernard Ladkin, Faculty of Technology, University of Bielefeld, 33594 Bielefeld, Germany
Tel+msg +49 (0)521 880 7319  www.rvs.uni-bielefeld.de