Re: [sc] Road Traffic Safety



Date view Thread view Subject view Author view Attachment view

From: Peter B. Ladkin (ladkin@xxxxxx)
Date: Fri 24 Dec 2004 - 08:51:08 GMT


Andrew Shore questions how and why speed limits are set.

I don't propose to get into a discussion of whether and
how road traffic engineers working for a democratic government
represent a choice by society. But I think it is important to
understand some engineering and behavioral reasons for setting
speed limits.

Engineering reasons are environment, physics, and planning.

Environment. If obstacles are likely to appear in your path,
if there are corners, if there are places beyond which a driver
cannot see, if there is multiple use (pedestrians, cyclists,
including children) and so on, then car drivers must be able
to react and stop before an obstacle which appears.

Physics. Newton's laws of motion. Coefficient of friction.
Manoeuvrability.

Planning. Signs and warning devices, curves, turn-outs all
have to be planned assuming a given (maximum) traffic speed.

All of these put constraints on driving behavior which are
not understood or appreciated by the majority of drivers.

To illustrate this point, consider the following. Tell me the
stopping distance of your car under heavy but normal braking
at 35mph in the dry. Now again in the wet.

Most people can't. (And yes, I have calculated and memorised
the theoretical stopping distances of my bicycles. But I was
still surprised yesterday in moderately heavy rain; I hadn't
taken into account for this particular bike - I have a few -
the time it takes for the blocks to wipe the rims under
heavily-wet conditions.)

But traffic engineers know such figures when they are devising
clearways for automobiles and other road users. They
must devise these under certain assumptions, such as a maximum
speed of traffic at certain points, and these assumptions
must be codified into a speed limit in order to ensure their
continued validity and thus the adequate safety of the construction.

We assume those traffic engineers perform an appropriate analysis
of the environment and the physics while performing their planning,
and we may assume that they spent more time doing so than the few
seconds which a car driver who is using that facility typically
has in which to assess the situation.

Now social reasons.

Mark Bowell has indicated how individual decision making by
drivers on what is appropriate or not can lead to interactive
conflict. So leaving it up to individual drivers to determine
what is adequately safe or not in given circumstances is not
a modus operandi for which will ensure adequately safe traffic
operation.

There is a fifth factor, also a social reason, why it is
necessary to adhere to speed limits. Road traffic is a social
situation in which there are many agents. We must make assumptions
about how other road users are going to behave. Car drivers passing me
within the 1.5m-2m clearance that they are supposed to leave are proceeding
under the assumption that I will not wobble left into their vehicle,
and so far they have been correct in that assumption. One
very important cognitive component to decision making, especially in towns,
is to judge the time in which oncoming traffic will reach me, and to judge
general manoeuvring capabilities of others. It is necessary to set a speed
limit in order for all road users to make cognitive assumptions about those
parameters that will enable accurate decision making: when I see a car
*there*, I know I can cross without hurry and in adequate time to be
seen myself. When much closer, then not. There may be no "right" limit 
according to this criterion: 30kph may work, also 40kph, even 50kph. But
there has to be one. And people travelling faster than this limit are
violating the assumptions that others are entitled to make about their
behavior; they are ipso facto creating a situation which is less safe than
planned or expected.

One example from a few years back. When my son was tiny, I hauled
him about in a trailer behind my recumbent. The total vehicle was long,
some 4m or so. And we accelerated slowly (braking was good, because
of the dynamics of recumbents). Although we were long, slow and wide,
I suspect it was actually a lot less risky than travelling about on
my own, because were were very yellow and comparatively large, so could
be seen well under all conditions, and represented an object about which
others had to manoeuvre. There is also a special taboo about endangering
very young children. I cannot recall anyone passing too close in three+
years of using it.

However, there was one major incident. There is a particular intersection
with a middling-major road which we used to traverse regularly. Since
traffic was parked at the side of that road, line of sight was impaired.
But one gets judges the distance one must be able to see, and the time one
has to cross, even allowing for those who might be travelling faster than
the 50kph limit. Once, I performed the usual stop-and-carefully-look, saw
no traffic at all, proceeded to cross. I was about two meters into
traffic when I suddenly saw, and heard, a car travelling towards us at
what must have been 80kph or so, well over the limit of 50kph and much
faster than any traffic I had ever seen there. Brakes went on; I stopped in
about another meter. But there I was, some three meters into the roadway
with Simon behind me, with no further manoeuvring possibility. The car
swerved into the opposite lane (there was no other traffic in sight). It
was full of yelling youths.

They had violated the assumptions under which I made my decisions. Were
this a road in which such things frequently happened, then I would have
been in the habit of using the intersection differently; in fact, I
would have used a different route some 40m to the right, which I started
to do after this incident, which has extended line of sight.

If I had been in a car, I would likely have accelerated faster, taken
longer to stop (constant reaction time at slightly faster speed; slightly
lengthened stopping distance) and there could well have been a collision,
since there would have been less room for the other to manoeuvre away.
And the alternative route, which is a pathway, would not have been
available to me in the future.

Speed limits set assumptions under which roads can be effectively
jointly used. There is no way of judging a speed "safe" or not from
an individual perspective if it is higher than the set limit unless
one is a very thorough socio/anthropologist of precisely that
microenvironment.

To summarise: because of the constraints of environment, physics,
and planning assumptions under which roadways are engineered,
judgments by individual drivers, no matter how smart or aware
they may be, on "safe" speeds are necessarily unreliable. Further,
Mark Bowell's example shows that coordination of many road users
does not follow from individual optimisations, but must be
superimposed through general constraints (such as speed limits).
Finally, road users develop practical expectations of the behavior of
others, which is constrained by a maximum expected speed, and violations
of those practical expectations invalidate individual decision making and
thereby increase risk in many situations.

Those are five good reasons why it cannot be left to individuals, even
smart, experienced individuals, to judge appropriate traffic speeds as
they pass through.

On another matter raised by my example: there is quite a lot to be said
from the safety point of view for raising the driving age to 25. And
for introducing a year or two of compulsory civil service, especially in
ambulances and clean-up crews for males, for all citizens of ages 18-20 or
so.

PBL

-- 
Peter B. Ladkin, Professor of Computer Networks and Distributed Systems,
Faculty of Technology, University of Bielefeld, 33594 Bielefeld, Germany
Tel+Msg +49 (0)521 880 7319     www.rvs.uni-bielefeld.de

Date view Thread view Subject view Author view Attachment view