[SystemSafety] Self-Driving Car Had a Fatal Accident ... (Norman, R 30 61 & many R 30.62)

[Peter Bishop]

On 01/04/2018 09:38, Peter Bernard Ladkin wrote:
> It is a little odd to see colleagues ignoring the difference
> between statistics and numerology. Statistics involves drawing
> robust conclusions from data to a specified level of confidence.
> Numerology is doing arithmetic on numbers.
> 
> One incident is not data, it is, at best, a datum. Consider: if
> the Risks discussion on the numbers associated with self-driving
> cars had taken place on March 17th, it would have "concluded" that
>  self-driving cars were the safest form of road use ever known to 
> mankind. On March 19th and since, people are "concluding" that it
> is an order or two of magnitude "worse" than human-driven
> automotive vehicles. The term "robust" is not applicable to
> inferences that flip so drastically between one day and the next.

I would take issue with this assertion.
This is only the case if we naively divide accidents by mileage.

Statistically, if we had 1 million miles and zero accidents, we could
only be 63% confident the rate is better than 10-6 /mile. So we cannot
conclude "it is the safest form of transport known to man".

If we had one accident in the next mile, the confidence drops to 26%
in 10-6 /mile, but we would have 60% confidence the rate is better
than 2x10-6 /mile.

So there is no massive step change for a specific confidence level.

For human driving, the 63% confidence figure is around 10-8/mile.

So for similar confidence levels, self-drive technology *is* a couple
of orders of magnitude worse than than the human (at the moment).
> 
> Not only that, but the technology of the accident vehicle may well 
> change in response to the one event, such that that event would
> not have occurred with the newer technology (operating as
> designed). So those gathering numbers can start over again from 0.
> And this can go on and on. Indeed, there is an argument (see below)
> that (a) this is how commercial aviation has addressed digital
> automation, and (b) commercial aviation is by far the safest mode
> of travel by a number of measures, and (c) continues to improve its
> safety record for high-performance jet transports using any
> measure.

True, lumping accidents together only gives an average of a
potentially improving rate. Mind you, the technology is being given an
easy ride at the moment (benign weather conditions, reasonable roads)
so the average could get worse as the operational scope increases.

-- 

Peter Bishop
Chief Scientist
Adelard LLP
24 Waterside, 44-48 Wharf Rd, London N1 7UX
http://www.adelard.com
Recep:  +44-(0)20-7832 5850
Direct: +44-(0)20-7832 5855

Registered office: Stourside Place, Station Road, Ashford, Kent TN12 1PP
Registered in England & Wales no. OC 304551. VAT no. 454 489808

This e-mail, and any attachments, is confidential and for the use of
the addressee only. If you are not the intended recipient, please
telephone 020 7832 5850. We do not accept legal responsibility for
this e-mail or any viruses.