[SystemSafety] Looking for information about safety-critical software faults in road vehicles

[Mario Gleirscher]

Hi all,

the following could be seen as a more general comment to that request:

About two years ago, I had mentored a master student whose task it was
to go through case reports (reports on accidents, incidents, and the
like, you know) to find out about software-related issues. The student
went through say about 300 such reports from various databases, one of
the larger automotive case report databases was very nice to search in
but of little use to get informed about root causes.

One of our conclusions was that quality of root cause documentation in
case reports is varying drastically across domains. The situation is
quite ok for e.g. train systems, but comparably immature in the
automotive domain. We were first surprised and had a few discussions,
but of course realized that there might be quite a few reasons for not
disclosing such information in case reports beyond a low level of detail.

However, anecdotal evidence (from cases we all know about and from
further interviews we were able to conduct) suggested that SW might be
more often part*) of the root cause than we observed from the reports.
However, we were not able to have this, at least in pieces, confirmed
from the case reports. Frankly said, that was a bit frustrating!

*) And I like to make that clear: I am not only talking about the more
directly tangible issues in older/more mature domains like motor control
or the braking loop (a few years ago), but particularly about
interaction phenomena in the distributed networked systems that are
known to be almost impossible to handle without proper methodology and
regularly generate tremendous amounts of analysis effort (not to talk of
the costs from that, weekly annoying car manufacturers).

I am curious to hear more about the cases people on this list know?

Best wishes,
Mario

On 20/08/18 18:20, Andrew Banks wrote:
> I'm not sure how many publically available responses this will generate...
> 
> I am aware (as a developer at and/or a consultant to, several companies) a number of issues that were caught late but fixed without fanfare - but I'm not at liberty to discuss in depth.
> 
> Particular areas that have come up repeatedly are to do with battery management (eg triggering thermal runaway) and motor control, both in electric/hybrid vehicles.
> 
> A
> 
> -----Original Message-----
> From: systemsafety [mailto:systemsafety-bounces at lists.techfak.uni-bielefeld.de] On Behalf Of Martyn Thomas
> Sent: 11 August 2018 10:11
> To: systemsafety at lists.techfak.uni-bielefeld.de
> Subject: [SystemSafety] Looking for information about safety-critical software faults in road vehicles
> 
> Other than the "uncommanded acceleration" reports and the remote control of vehicles by researchers, can anyone give me details of safety-critical software faults discovered in road vehicle software?
> 
> I well understand that manufacturers may have corrected such faults (or
> not) without a recall or publicity and that details may be very sensitive. I will give strong assurances of confidentiality - I'm just trying to understand the scale of known issues.
> 
> Martyn
> 
> 
> 
> _______________________________________________
> The System Safety Mailing List
> systemsafety at TechFak.Uni-Bielefeld.DE
> 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/pkcs7-signature
Size: 5053 bytes
Desc: S/MIME Cryptographic Signature
URL: <https://lists.techfak.uni-bielefeld.de/mailman/private/systemsafety/attachments/20180820/9cd3fd4f/attachment.bin>