[SystemSafety] FMEA draft international standard

[Peter Bernard Ladkin]

On 2014-07-15 21:16 , Barrie Reynolds wrote:
> PBL said
>> But getting such a process underway is not trivial. First, people have to obtain a copy of the CD.
> 
> The UK committee is BSI DS/1
> Members of the supporting organisations should obtain a copy direct from their representative on DS/1
> The Supporting organisations are at the bottom of the page at:
> https://standardsdevelopment.bsigroup.com/Home/Committee/50001626

If I may take the opportunity to use Barrie's helpful message for a purpose for which it was not
intended.....

I am very concerned, as are other colleagues, about the restrictive practices involved in standards
development. There are two sides to it.

One is that some form of organisation is required, resources for any activity have to be sourced,
especially an activity such as codifying the state of the art in a specific technology while
controlling as far as possible for special interests, and the world is a very complicated place.

Another is that the sum total of practices discourage much participation in the endeavor, and the
results suffer technically. Many standards in some areas are unnecessarily technically poor.

For example, John Knight has complained publicly that he cannot obtain copies of standards to which
he has himself contributed as a committee member without paying out large sums of money to which he
does not have access in his academic role. And his students thereby cannot study them. He is right.
Almost all academia is hindered from using actual - even past - standards in their teaching, at
least in the area in which I work. That seems to be absurd if the standards' claim to codify current
state of the art is true.

Some national standards organisations do make provision for this. In Germany, a flat-rate
subscription somewhat less than the current cost of a leading journal buys an organisation access to
any German standards relevant to its area of interest. So I could get anything produced by IEC SC
65A (the functional system safety people) or IEC TC 56 (the Dependability people, which in the IEC
definition does not include safety, although in the IFIP glossary - Laprie, Randell and co - it
does) - but only in its German translation (to which the responsible national standards organisation
DKE has copyright - the official German standard is the translation of the IEC one).

Then there is the business of soliciting input and commentary from those with expertise who are not
involved in development.

BSI appears to distinguish two phases, a Committee Draft and a Draft for Public Comment. The
proposed IEC 60812 is not available as a Draft for Public Comment on the BSI WWW site. The German
DKE does not distinguish such phases, as far as I know. A Committee Draft is made available to any
who have registered a legitimate interest in the work of a given committee (the German correspondent
to BSI DS/1 is DKE K132). Registering an interest is open to anyone, and you will be accepted if it
looks as if you work in the field or a related field, and often even if you don't.

For example, I imagine any member of the VDE (the German electrotechnical engineering organisation)
would be enabled to register an interest in any standards area for which DKE is responsible. Whereas
I note that the British equivalent, the IET, is not a supporting organisation of DS/1, so you don't
get to comment on the 60812 CD merely through being an IET member with an interest. (Incidentally, I
find it very odd that the IET is not. I think we should fix that. HAZOP and FMEA and FTA and ETA,
let alone Markov modelling and so on are important to electrotechnical engineers!)

IEC CDs, NWIPs and so on are distributed, and comments are requested - commentary is expected to be
solicited by those who have registered an interest. So for example I gave a recent CD of another
standard to a particularly talented student for commentary; he sent in extensive (and worthwhile)
comments, which were forwarded to the IEC MT (though had not been addressed by the time I resigned
my MT membership, which was at the time that the subsequent CD was being distributed - one
procedural anomaly amongst many, which is why I resigned) and he was invited to participate in the
subsequent K 132 meeting. Comments are sent direct to the named German Representative on the IEC
Technical Committee (each participating country has just one Representative on a TC; this is a
different role from that of participating in a specific Project Team or Maintenance Team, which is
nominally not tied to nationality).

So I regard the German regime as appropriately technically liberal in this regard; the British and
US ones much less so.

Then there is the matter of affordances/hindrances, to use the Nielsen/Norman terminology.

As I mentioned, the IEC (and by transitivity the national standards organisations) only accept
comments on their comments form. This is commendably targeted in that it requires not only a general
comment (usually expressing a criticism) but also a specific proposed change, and has a column for
committee response. Formally, a committee response to each comment is required. The "committee" of
first instance here will often be the national committee, which will decide which comments to
forward to the IEC PT/MT. The affordance here is that comments get logged and responded to in a
traceable manner.

*But* the form is in a proprietary document-preparation format, and is not formatted for easy use on
the screens of computers which most of us use. I just got a brand-new Apple MacBook Pro, which
renders the file automatically using its own doc-prep SW Pages, and it is as usual all over the
place. I can of course deal with that and will. But I am involved in standardisation work, and many
people whom we hope will comment are not. Many of them are not going to take the half-hour to couple
of hours to configure their computers just to use a form. It's a hindrance.

What happens in Germany is that free-form comments are de facto accepted, and someone usually
volunteers to wedge them into the IEC format for further consideration, all very informal. As far as
I know, that doesn't seem to happen in Britain.

All in all, a reverse of the usual cultural stereotypes. :-)

John is concerned about access to standards and standards development for engineering students, as
am I (but as I mentioned, we have some sort of a solution, at least for IEC and ISO although not for
RTCA or EUROCAE, in Germany). I am very concerned about the technical quality of published standards
in our area, as is John, because it so often is poor. I think FDISs (the final draft standard)
should be subject to eminent-peer review. I think the EU project regime here is quite effective.
Named outside reviewers who engage face-to-face with representatives of the Project Team and who
have final say on whether Deliverables (documents and SW) are accepted. (It is more effective than
academic-style anonymous reviewing, because *everybody's* reputation is publicly on the line at
every step. For reviewers to miss things is just as consequential as for the PT to miss things.)
Resourcing this is an issue, but one imagines that the IEC could find a way from the kinds of
four-figure sums it charges for the likes of IEC 61508.

John's paper at SSS2014 is available at
http://scsc.org.uk/paper_126/protect_reg_01-Knight.pdf?pap=933 . I have a White Paper on it from a
year ago at http://www.rvs.uni-bielefeld.de/publications/WhitePapers/RVSsfssPrinciples.pdf

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