[SystemSafety] FMEA draft international standard

[Peter Bernard Ladkin]

Many people here know I am involved in national and international standardisation activities.

Standards work best when there is a need to agree to get something done. People supplying
electricity and other people supplying electrical devices need to agree on what type of supply is
delivered and expected. In each land, plugs on your XYZ kettle fit sockets in the wall and get what
current they want at the voltage and phase in which they want it. Maybe you need a cheap "adapter"
for another country, for the same hairdryer. Compare with the plethora of chargers for mobile phones
and realise what an achievement this is!

In the last two years, I have learnt just how messy and sometimes downright nasty the process can
be. I have been deliberately humiliated at both international and national level. Politics, with a
big "P" (if any engineers ever tell you they don't do "politics", don't believe it). Still, it's
only words and I can walk out whenever I want, and have done on a couple of occasions.

One of the problems is that we, systems people, don't actually make things such as plugs and sockets
which need to fit each other in order to operate. We standardise, not things like plugs, but ways of
doing things, like Fault Trees. As everyone who has tried to play football or play the piano knows,
there's more than one way to do most things.

There is a Committee Draft (CD) of a revised standard IEC 60812 on FMEA, Failure Modes and Effects
Analysis, now circulating for comments.

We (my tech-transfer company Causalis) have recently completed part of an FMEA for a client on what
looks at first sight like a piece of relatively simple kit. What is most important to our client is,
first, that the analysis is right, and, second, that it is thorough.

We have ways to do that. However, the CD of IEC 60812 has no word on methods to check whether an
FMEA is right. Nor, indeed, on methods to check whether it is thorough (see Footnote). How can that
be? (In this case, I actually know first hand how that can be.)

My colleague nominally on the committee (known as a Maintenance Team or MT) suggests with some
lytotes that the CD does not represent the current state of the art. He also suggests that the most
effective way of dealing with this issue is for as many engineers as possible to comment on the
draft. The MT is required to consider each and every comment and to respond.

But getting such a process underway is not trivial. First, people have to obtain a copy of the CD.
Then they have to download and engage with ("fight with" is maybe a more appropriate phrase) the IEC
comment form, which is easy to use if your computer screen has A3 landscape format and monstrously
frustrating if it doesn't. Then they have to send the comments to their National Committee
representative on the IEC 60812 MT, so they have to find out who that is from their national
standards organisation.

Let me try to ease that somewhat. If you wish to read and comment the CD of IEC 60812 FMEA, let me
know by email by this Friday, 18 July, and I'll see what we can figure out to get your comments into
the process chain. Time is short. I know comments in Germany have to be in by 8 August.

PBL

Footnote: there is of course a technical term for "right", as well as one for "thorough". Namely,
correctness, resp. completeness. Next time someone tries to tell you "completeness is impossible",
think about what they are really saying!

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