Quality in Design - Medical Device Design: Innovation from Concept to Market - page 133

as we would like to save as many as possible. If we needed to reduce something, say power
usage, the arrow would point down. This informs you of how the targets are to be met.
At last we come to the attic. This layout enables you to compare each technical specification
item against another. It works in the same way as a football chart showing who is playing
who over a season. In these boxes we add one of four symbols. Double arrows up mean
that the two are correlated and the change is strong and improving. A single arrow up
means the change is weak. No symbol means there is no link. A single downward arrow
means the change is not improving but is detrimental; a double downward arrow is strongly
detrimental. The roof is important as it is the first time you look at the interaction of your
decisions. The best way to exemplify this is with an automobile engine. If we have a target to
have low consumption (i.e., more miles per gallon), but also a target to have an engine with
more power output, you can see the two are correlated but increasing the power output will
obviously decrease the miles per gallon. Hence it forces you to look at ways of turning the
arrows upward or at least making them blank.
As with previous sections this is only a taste of HoQ. If you want to find out more about
this subject area then you can ind a liberal amount of topics on the subject under the title
QFD (Quality Function Deployment); there is also a liberal amount of good information
on the web.
7.5 Failure Mode and Effect Analysis (FMEA)
This is one of the most valuable tools a designer can have in their toolbox. It is, literally,
the place to imagine all the nasty things that can go wrong with your design and then to
make them unlikely. Some people, incorrectly, think of it as risk analysis…we are not
assessing risk (that comes later in this topic). We are assessing how the device will fail,
what will cause it, what the effect will be, and whether the effect is detrimental or not;
and then using this analysis to improve our design (at the design stage!). It should not be
underestimated that we are talking about foresight: a poor design that leads to a product
recall is all about hindsight…a good designer never relies on hindsight! From now on do
not let me hear you utter the phrase “ Well in hindsight we…. ”; a good FMEA will remove
this phrase from your phrasebook.
As an example let us consider the humble wheelchair. We may imagine many failure modes
but let us imagine a person sitting in the chair, the fabric base ripping open and the person
falling to the floor. The effects are obvious and traumatic; the failure is obvious - it ripped.
However the causes are manyfold. One cause may be that the person was too heavy for the
chair; another may be that the fabric was not strong enough to begin with; a third could be that
the material was already ripped from bad treatment by the previous user. All three can exist
but what do we do as a designer? We need to examine all the failures and try to design them
out. If we are unable to design them out then we need to build limits into our device's usage.