The Pursuit of Excellence
A Manager's Guide to Quality
|
Specifications Revisited
|
The figures which we have given throughout this example have been based on
statistical theory, and the application of theory to real life situations
is rarely straightforward. Before leaving this section, there are a few
points which should be stressed, some of which have already been mentioned
earlier:
- It would be impossible to take all the different variables
into account when presenting such an example. Consequently, we
have given a highly simplified model in which the only variable
that was considered was the radius of each of the components.
- The processes which produced the shafts and bearings were
initially considered to be uncontrolled, but stable. This is an
inescapable paradox. If we assume that the processes are out of
control, then there are assignable causes present, and it would
be possible to reduce the overall variation. However, in such a
situation, the process average and standard deviation would almost
certainly be fluctuating, and it would be impossible to calculate
probabilities in any consistent manner. Therefore, we had to use a
fixed standard deviation and, assume that it would be possible to
keep the processes centred on the target value.
- None of the figures given in this example should be taken as
precise predictions - they are likely estimates calculated from a
theoretical normal distribution. It is reasonable to expect that
the figures for a real process would be somewhere in the region of
those presented here, no more than that.
|