The impact of variation on producibility is
a central issue in determining the ease of manufacturing a product. For
example, consider a certain critical product characteristic Y_{1}.
Y_{1} can be any characteristic such as the length of a mechanical
part, voltage in a circuit, or resistance of a material. Suppose that
Y_{1} exhibits a large amount of variation relative to its performance
standard. Under this condition, manufacturing ease is low when compared to a
situation when the variation in Y_{1} is small. Therefore, as variation
increases, producibility decreases because the probability of non-conformance
increases. In turn, this situation produces additional rework, scrap, cycle
time, and cost. As rework and scrap increase, direct labor also increases. The
ease with which a product can be manufactured is consequently diminished as
variation increases. It can therefore be stated that variation and producibility
are inversely related - but from where does such variation come?

Product variation emanates from the
underlying network of causal variables. This concept is better expressed as
Y=f(X_{1},..., X_{n}), where Y is some
product characteristic and (X_{1},...,X_{n}) describes all of the variables
in the cause system. The many variables within most cause systems can be
classified into three primary sources of causation:

The independent effects of the numerous design, process, and
material variables and their combinational interactions form the foundation of
product quality. When the vital few effects are controlled such that variation
is minimized, product quality is improved. And as quality is improved,
producibility is enhanced. The interplay between product quality and
producibility is therefore a pivotal issue.

Variation represents the crucial drawback to producibility and
must be numerically determined if the ease by which products are manufactured
can be studied and ultimately improved. In particular, variation must be
quantitatively assessed with respect to design margins, material/component
specifications, and process control/capability. Only then can the issue of
producibility be addressed in a practical and meaningful way.

Prediction is not possible if producibility is not studied before
production begins. Therefore, when the amount of variation is large relative to
the performance standards, the ability to predict the frequency of
non-conformance diminishes. As a logical consequence, manufacturing viability is
lost.

The accurate prediction of producibility is
highly dependent upon a measure of the interplay within and between the three
circles depicted in Figure 4-1. In addition, producibility optimization during the design phase
requires that engineers design for producibility so that product designs will
be relatively impervious - or robust - to natural, unavoidable sources of
process, component, and material variation. To accomplish this, there must be
quantitative knowledge of process, component, and material capabilities that
is continually updated and renewed. Product and process designs in accordance
with known production capabilities must be numerically optimized so that
products are designed in the light of that inevitable variation. Equally
important is that design viability be assessed before release for production.
The synergistic effect of designing for manufacturability and producibility
assessment reacts so that the whole is greater than the sum of the parts
(Figure
4-1).