Page 252 - Design for Six Sigma a Roadmap for Product Development

222 Chapter Seven

of performance (goal rating/current rating). Hauser and Clausing (1988)
view this matrix as a perceptual map in trying to answer the following
question: How can we change the existing product or develop a new one
to reflect customer intent, given that the customer is more biased toward
certain features? The product of customer value, the targeted improve-
ment ratio for the raw (feature), and the sales point, which is a mea-
sure of how the raw feature affects sales, will provide a weighted
measure of the relative importance of this customer feature to be con-
sidered by the team.

7.4.6 HOWs correlation (the roof)
Each cell in the roof is a measure of the possible correlation of two
different HOWs. The use of this information improves the team’s
ability to develop a systems perspective for the various HOWs under
consideration.
Designing and manufacturing activities involve many trade-off deci-
sions, due mainly to the violation of design axioms (Chap. 8). The cor-
relation matrix is one of the more commonly used optional extensions
over the original QFD developed by Kobe engineers. Traditionally, the
major task of the correlation matrix is to make trade-off decisions by
identifying the qualitative correlations between the various HOWs.
This is a very important function in the QFD because HOWs are most
often coupled. For example, a matrix contains “quality” and “cost.” The
design engineer is looking to decrease cost, but any improvement in
this aspect will have a negative effect on the quality. This is called a
negative correlation and must be identified so that a trade-off can be
addressed. Trade-offs are usually accomplished by revising the long-
term objectives (HOW MUCHs). These revisions are called realistic
objectives. Using the negative correlation example discussed previously,
in order to resolve the conflict between cost and quality, a cost objective
would be changed to a realistic objective. In the correlation matrix, once
again, symbols are used for ease of reference to indicate the different
levels of correlation with the following scale:

Trade-Offs
Synergy + 1.0
Compromise – –1.0

In a coupled design scenario, both positive and negative interaction
may result. If one HOW directly supports another HOW, a positive cor-
relation is produced.
Correlations and coupling can be resolved only through conceptual
methods such as TRIZ (Chap. 9) and axiomatic design (Chap. 8). Other-
wise, a couple design results and trade-offs are inevitable, leading to
compromised customer satisfaction with design physics.