Page 167 - Design for Six Sigma a Roadmap for Product Development

Design for Six Sigma Project Algorithm 141

expectations and delights together with corporate and regulatory
wants. This understanding is then translated into functional require-
ments (FRs) with design specifications (tolerances), which then cas-
cade to all levels of design hierarchy. The power of first gaining
complete understanding of requirements and then translating them
into specifications is highlighted by Pugh (1991). This notion is also
the basis of the strategy commonly associated with quality function
deployment (QFD).

5.3.7 Map CTSs into functional
requirements (FRs) (DFSS algorithm step 2)
The first formal mapping, in a QFD format of customer requirements
to design characteristics was done in 1972 by the Kobe Shipyard of
Mitsubishi Heavy Industries. This start led to the evolutionary devel-
opment of the four phases of QFD. QFD phase 1 translates the customer
needs and expectations into the CTSs. Subsequently, the CTSs must be
converted into design actions. This conversion is completed by con-
structing QFD phase 2, a new house of quality, on which the WHATs
are the CTSs and their target values from Fig. 5.6 (phase 1, house of
quality). The HOWs and HOW MUCHs of each matrix are progres-
sively deployed as WHATs on the charts or matrices that represent the
next phase of the design development cycle. This conversion of HOWs
to WHATs is continued from design planning to production planning.
While we recognize the mapping conducted in each of the four
phases of QFD, we propose limiting the QFD exercise to only phases 1
and 2. We believe that the zigzagging method of axiomatic design is
more powerful when armed with design axioms and vulnerability
reduction techniques. Therefore, in the DFSS algorithm, we propose
the following mappings as depicted in Fig. 5.7:
■ Perform QFD phase 1 by mapping customer attributes to critical-to-
satisfaction (CTS) requirements (step 2 of the DFSS algorithm).
■ Perform QFD phase 2 by mapping CTS to functional requirements
(FRs) (step 2 of the DFSS algorithm).
■ Perform zigzag mapping of axiomatic design between the functional
requirements (FRs) and design parameters (DPs) (step 6 of the
DFSS algorithm).
■ Perform zigzag mapping of axiomatic design between the design
parameters (DPs) and the process variables (PVs) (step 6 of the
DFSS algorithm).

The first mapping begins by considering the high-level customer
attributes for the design. These are the true attributes, which define