Page 59 - Six Sigma Advanced Tools for Black Belts and Master Black Belts

OTE/SPH
OTE/SPH
Char Count= 0
JWBK119-04
44 August 31, 2006 2:54 The Essence of Design for Six Sigma
One-dimensional elements (OD) are those where there is a strong correlation be-
tween satisfaction levels and the degree of presence or absence of a feature or
service.

The general rule of thumb is to fulfill all MB needs and to establish leadership position
in certain OD needs in the market. The ATT needs are good-to-have features and can
be used to differentiate a product from others. It is worth noting that some ATTs may
later evolve into future MB needs.
In essence, the QFD matrix can be completed by taking the following steps:

1. List the customers’ needs (i.e. the whats).
2. Classify and assign priority index to the customers’ needs using the Kano model.
3. List the technical attributes of the design to fulfill the needs (i.e. the hows).
4. Determine the relationships between the whats and the hows; strong relationships
are given a score of 9, moderate a score of 3, and weak a score of 1 (see Table 4.2).
5. Calculate scores for each how. This is done by multiplying the priority index column
with the relationship score under the respective how column. For example, under
‘resolution’ in Table 4.1, the raw weight is given by 5 × 9 + 4 × 3 + 4 × 3 + 5 × 3 =
84. The raw weights are then normalized to a scale of 1 to 5 by dividing each raw
weight by the maximum value.
6. Define the targets and limits on the hows (not shown).
7. Perform technology, measurement, and competitive analyses (not shown).
8. Determine the course of action.
9. Flow down requirements.

QFD is a very useful tool to facilitate and to document the thinking process in ful-
filling customer needs. It also provides a common ground for teamwork and triggers
subsequent actions in satisfying the needs. During the product development process,
information not only flows down to the lower-level matrix but also flows up in order
for certain design limits to be revised to achieve the desired trade-off in manufacturing
cost constraints and customer satisfaction. There has been some work on performing
such trade-off analysis; notably, Tang and Paoli proposed a spreadsheet model using
linear programming to obtain the Pareto optimal solutions. 3

4.2.2 Design

While the identify phase is mainly strategic in nature, the design phase requires major
engineering effort to be integrated with statistical skills, as this is where conceptual
design takes place and a feasible design that focused on customers’ requirements
emerges. Compared to the traditional DMAIC roadmap, the tools in the design phase
are a subset of the most powerful toolset from the measure, analyze and improve
phases of DMAIC where the transfer function between the key process input vari-
ables (KPIVs) and key process output variables (KPOVs) is developed for product
characterization rather than for process optimization purposes. Some advanced con-
cepts and techniques will be presented in the later chapters.
The key tasks in the design phase include the following:

1. Formulate conceptual designs and develop into possible design solutions.
2. Evaluate alternative designs and quantify their associated risk.