INTRODUCTION TO SOFTWARE DESIGN FOR SIX SIGMA (DFSS) - Software Design for Six-Sigma: A Roadmap for Excellence - page 174

of a compatible systemic approach to find ideal solutions, the ignorance of the de-
signer, the pressure of the deadlines, and budget limitations. This can be attributed, in
part, to the fact that traditional quality methods can be characterized as after-the-fact
practices because they use lagging information for developmental activities such as
bench tests and field data. Unfortunately, this practice drives design toward endless
cycles of design-test-fix-retest, creating what broadly is known as the “fire fighting”
mode of the design process (i.e., the creation of design-hidden factories). Companies
who follow these practices usually suffer from high development costs, longer time-
to-market, lower quality levels, and marginal competitive edge. In addition, corrective
actions to improve the conceptual vulnerabilities via operational vulnerabilities im-
provement means are marginally effective if at all useful. Typically, these corrections
are costly and hard to implement as the software project progresses in the devel-
opment process. Therefore, implementing DFSS in the conceptual stage is a goal,
which can be achieved when systematic design methods are integrated with quality
concepts and methods upfront. Specifically, on the technical side, we developed an
approach to DFSS by borrowing from the following fundamental knowledge arenas:
process engineering, quality engineering, axiomatic design (Suh, 1990), and theories
of probability and statistics. At the same time, there are several venues in our DFSS
approach that enable transformation to a data-driven and customer-centric culture
such as concurrent design teams, deployment strategy, and plan.
In general, most current design methods are empirical in nature. They represent the
best thinking of the design community that, unfortunately, lacks the design scientific
base while relying on subjective judgment. When the company suffers in detrimental
behavior in customer satisfaction, judgment and experience may not be sufficient
to obtain an optimal Six Sigma solution, which is another motivation to devise a
software DFSS method to address such needs.
Attention starts shifting from improving the performance during the later stages
of the software design life cycle to the front-end stages where design development
takes place at a higher level of abstraction. This shift also is motivated by the fact
that the design decisions made during the early stages of the software design life
cycle have the largest impact on the total cost and quality of the system. It often is
claimed that up to 80% of the total cost is committed in the concept development stage
(Fredrikson, 1994). The research area of design currently is receiving increasing focus
to address industry efforts to shorten lead times, cut development and manufacturing
costs, lower total life-cycle cost, and improve the quality of the design entities in
the form of software products. It is the experience of the authors that at least 80%
of the design quality also is committed in the early stages as depicted in Figure 8.1
(El-Haik & Roy, 2005). The “potential” in the figure is defined as the difference
between the impact (influence) of the design activity at a certain design stage and the
total development cost up to that stage. The potential is positive but decreasing as
design progresses implying reduced design freedom over time. As financial resources
are committed (e.g., buying process equipment and facilities and hiring staff), the
potential starts changing sign, going from positive to negative. For the cunsumer, the
potential is negative and the cost overcomes the impact tremendously. At this stage,
design changes for corrective actions only can be achieved at a high cost, including