Page 287 - Six Sigma Advanced Tools for Black Belts and Master Black Belts

OTE/SPH
OTE/SPH
August 31, 2006
JWBK119-18
Taguchi Methods
272 3:6 Char Count= 0
Table 18.1 Advances in the application of statistics for quality improvement.
Stage I II III
Philosophy Quality by inspection Quality by monitoring Quality by design
Strategy Passive Defensive Pre-emptive
Approach Defect detection Defect prevention Defect elimination
Objective Damage control Status quo Optimization
Technique Sampling inspection Process control Robust design
Tools Sampling plans Indexes and charts Design of experiments
Basis Data based Information based Knowledge based
Location Check point On-line Off-line
Application Product Process Product and process
Mode Batch by batch Line by line Project by project
Currency 1940s 1960s 1980s

18.4 THE TAGUCHI APPROACH

Before Taguchi methods spread throughout the USA, design of experiments and the
associated techniques were treated as mathematical tools, more like an adjunct to an
engineer’s technical resources for the study of product and process characteristics.
In fact, they were taught and studied largely in this manner both in the universities
and in continuing education courses for engineers. Parallel to this, in the pre-Taguchi
period, was a parade of personalities in the USA preaching a variety of philosophies,
viewpoints, and strategies for quality -- Deming, Juran, Crosby and Feigenbaum be-
ing some of the better-known names. The target audience of such quality ‘gurus’ was
mostly senior management; engineers who were responsible for design and man-
ufacturing were largely left to their own devices -- mostly technological -- to raise
quality performance. Thus with reference to Figure 18.1, many organizations during
this period were operating with a disjointed rather than integrated front for qual-
ity improvement which, in the face of competition from superior Japanese quality
performance, constituted a backdrop to the ultimate cry in the summer of 1980: ‘If
Japan can, why can’t we?’ (this was also the title of a much publicized NBC television
documentary).
A partial answer to the question can actually be glimpsed from Taguchi methods.
Although ‘Taguchi methods’ is a term coined by Americans for the convenience of
Americans, it does serve as a convenient reference to the approach with which the
Japanese have been able to harmonize and integrate management, technology, and in-
formation capabilities to enhance quality, reliability, and profitability in manufactured
products. Some relevant background is as follows.
After the War, Genichi Taguchi used and promoted in Japan statistical techniques
for quality from an engineer’s perspective, rather than that of a statistician. The start-
ing point in Taguchi methods is his unconventional definition of quality. In contrast
to concepts such as ‘fitness for use’, ‘conformance to requirements’, and ‘customer
satisfaction’ circulated in the West, Taguchi’s ‘loss to society’ definition reflects two
common oriental values, namely aspiration to perfectionism and working for the col-
lective good. Thus, quality is not cast as a one-off performance orchestrated to suit an