Page 298 - Six Sigma Advanced Tools for Black Belts and Master Black Belts

OTE/SPH
OTE/SPH
Char Count= 0
August 31, 2006
3:6
JWBK119-18
Cultural Insights 283
18.6.1 Originality of statistical ideas
Many adverse comments have been made as regards the originality, or the absence of
it, of the experimental design principles at the core of Taguchi methods. 15 To set the
record straight, there is no evidence to suggest that Genichi Taguchi created the label
‘Taguchi methods’ for himself. In Japan, Taguchi’s contributions to applied design of
experiments are well known, but the subject is generally recognized by its generic
name, not by ‘Taguchi methods’. As for fundamental contributions, it is fair to state
that the Japanese have not been known as great originators of technology anyway;
most patents in technology breakthroughs in the past have their origins in the USA
and Europe.
However, in modern times the Japanese have been exceedingly successful in me-
thodically adapting, transforming, packaging and marketing acquired technology.
This has been amply demonstrated by their success with consumer goods, for exam-
ple electronic products and automobiles. In the field of quality control, the West has
already witnessed the acquisition, simplification, and promotion of statistical tools
in Japan, 46 and the subsequent exporting and marketing of such tools back to their
very places of origin. The emergence of Taguchi methods is essentially only the latest
exemplification of this familiar cycle.
Although it is academically correct to point out that Taguchi’s orthogonal arrays and
fractional factorial experimentation techniques have existed for decades, and ideas
such as effects of inaccuracies in parameter setting (error transmission) are not new
and could be more rigorously studied, 47,48 it may be worthwhile to ponder on the prac-
tical ramifications of Taguchi methods: stripping experimental design methodologies
of their mathematical cloaks, casting them in standard formats, and putting them
within the reach of rank-and-file technical personnel striving to meet quality and reli-
ability objectives in design and manufacturing. Experimental design can now become
an everyday occurrence, not something to be seen on special occasions. This by itself
is a novel strategy for pervasiveness in the realization of the power of experimental
design in industry, in much the same way as quality should ultimately be pervasive in
society.

18.6.2 Cookbook instructions for users
One of the well-known Japanese cultural traits is the propensity for uniformity and
doing things by standard examples. Foreign visitors to Japan are invariably struck by
the omnipresence of realistic wax models (mihon) of food dishes exhibited in show-
cases outside restaurants. At most establishments only what is on display is offered,
and a customer is expected to order from the standard fare with no possibility of vari-
ations -- and within minutes he gets exactly what the model shows. Virtually all eating
places also offer set meals (teishoku), saving the customer the trouble of making de-
cisions and creating combinations of orders, and the cooks some preparation time as
well. It would not be trivial at all to draw the parallel between the mind-sets governing
the way restaurant dishes are offered, chosen and consumed, and the way Taguchi’s
parameter design techniques are presented (through standard orthogonal arrays),
adopted (by means of a collection of artfully crafted linear graphs), and executed
(step-by-step marginal analysis, graphical presentations, and analysis of variance).