SHAININ Method: Edge Over Other DOE Techniques

’,
A.K. Verma A. Srividya A.V. Mannika?, V.A. Pankhawala K.J. Rathanraj3 ’,
’ Reliability Engineering Group, Indian Institute of Technology, Bombay, India
Automotive Research Association of India (ARAI), Pune, India
’BMS College of Engineering, Bangalore, India

Abstract- Shainin methods refer to a collection of

principles, which make up the framework of a continually 11. SHAININ VARIABLE SEARCH METHOD
evolving approach to quality. After the classical design of
experiments (DOE) and Tagnchi DOE, the third approach According to Bhote [1], the variable search method
is Shainin DOE, which is a collection of simple, but
powerful techniques invented or perfected hy Dorian can be classified into four stages. In the first stage,
Shaioin of the United States. In this paper, three cases of objective is to determine and to select the right variable
Taguchi experiments have been taken from literature and and the right levels for each variable for the experiment.
the above method has been tried to find out whether the Here, after selecting the factors for the experiment,
authors have got the positive results from their assign two levels to each factor-a best level, which is
experiment. If not, authors emphasize on the importance likely to contribute to a best response/output and a
of giving the check in the start of the experiment marginal level, indicative of a likely deviation from the
(screening experiment) with minimum number best level in day-to-day production with normal
experiments prior to the Taguchi approach. maintenance. Once the factors and levels are fixed, two
experiments are run, first with all factors at their best
Keywords- Shainin, Taguchi DOE levels, second with all factors at their marginal levels. If
there is a large difference between the response of the
all-best and the all-marginal combinations of factors, it
1. INTRODUCTION is an early indication that one has captured the right
factors. If the difference in response is small, the
The Shainin method is gaining popularity now
chances are that one has not captured (1) the right
because of the simple tools, which can give substantial
factors; or (2) the right levels of these factors; or that
good results at low cost and time. In fact, Motorola has
(3) the first dominant cause is being cancelled by a
a saying: “ Without Deming, the U S would not have
strong second dominant cause; or (4) the first dominant
had a quality philosophy; without Juran, it would not
factor is an interaction among an even number of
have had a quality direction; without Shainin, it would
factors. Thus necessary corrective action can be taken if
not have solved quality problems!” that sums up the
contributions of America’s three greatest quality gurus the difference in response is small. If the difference
between the responses is large, for confirmation of the
[I]. Unfortunately, like Demings approach and Taguchi
response values, two all-best levels of all factors and
methods, Shainin techniques have not received the wide
two more all-marginal levels of all the factors
publicity and use they deserve.
(randomized to avoid bias) can be conducted. Then Did
In this paper, authors would like to emphasize on
ratio is found, which must be greater than or equal to a
the screening test or the pilot test, which is being
minimum of 1.25. (D is the difference between the
carried out at the start of the experimentation. This will
median values of the best and the marginal responses
give an initial check to the parameter selection to level
and d is the average of two differences (or ranges)
selection. Shainin technique gives a tool to check the
within the all-best responses and the all-marginal
screening test results so that one can think of going
responses). If Did ratio value is more than the said
ahead with the full experimentation. This is a limitation
value, experiment can be moved to second stage. This
of both classical and Taguchi DOE where experiments
also means that right factors have been captured.
are to be carried out fully before one could realize that
Subsequent stages are similar to other DOE techniques.
experimental result are of use or not. Thus in this paper,
three cases of Taguchi experiments have been taken
from literature [2,3,4] and the above method has been
111. CASE STUDIES
tried to find out whether the authors have got the
positive results from their experiment. If not, authors
A. Case study 1:
emphasize on the importance of giving the check in the
According to Srivastava [2], for the optimization of
start of the experiment (screening experiment) with
resistance spot-weld process, three main parameters
minimum number experiments prior to the Taguchi
selected were % Heat, weld time and hold time and
approach.

1110

0-7803-8519-5/04/$20.00 @ 2004 I E E E

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY BOMBAY. Downloaded on February 9, 2009 at 00:46 from IEEE Xplore. Restrictions apply.
 One can observe that there is a large difference within
the replications than the Run 1 and Run 8. There is no
substantial success achieved out of this experiment in
identifying the influence of variations in weld
Run No. Mis Alpha Company Parameters parameters. The Shainin technique uses the
%Heat Weld Hold fundamentals of the Analysis of variance that ratio of
Time Time the difference between the group variability and
variability within must be significant otherwise the
experiment will not achieve the intended result.

B. Case study 2:
Kusiak and Feng [3], considered the tolerance design
problem for the machining dimensional chain and each
dimension includes a nominal value and the tolerance.
Assuming that the nominal values have been selected,
tolerances are to be allocated so that the manufacturing
cost is minimized. In Tolerance synthesis problem
Results of which are depicted in the Table 2 [ref2 pp 88 (TSP), the unit of tolerance is p m, and there is no unit
Table 4A], which wnsists of 20 repetition of each run. for cost as it is a relative value. The robust TSP is to
Observing the below table and finding the difference assign a tolerance stackup so that the tolerance stackup
among the run 1 and run 8 which are the Marginal is not greater than an upper limit (20 p m) and the total
levels and the Best levels of the orthogonal array. cost manufacturing does not exceed a preset limit.
All best level - Run 1: median value = 6.20 From the Table 3 and applying the screening test of
All marginal level -Run 8: median value = 6.185 Shainin method
Therefore, D = 6.20 - 6.185 = 0.015 All hest level - Level 0: median value = -1.57
d = Average lack of repeatability in each assembly All marginal level -Level 1: median value = -0.41
= ((6.28-6.13)+(6.320-5.987))12 = 0.2415
Did = 0.01510.2415 = 0.062, which is less than 1.25:1.
This implies that the other experiments between Run 2
to Run 7 were not required to he conducted.

International Engineering Management Conference 2004 1111

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY BOMBAY. Downloaded on February 9, 2009 at 00:46 from IEEE Xplore. Restrictions apply.
 TABLE 3

-
So D = -1.57 (-0.41) = 1.16 Did = 17.47i1.22 = 14.31, which is greater than 1.25:l
d = Average lack of repeatability in each assembly So this experiment is a success.
= (9.29 + 8.12)lZ = 8.71
Did = 1.1618.71 = 0.13, which is less than 1.25:l. So
this experiment might have not yielded significant IV. CONCLUSION
result.
One can observe that there is a large difference within In this paper, three case studies have been discussed
the replications than the level 0 and levell. There might to show that how screening test of Shainin method can
be an improvement in the form of cost reduction and be applied on the Taguchi’s orthogonal array to assess
tolerance. But one can say that it is not substantial. the adequacy of the experiment. It is observed that case
1 and case 2 show Did ratio less than 1.25 and therefore
C. Case study 3: it can be concluded that the experimentation result be
Table 4 below from Lin [4] that is a four factor two successful in identifying the influence of variations in
level experiments, which discusses about Taguchi’s parameters. In fact, it goes to assert adequacy of
concept of orthogonal arrays. Looking at the below tolerance levels (i.e. the tolerance variations are not
orthogonal array one can make out the Run 1 which has influencing the quality). However analysis of these two
all level at -1 and Run 8 which has all level at 1 can be case studies do emphasis that the additional
considered All-best and All-marginal level. Applying experiments conducted were not really required and
the screening test of Shainin method to below Table 4: could have been eliminated using Shainin variable
All hest level -Level -1: Average value = 32.27 search method at screening stage only. While case 3
(As min. three values are not available to consider the shows higher ratio of Dld and hence it can be concluded
median value) that the experimentation might give significant result.
All marginal level -Level 1: Average value = 14.8 Thus the Shainin variable search method can be
-
S O D= 32.27 14.8 = 17.47 effectively used to find and fix the few important
d = Average lack of repeatability in each assembly facton as well as its levels by conducting minimum
= (1.64 + 0.8)/2 = 1.22 number of experiments in the screening test.

TABLE 4

1112 International Engineering Management Conference2004

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY BOMBAY. Downloaded on February 9, 2009 at 00:46 from IEEE Xplore. Restrictions apply.
 REFERENCES

[I] Kelii R. Bhote, Adi K. Bhote, “World Class

Quality: using Design of Experiments to make it
happen”, 2”‘ edition, American Managemenr
Association, New York, 2000.
[2] Srivastava, Bipin B., “ Reliability studies on
resistance spot weld process parameters in
automobile components” Ph.D thesis, UT, Bombay,
Mumbai, 2000.
[3] Kusiak A., Chang-Xue Feng, “ Robust Tolerance
Design for Quality” Journal of Engineering for
Industry, Transactions of the ASME, Vo1.118, Feb
1996, pp 166-169.
[4] Dennis K. J . Lin, ‘‘ Making full use of Taguchi’s
orthogonal arrays” Qualify nnd Reliabiliry
Engineering International, Vol.10, 1994, pp 117-
121.

Internotional Engineering Management Conference 2004 1113

Authorized licensed use limited to: INDIAN INSTITUTE OF TECHNOLOGY BOMBAY. Downloaded on February 9, 2009 at 00:46 from IEEE Xplore. Restrictions apply.