Professional Documents
Culture Documents
Industrial Scrap Reduction Using Shainin
Industrial Scrap Reduction Using Shainin
Abstract— An alternative to the Classical and heralded SS tools, pre-control and variables search.
Taguchi experimental design is the lesser known There may be new developments not yet in the
but much simpler Shainin DOE approach public domain. Steiner and MacKay (2005) build
developed and perfected by Dorian Shainin on what we think are the best elements of SS.
(Bothe and Bothe, 2000), consultant and advisor In assessing the Shainin System, it is important to
to over 750 companies in America and Europe. differentiate between the overall approach that we
Shainin’s philosophy has been, “Don’t let the think is strong, and the specific analysis methods
engineers do the guessing; let the parts do the that are sometimes weak.
talking.” Shainin methods refer to a collection of Shainin DOE basically works at eliminating
principles, which make up the framework of a suspected process variables (Xs) mostly by using
continually evolving approach to quality. After seven different tools:
the classical design of experiments (DOE) and Multi-Vari Analysis
Taguchi DOE, the third approach is Shainin Component Search
DOE, which is a collection of simple, but Paired Comparison
powerful techniques invented or perfected by Component Search
Dorian Shainin of the United States. Full Factorials
B vs. C (Better vs. Current) Analysis
Index Terms—Shainin Techniques Taguchi Root Cause Measurements
DOE, Multivari analysis, B vs. C, Component
Search, Paired Comparison
II. PHASES OF SHAININ TECHNIQUE
I. INTRODUCTION 2.1. Multi-Vari analysis
Multi-Vari analysis is used to reduce a large
T he goal of this paper is to provide
number of suspected sources of variation to a
smaller family of variables containing the dominant
overview of Shainin concepts and how it is source of variation. A multi-Vari analysis is based
used for reducing the Industrial scraps. on the assumption that any process variation can
Bhote and Bhote (2000) and Bhote (1991, 1988) occur as a result of three factors, viz.
give the most complete (although not
comprehensive) treatment of SS. We agree with Positional or Part-to-Part Variation
reviewers (Nelson 1991, Moore 1993, Hockman Temporal or Time-to-Time Variation
1994 and Ziegel 2001) that these books make many Cyclical or Stream-to-Stream Variation
unsubstantiated, exaggerated claims. What is
worse, we believe that these books are a disservice Each type of variation is individually measured
to SS, since the hyperbole hides many of the using a run of approximately 3 to 5 units produced
genuinely useful ideas. A less technical and less consecutively at any given time. After a time lapse,
controversial reference that includes many case another run of 3 to 5 units are produced. This
studies is Traver (1995). process is repeated until 80 per cent of the out-of-
Overviews of SS have been published in control variation in the process is captured. A plot
conference proceedings; see Shainin (1992, 1992b, of these results indicates which one out of the three
1993, 1993b, 1995), Shainin and Shainin (1990) variations is maximum. Simple tools like the
and Shainin et al. (1997). Other review articles average and the range are then used to identify
include Logothetis (1990) and De Mast et al. which type of variation out of these three is the
(2000). Does, Roes and Trip (1999) cover many of highest.
the specific tools associated with the Shainin
System but not the overall strategy. Ledolter and
Swersey (1997a, 1997b) review two widely
Table 2.2: Month wise component rejections
Month Rejection Qty % Rejection
Sep “10 4536 10.272
Oct “10 5656 12.89
Nov “10 5209 11.796
Dec “10 8913 20.184
Jan “11 8981 20.338
Feb “11 7237 16.389
Mar “11 3626 8.211
Initial value 1 10
discussion of feasible alternatives. In our Greg Blue “Six Sigma for Managers” © McGraw
experience, there is also a rigidity with which the Hill 2002 0-07-138755
methodology is presented.
In many situations, other statistical tools, such as Montgomery D.C. (1996), Introduction to
regression, time series, and analysis of variance, Statistical Quality Control, 3rd edition, John Wiley
could be very useful, but are not employed because and Sons, New York.
they are not formally part of the SS tool bag.
Peter.S.Pandey Robert P, Neumann Roland R
Cavanagh. “The Six Sigma Way” 2nd edition
REFERENCES McGraw-Hill 2000 0-07-135806-4