World Grain - March 2020 - 70

MILLING OPERATIONS

Influence of lab milling
on flour quality tests
Comparison of flour milled on Bühler Tandem mill
to control wheat milled at a constant setting
by Jeff Gwirtz

In my previous article, "More About
Laboratory Milling," in the November
2019 issue of World Grain, results of
varying Bühler Tandem mill settings
on the same wheat and impact on mill
EDODQFHDQGÀRXUDQDO\VLVZDVSUHVHQWed. The purpose and use of laboratory
milling to assist in evaluating potential
milling performance of wheat in commercial milling was presented, as was
the importance of controlled laboratory
PLOOLQJ IROORZLQJ VSHFL¿F SURFHGXUHV
In this article, we will investigate the reODWLRQVKLSEHWZHHQYDULRXVÀRXUTXDOLW\
WHVWLQJ PHWKRGV RQ WKH ÀRXU UHVXOWLQJ
from the Bühler Tandem compared to
ÀRXUPLOOHGIURPFRQWUROZKHDWPLOOHG
at a constant setting.
The values shown in Figure I, page
72, are developed on 22 replications of
a control wheat milled each day during
WKHFURSTXDOLW\VXUYH\RIKDUGUHG
winter wheat in the United States. The
average, standard deviation, minimum,
PD[LPXPUDQJHDQGFRH൶FLHQWRIYDULation are provided for the attributes idenWL¿HG,QWKHSUHYLRXVDUWLFOHRQO\WKHDYerage of three replications were provided
for the various mill setting. The purpose
RI WKLV ¿JXUH LV WR SURYLGH VRPH EDFNground relative to the variability of these
TXDOLW\PHDVXUHVZKHQDODUJHUQXPEHU
of replications are available using the
same wheat and milling procedure.
It should be clear that when a single
sample is milled, and results reported,
that the single observation of the mea70

VXUH FRXOG ¿W LQWR PDQ\ GLVWULEXWLRQV
about the true average value estimate for
the sample. Given a single observation,
there is no estimate of central tendency
(average) or variability (standard deviation) and one should feel hesitant about
drawing conclusions relative to these
data when comparing annual or regional
production test results.
7KHTXDOLW\GDWDLQWKLVUHSRUW)LJXUH
I) includes the standard deviation, maxiPXPDQGPLQLPXPYDOXHUDQJHGL൵HUence between the minimum and maxiPXPYDOXHFRH൶FLHQWRIYDULDWLRQDQG
the value needed to calculate the 95%
FRQ¿GHQFH LQWHUYDO IRU WKH PHDQ $V
noted, N=22, which is the number of
replications for the control wheat sample millings under the control setting.
7KHFRQ¿GHQFHLQWHUYDOLVFDOFXODWHGXVing the t statistic as N < 30.
7KHTXDOLW\GDWDLQWKHSUHYLRXVUHSRUW
provides only the average of three replications to provide a stronger estimate of
the mean or average value for the mill
setting. The standard deviation statistic
tells us about the variation of the individual observations and can be used to
HVWLPDWHDFRQ¿GHQFHLQWHUYDOIRUREVHUYDWLRQVDVZHOODVDFRQ¿GHQFHLQWHUYDO
for the estimate of the average value. It
VKRXOGEHQRWHGWKDWWKHFRQ¿GHQFHLQWHUYDOIRUWKHPHDQYDOXHLVVLJQL¿FDQWO\
smaller than for individual observations.
$VLQJOHREVHUYDWLRQRI\LHOGRUTXDOLW\
measures as provided in a crop report is
an observation and does not provide an

estimate of the mean value or variation
but may provide directional tendencies.
Statistics tell us that approximately
68%, 95% and 99.7% of the observations fall within plus or minus one,
two or three standard deviations of the
mean. As a side note, given a range of
values for multiple observations of an
attribute, one may divide by 4 to estimate the value of the standard deviation
for the observations. Fortunately, many
crop reports segregate samples by protein level with multiple milling samples
and analysis that report the average values in the crop report, which often are
compared to a multiple-year average.
The crop report is therefore providing
a good estimate of the average but is
VRPHZKDWODFNLQJZLWKUHVSHFWWRYDULability a miller might observe with individual lots of wheat arriving at the mill.
FIXED CONTROL SETTING COMPARISON
A comparison of product stream information from Bühler Tandem Mill setWLQJUXQVFRPSDUHGWR¿[HGFRQWUROVHWting is provided in Figure II, page 74.
The wheat used for the various settings
was a 50/50 control wheat blend made
RIFOHDQGU\XQWUHDWHGFHUWL¿HGVHHGRI
NQRZQYDULHWLHV$VLQGLFDWHGLQWKHSUHvious article, higher yield and ash levels
were accomplished by tightening both
the second roll pair (R1, B4, B4) in addition to the last roll pair (C4, C5, C6).
7KHVHFKDQJHVUHÀHFWWKHW\SHRIFKDQJes that might be observed by a laboraMarch 2020 / World Grain / www.World-Grain.com
http://www.World-Grain.com
World Grain - March 2020

Table of Contents for the Digital Edition of World Grain - March 2020
