MSU Barley Quality Lab Repeatability

August 2017
FG M M CG M M M M M M M M M M M M M M Color M M M M M M M M
Notes: Date Sample Description Measure # Extract
i
n
a
x Extract
i
n
a
x
Turbidity i
n
a
x
B-Glucan i
n
a
x
FAN i
n
a
x
Malt Protein i
n
a
x
S. Protein i
n
a
x
Moisture i
n
a
x (Gallery)
i
n
a
x
AA i
n
a
x
DP i
n
a
x
Friability i
n
a
x
This template is designed to assist you in starting Statistical Process Control 1/1/2017 Example 1 1 80.7 79 82 78.4 76 83 10 1 11 150 63 179 145 119 175 11.1 10 12 4 4 5 6.7 4 8 2.5 2 3 55 39 66 135 92 152 81.9 73 97
(a way to track and determine if you are statistically within your goal specs Example 2 2 80.6 79 82 78.3 76 83 8 1 11 135 63 179 140 119 175 10.6 10 12 4.2 4 5 6 4 8 2.6 2 3 60 39 66 120 92 152 90.1 73 97
for a malt, your system, etc.) This can help you monitor and detect problems, 1/8/2018 Example 3 3 78.7 79 82 76.1 76 83 6 1 11 140 63 179 160 119 175 11.2 10 12 3.9 4 5 6.6 4 8 2.8 2 3 62 39 66 112 92 152 76.6 73 97
ie. help you decide when you are seing normal variation of a process vs. Example 4 4 80.4 79 82 79.8 76 83 8 1 11 148 63 179 144 119 175 12.0 10 12 3.9 4 5 5.69 4 8 2.9 2 3 45 39 66 113 92 152 81.0 73 97
when you must act to adjust something. 1/15/2018 Example 5 5 80.4 79 82 79.9 76 83 4 1 11 131 63 179 142 119 175 11.3 10 12 4 4 5 4.92 4 8 2.3 2 3 49 39 66 119 92 152 86.0 73 97
Example 6 6 79.4 79 82 79.3 76 83 3 1 11 84 63 179 170 119 175 11.7 10 12 4.7 4 5 5.75 4 8 2.3 2 3 55 39 66 136 92 152 93.0 73 97
Tests included here are typical of what a Malster might track for a particular 1/22/2018 Example 7 7 81.3 79 82 81.3 76 83 4 1 11 105 63 179 124 119 175 11.0 10 12 3.9 4 5 4.62 4 8 2.4 2 3 43.0 39 66 98 92 152 90.5 73 97
malt - you can add and delete tests to track as you see fit. Example 8 8 80.9 79 82 81.4 76 83 6 1 11 75 63 179 155 119 175 10.2 10 12 4.3 4 5 7.29 4 8 1.7 2 3 49.0 39 66 143 92 152 81.0 73 97
9 79 82 76 83 1 11 63 179 119 175 10 12 4 5 4 8 2 3 39 66 92 152 73 97
Hints: 10 79 82 76 83 1 11 63 179 119 175 10 12 4 5 4 8 2 3 39 66 92 152 73 97
*For a good explanation of this type of tracking visit the Craft Maltster's 11 79 82 76 83 1 11 63 179 119 175 10 12 4 5 4 8 2 3 39 66 92 152 73 97
Guild members only section and view the "Building a Malthouse QC 12 79 82 76 83 1 11 63 179 119 175 10 12 4 5 4 8 2 3 39 66 92 152 73 97
Program" where Aaron MacLeod discusses basic statistics in the second half: 13 79 82 76 83 1 11 63 179 119 175 10 12 4 5 4 8 2 3 39 66 92 152 73 97
14 79 82 76 83 1 11 63 179 119 175 10 12 4 5 4 8 2 3 39 66 92 152 73 97
15 79 82 76 83 1 11 63 179 119 175 10 12 4 5 4 8 2 3 39 66 92 152 73 97
[Link] Mean Mean 80 79 6.13 121 147 11.1 4.1 5.9 2 52 122 85
macleod-hartwick-college-christian-stanley-valley-malt/ SD SD 0.84 1.73 2.42 29.17 14.02 0.57 0.27 0.90 0.38 6.86 14.85 5.77
Min Allowance (-2* SD) 78.61 75.85 1.29 62.71 119.41 9.99 3.56 4.14 1.67 38.53 92.28 73.46
*In order for this template to work properly it needs continual care to ensure Max Allowance (+2*SD) 81.98 82.79 10.96 179.39 175.49 12.27 4.66 7.75 3.20 65.97 151.69 96.56
that new data is being included, the graphs are showing new data and/or
encompassing the data range, the means/SDs are calculating the correct FG Extract CG Extract Turbidity A-Amylase
data etc. - if you are not familiar with Excel functions it is highly recomended 85 70
14
to do some online tutorials through sites like [Link] or by google 83
84
12
searching Excel video tutorials. 82
10
60
81 80 50
8
About this template: 79 78 6 40
The green rows below the data are set to automatically calculate your mean 77 76
4
2
30
and Standard Deviation for the values of each test. The graphs below will 75
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
74 0 20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
plot your points. The upper and lower limit lines of the graph are set to 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

automatically reflect values that are 2x the SD above and below your mean
(these adjust with each measurment added) - in a stable process you would B-Glucan FAN Protein D-Power
only expect to see values outside of these limits 1 in 20 measurments. 195 200 6.0 200
175 5.5 180
180 5.0
Instructions: 155
160 4.5
160
135 140
The file starts loaded with example data so that you can visualize the goal - 115
4.0
120
140
when you are ready, delete the examples and begin filling with your own 95
3.5
3.0 100
data. 75
120
2.5 80
55 100 2.0 60
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Step 1: Start with a process that you believe is 'in control', make
measurements at an appropriate frequency - enter your values in the Moisture Color Friability Malt Protein
template. 8 measures is a reasonable set to start evaluation of a process 10 100 13
with. 9
4
95
Step 2: Consider your mean and SD. Is your process stable, ie centering 8 3
90
12

around a mean and within the 2x SD limits? If not in what ways could you 7
2 85
11

tighten the process to get more consistent results? 6

80
10
Step 3: Once you have a stable process, compare new measures over time. 5
4
1
75 9
This chart can be used to determine the effectiveness of a process change 3 0 70 8
you put in place, or can alert you if changes occurs that you are unaware of. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Step 4: As time goes on and you become comfortable with your process you
can adjust the values included in the mean and SD calculation to disclude
older data points to best reflect your current practices.
You could make separate sheets for different malts, even add a sheet to track
parameters of your system like temp or grain moisture at steep out etc
When to act - when to wait:
In a stable process you should have an equal number of values falling above
and below your mean - this is normal variation that does not require action.
If you begin to see the values trending above or below the mean it indicates
there may need to be adjustments made. If you consistently begin to see
values outside the 2x SD range (rememer 1 in 20 values is normal) action is
required to bring the process back into stability.