You are on page 1of 6

XmR Control Charts - Vendor Invoices Not Entered

Cost Data Integrity Project


Data Entry Cell >
1. Sample Information
Take a representative sample from the core projects that are part of this process improvement project and
plot the number of timesheets that are not entered over a continuous period of time. A sample size of at
least 20 is recommended in order to calculate control limits.

2. Purpose of Control Chart


Need to understand overall behavior within the process over time. How stable is the process and what effect
does a change have on the average value?

3. Important Elements

X-Bar: Centerline on the X-Bar Control Chart


R-Bar: Centerline on the R (Range) Chart
UCL: Upper Control Limit determined by multiplying a constant of 2.66 x R-Mean and adding to the Mean Value
LCL: Lower Control Limit determined by multiplying a constant of 2.66 x R-Mean and subtracting from the Mean Value
R-Mean: Average value for range values
Upper Range Limit: Upper Control Limit for Range Chart, calculated by multiplying a constant 3.27 x R-Mean

Note: The data selected for charting is often skewed and not very normal. One way to compensate for
this problem is to use the Median method as opposed to the Mean method when calculating your XmR
Chart. To keep it simple, we will stick with the Mean method despite skewness in the data.

4. How to Use the Charts


X-bar is the average or mean - your values should fluctuate around this value. There are rules that most people
follow to determine if the process is "out of control" such as:

One or more control points is outside (+ or -) 3 standard deviations from the mean
Two of three control points is outside (+ or -) 2 standard deviations from the mean
Four of five successive control points is outside (+ or -) 1 standard deviation from the mean
Six or more successive control points clearly shows an up or down trend from the mean

Also, you should make sure the process is stable by looking at the Range Chart - is this chart out of control? If yes,
then the Upper and Lower Control Limits that have been calculated are not reliable.

5. Collect and Calculate the Data for the Control Chart


Lower Upper
Test Number Upper Control Control
Period of Defects Range Mean R-Mean R-Limit Limit Limit
1 1 n/a 1.55 0.74 2.41 -0.41 3.51
2 2 1 1.55 0.74 2.41 -0.41 3.51
3 2 0 1.55 0.74 2.41 -0.41 3.51
4 0 2 1.55 0.74 2.41 -0.41 3.51
5 2 2 1.55 0.74 2.41 -0.41 3.51
6 2 0 1.55 0.74 2.41 -0.41 3.51
7 1 1 1.55 0.74 2.41 -0.41 3.51
8 2 1 1.55 0.74 2.41 -0.41 3.51
9 2 0 1.55 0.74 2.41 -0.41 3.51
10 2 0 1.55 0.74 2.41 -0.41 3.51
11 2 0 1.55 0.74 2.41 -0.41 3.51
12 1 1 1.55 0.74 2.41 -0.41 3.51
13 1 0 1.55 0.74 2.41 -0.41 3.51
14 2 1 1.55 0.74 2.41 -0.41 3.51
15 0 2 1.55 0.74 2.41 -0.41 3.51
16 2 2 1.55 0.74 2.41 -0.41 3.51
17 2 0 1.55 0.74 2.41 -0.41 3.51
18 2 0 1.55 0.74 2.41 -0.41 3.51
19 2 0 1.55 0.74 2.41 -0.41 3.51
20 1 1 1.55 0.74 2.41 -0.41 3.51
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Count > 20
ment project and
mple size of at

ss and what effect

ng to the Mean Value


acting from the Mean Value

ant 3.27 x R-Mean

way to compensate for


n calculating your XmR

ules that most people

art out of control? If yes,


XmR Control Charts - Vendor Invoices Not Entered
Cost Data Integrity Project

Range Chart

Range Chart
3.00

2.50

2.00
Range Values

1.50

1.00

0.50

0.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Test Periods

X-Bar Chart

X-Chart
4.00
3.50
3.00
2.50
2.00
Defects

1.50
1.00
0.50
0.00
-0.50
-1.00
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Test Periods
17 18 19 20

17 18 19 20