Lecture Week 7:

DEVELOPING VARIABLE
CONTROL CHARTS
 VARIATION

Variation occurs in
anything we do including
manufacturing processes.

A process is statistically in
control if its variation is due
to random causes only.
 Variable control charts
B. Objectives of using variable control
charts

1. To establish specification limits.

2. Monitor the process to confirm products
being made conform to specifications.
3. To improve quality whenever
possible.
4. To prevent high costs of holding or
rejecting lots.
5. To lower final product inspection
Variable control charts (cont.)
C. Formulas and Factors needed to develop variable
control charts (X, R) Control Limits:

For X : Upper Control Limit (UCLx) = X + A2R

Lower Control Limit (LCLx) = X - A2 R
Central Line (CLx) = X
where,
 A2- values are derived and decrease as the
number of product units in each subsample (Ns)
increases.
 X is the calculated grand average
 R is the calculated range average
Variable control charts (cont.)

For R:
Upper Control Limit (UCLR) = D4R
Lower Control Limit (LCLR) = D3R
Central Line (CLR) = R

where,
 D4 and D3 can be found on factor chart
below.
 R is the calculated range average.
Table of derived factors: Factors for Average-Range
Sample Size A2 D3 D4
(Ns)
2 1.880 0 3.267
3 1.023 0 2.575
4 0.729 0 2.282
5 0.577 0 2.115
6 0.483 0 2.004
7 0.419 0.076 1.924
8 0.373 0.136 1.864
9 0.337 0.184 1.816
10 0.308 0.223 1.777
11 0.286 0.256 1.744
12 0.266 0.284 1.716
13 0.249 0.308 1.692
14 0.235 0.329 1.671
15 0.223 0.348 1.652
16 0.212 0.346 1.636
17 0.203 0.379 1.621
18 0.194 0.392 1.608
19 0.187 0.404 1.596
20 0.180 0.414 1.586
21 0.173 0.425 1.575
22 0.167 0.434 1.566
23 0.162 0.443 1.557
24 0.157 0.452 1.548
25 0.153 0.459 1.541
 Variable control charts (cont.)

E. Step-by-step construction of the

control charts.

1. - Collect data: it is preferable to have

15-20 sub-sample with 4-6 product
units each.
- Sample Data & Calculations
 Sub-
sample
Product Units Calculations
(day)
1 2 3 4 5 Average Range
(X) (R)
1 7 10 8 2 8 7 8
2 6 9 3 4 3 5 6
3 6 7 2 6 4 5 5
4 4 8 5 7 7 6.2 4
5 10 3 5 6 4 5.6 7
6 8 11 9 6 6 8 5
7 5 7 5 6 7 6 2
8 9 5 8 7 6 7 4
9 6 5 4 5 5 5 2
10 1 3 1 0 3 1.6 3
11 3 4 6 5 4 4.8 3
12 5 3 6 3 3 4 3
13 5 6 8 9 7 7 4
14 7 7 8 7 6 7 2
15 7 7 6 7 7 6.8 1
16 7 7 7 9 9 7.8 2
Total 93.8 61
X=5.86 R=3.81
Avg.
 Variable control charts (cont.)

From table
 A2 for (Ns=5) = 0.577

 D4 for (Ns=5) = 2.115

 D3 for (Ns=5) = 0
 Calculation for X/R charts

a. Calculate average (X) for each row :

X = X/Ns
ex: for sub-sample 1: 7 10 8 2 8
X = 35 / 5 = 7
b. Calculate grand average (X): X =
X/N sub-samples
ex: X = 93.8 / 16 = 5.86
 Calculation for X/R charts (cont.)

C. Calculate range (R) for each sub-sample:

R=largest x – smallest x in sub-sample:
ex: for subsample 1: 7 10 8 2 8
R = 10 – 8 = 2

D. Calculate average R :
R = R / N subsamples
ex: R = 61 / 16 = 3.81
 Calculation for X/R charts (cont.)
E. Calculate control limits using the formulas following and
table values for A2,D4 and D3 .

X chart: UCLx = X + A2R = 5.86 + 0.577 (3.81) = 8.06

LCLx = X - A2R = 5.86 - 0.577 (3.81) = 3.66

CLx = X = 5.86

R chart: UCLR = D4R = 2.11 x 3.81 = 8.05

LCLR = D3R = 0.00 x 3.81 = 0

CLR = 3.81
 Calculation for X/R charts (cont.)
F. Warning limits can be also be calculated:

 Upper 1  =CL + (UCL-CL)/3

 Upper 2  = CL+ 2(UCL-CL)/3
 Upper 3  = UCL
 Lower 1  = CL- (UCL-CL)/3
 Lower 2  = CL- 2(UCL-CL)/3
 Lower 3  = LCL
 Variation,σ Distribution
Probabilities
>+3
0.13
Upper Control Limit +3 2.1
+2 13.6
(UCL) +1 34.0
Central Line (CL) X
-1 34.0
-2 13.6
Lower Control Limit -3 2.1
(LCL) >-3 0.13
Plotting of Control Charts

• X chart:

a. Plot all warning limits.

b. Plot averages (X) for each sub-

samples vs. sub-samples number,
time, month etc.
 Plotting of Control Charts (cont.)

 R chart:

a. Plot all warning limits

b. Plot individual range (R) vs. sub-sample
number, time, month, etc.
 Plotted Charts
X – chart
 Plotted Charts (cont.)
R – chart