Control Charts

Statistical Process Control (SPC)

• SPC was developed by Walter A. Shewhart in 1924

• Historically, SPC has been used to monitor & control output ‗Y‘

• In DMAIC, we apply SPC to control X‘s (remember ‗Y‘ is only monitored)

• However, sometimes applying SPC to ‗Y‘ can also be beneficial in detecting trends

• About SPC
– Aids visual monitoring & controlling
– Depends heavily on data collection

2
Foundation of SPC

• It forms data into patterns which can be statistically tested and, as a result, leads to
information about the behavior of process output / control variable characteristics

• It graphically represents output / control variable performance

• It detects assignable causes which affects the central tendency and/or variability of the
cause system

• It serves as a probability-based decision making tool

• It points out where action can be taken with known degrees of risk and confidence

3
SPC Tools

• SPC primarily uses ‗Control Charts‘

Upper Control Limit

Characteristi (UCL)

Lower Control Limit

cs

(LCL)

Time /
Number

4
Basics of Control Charts

• Control charts are useful for tracking process statistics over time and detecting the
presence of special causes

• A process statistic, such as a subgroup mean, individual observation, or weighted

statistic, is plotted versus sample number or time. A ―center line‖ is drawn at the
average of the statistic being plotted for the time being charted. Two other lines—the
upper and lower control limits—are drawn, by default, 3σ above and below the center
line

• A process is in control when most of the points fall within the bounds of the control
limits, and the points do not display any nonrandom patterns

5
Purpose of Control Limits

Special Cause
Variation UCL

Variation
Cause
Common

LCL

Special Cause
Variation

6
Purpose of Control Limits

0.135%
UCL = µ +
3σ

99.73%

LCL = µ -
0.135% 3σ
Out of control
point

Control Limits define a probabilistic

level of occurrence of an ‘out of
control’ point

7
Setting the Control Limits

• A standard control chart uses control limits at three standard deviations form the data
mean. The probability of an out-of-control point when the process has not changed is
only 0.27%

• If control limits are set at two standard deviations, it increases the chance of type I
error

• If control limits are set at four standard deviations, it increases the chance of a type II
error

• Control chart should keep in mind both type I & type II errors

8
• Types of errors:

• Type I errors
When a point falls outside the control limits even though no
special cause is operating. The result is a witch-hunt for
special causes and adjustment of things are done here and there.
The tampering usually distorts a stable process as well as wasting
time and energy.

• Type II errors
When you miss a special cause because the chart isn't
sensitive enough to detect it. In this case, you will go along unaware
that the problem exists and thus unable to root it out.

• 3-sigma limits are a good balance between two types of errors:

9
• 3-sigma balance the risk of error as data points will fall
inside 3-sigma limits 99.7% of the time when a process
is in control.

• This makes the witch hunts infrequent but still makes it

likely that unusual causes of variation will be detected.

10
Top Nine Indications of an Out of Control Process

9. A single point outside control limits

8. Two out of three successive points between 2σ and 3σ on the same side of
the centerline

7. Seven successive points on the same side of the centerline

6. Nine out of ten successive points on the same side of the

centerline
5. Twelve out of fourteen successive points on the same side of the
centerline
4. Consistent increase or decrease in
levels
3. Fourteen points alternating up and
down
2. Four out of five successful points beyond 1σ on the same side of the
centerline
1. Eight points in a row with none between ±
1σ

11
Attributes and variables

• Variables are quality characteristics such as

weight, length time, temperature, voltage etc
that is measured.

• Attributes are quality characteristics which

concentrate on defects for eg. A lamp either
working or defective.

12
Central Limit Theorem

If the population is normal or the sample size is

large, then means of random samples coming
from that population are normally distributed with
mean  and standard deviation /n .

Where
 --- is mean of population
n---- sample size
----- is the standard deviation of population

13
• Control chart for Attributes
• p chart(control chart for proportion defective)
• np chart (control chart for number defective)
• c chart (chart for number of defects)
• u chart (chart for number of defects with
changing area of opportunity for defect)
• Control chart for Variables
• X bar chart (for average level)
• R chart (for spread)
• Moving range control chart for individual
measurements.

14
Choosing An Appropriate Control
Chart
Continuous Data

Individual Data Subgroup

Points
Pulling one sample at fixed Takingsperiodic grouped
frequency data

Variability of Variability of
I & MR individual average
characteristics over characteristics over ¯X & R
{Individual & Moving
time Range Chart} time when sub-group
size
is less than 8 when
dataVariability of
are normally
average
distributed
characteristics over X&S
time when sub-group
size
is more than 8
data are not normally
distributed

15
Choosing An
Appropriate Control
Chart Discrete Data

Defectives Defects

Constant Varying Constant Varying

Sub- Sub- Sub- Sub-
group group group group
Size Size Size Size

NP P C U

# of % of Average # of
# of
units units defects per
defects
rejecte rejecte opportunity
d d
Binomial Distribution with Poisson Distribution with
(n, p) (λ)

16
 P-Chart
• P-Chart or Control Chart-for proportion defective is used to
monitor the proportion of defective item being turned out by
a process.
• Overall proportion defective for all samples

p = Total number of defective

Total number inspected
• Control Limits

UCL = p+3 p(1-p) / n

LCL = p-3 p(1-p) / n

• Sample size : n>= 20 for p close to 0.5
np>=5 if p departs substantially in either direction
17
Example: On a given day periodic sample of 100 chips every half an hour were taken. Data on Defective Items for 25 samples of Size
100 is given below

Sample Number Number of Defective Items Proportion of Defective items

1 10 0.10
2 5 0.05
3 10 0.10
4 13 0.13
5 14 0.14
6 9 0.09
7 8 0.08
8 9 0.09
9 7 0.07
10 14 0.14
11 10 0.10
12 7 0.07
13 8 0.08
14 10 0.10
15 12 0.12
16 10 0.10
17 12 0.12
18
Sample Number Number of Defective Items Proportion of Defective
items

18 9 0.09
19 6 0.06
20 5 0.05
21 10 0.10
22 16 0.16
23 13 0.13
24 5 0.05
25 12 0.12
Sum= 245 Avg = 0.098

p = 245/2500 = .0980 ---- overall proportion defective

UCL = 0.1872 n( sample size)= 100

LCL= 0.0088

19
When sample size varies , control limits vary from sample to sample.
20
 np-Chart
np-Chart or Control Chart-for number defective is used to
monitor the number of defective item being turned out by a
process.
• Central line of np chart= n * p = 100 * 0.098 = 9.8

• Control Limits

UCL = n p +3  n p (1-p) = 18.72

LCL = np-3  np(1-p) = 0.88

For np chart the sample size should remain constant from

sample to sample.
21
22
 C-Chart
C- charts are used for number of defects when area of
opportunity remain same. Average no. of defects

c = Total number of defects

Total number inspected
• Control Limits

UCL = c +3c

LCL = c -3 c

If value for LCL calculated is negative then LCL is set to zero

23
 Number of Defects in 25 Rolls of
Coated Paper
Roll No. 1 2 3 4 5 6 7 8 9 10 11 12 13
No. of defects 5 6 3 5 3 6 6 7 5 5 4 3 5

Roll No. 14 15 16 17 18 19 20 21 22 23 24 25

No. of defects 4 9 11 3 7 5 5 12 4 6 7 5

Total No. of defects 139

c = 139/25 = 5.56

UCL = 5.56+ 35.56= 12.63

LCL = 5.56- 35.56= -1.51 0

24
25
 U-Chart
U- charts are used for number of defects when area of
opportunity varies with sample.
Average no. of defects per unit area of opportunity

= Total number of defects observed

Total area inspected

UCLi = u +3  u/ai
LCLi= u – 3  u/ai

If value for LCL calculated is negative then LCL is set to zero

26
 Area and Number of Blemishes in 25
Countertops
Counter 1 2 3 4 5 6 7 8 9 10 11 12 13
top No.
Area (sq. ft.) 18 19 17 37 47 23 28 26 27 38 25 24 29

No.of 4 10 1 9 14 3 9 12 3 10 4 8 8
blemishes
Counter 14 15 16 17 18 19 20 21 22 23 24 25
top No.
Area (sq. ft.) 28 30 29 38 40 25 26 29 31 19 50 42

No.of 14 8 13 8 12 6 8 13 9 5 8 17
blemishes

Total area 745

Total No. of defects 216

= 216/745= 0.2899,
27
28
X-Chart
X charts are used for for controlling average level of
variables
Computation of control limits requires a factor A2 which
depends on the subgroup size(n) which is normally 4 or 5.
Value of A2 is given in a table.

X = overall average

UCL = X + A2R

LCL= X- A2 R

29
 R-Chart
R charts are used for for controlling spread of variables
Computation of control limits requires two factors D3 and D4
which depends on the subgroup size(n) which is normally 4
or 5. Values of D3 and D4 are given in a table.

R = average of ranges

UCL = D4R
LCL = D3R

30
 Control Chart Factors
Subgroup A2 D3 D4 d2 d3
Size

2 1.880 0.000 3.267 1.128 0.853 A2 =3/{d2n}

3 1.023 0.000 2.574 1.693 0.888

𝑹
4 0.729 0.000 2.282 2.059 0.880 𝝈= ¿
𝒅𝟐
¿
5 0.577 0.000 2.114 2.326 0.864
= population
6 0.483 0.000 2.004 2.534 0.848 standard
deviation
7 0.419 0.076 1.924 2.704 0.833

31
 Example: Tensile Strengths for 25 Samples of
Plastic Cord Produced by PLASTEX
Sample Sample Values (Tensile Sample Average Sample Range
Number Strength) (X-Bar) ®
1 75 78 85 81 79.75 10
2 83 84 76 80 80.75 8
3 81 82 79 74 79.00 8
4 77 81 89 79 81.50 12
5 79 76 82 78 78.75 6
6 83 77 85 85 82.50 8
7 87 79 83 75 81.00 12
8 80 85 76 76 79.25 9
9 86 79 82 80 81.75 7
10 74 82 80 84 80.00 10
11 80 74 75 81 77.50 7
12 76 81 79 78 78.50 5
32
Sample Sample Values (Tensile Sample Average Sample Range
Number Strength) (X-Bar) ®

13 77 86 77 84 81.00 9
14 83 78 72 86 79.75 14
15 79 85 75 78 79.25 10
16 81 84 82 81 82.00 3
17 80 84 81 81 81.50 4
18 80 77 83 77 79.25 6
19 78 75 74 85 78.00 11
20 79 78 79 80 79.00 2
21 88 77 81 72 79.50 16
22 79 74 82 80 78.75 8
23 84 79 84 81 82.00 5
25 79 85 82 85 82.75 6
Average
X= 80.02 R=8.04

33
 X-Chart

UCL = 80.02 + 0.729(8.04) =85.88

LCL =80.02 - 0.729(8.04) = 74.16
34
 R-Chart

UCL= 2.282(8.04)= 18.35

35
LCL= 0(8.04) = 0
Interpretation and Inferences
• Caution
• sample plots may fall outside the limit, even though no special causes
are present
• Reason being that desirable standards may not be consistent with the
process conditions
• It is easy to meet a desirable target value for process mean than it is
for process variability (Range)
• Interpretation
• Difficult and needs thorough knowledge about different process
parameters on quality characteristic
• When R-chart is brought to control, many special causes for the Xbar
chart are eliminated as well
• Xbar chart monitors the centring of the process and a jump indicates
process average has jumped

36
 Control Charts for Individuals
• These charts are used in monitoring of variables when one
value of the variables is naturally available.
• Chart for individual measurement is used to control
average level.
• A moving range chart is used to control the spread.
X= average of individual measurements
R= average of moving range
Control limits for individual measurement
UCL= X +2.66 R ; LCL = X-2.66 R
Control limits for moving range control chart
UCL = 3.267 R ; LCL =0
37
Twenty five Consecutive Weekly Accounts Receivable Balances
Week Accounts Receivable Balances (000) Moving Range
1 47.0
2 41.3 5.7
3 64.3 23.0
4 83.3 19.0
5 48.0 35.3
6 87.8 39.8
7 65.3 22.5
8 40.8 24.5
9 55.5 14.7
10 50.8 4.7
11 47.8 3.0
12 66.3 18.5
13 39.0 27.3
14 31.5 7.5
15 48.3 16.8
38
 Week Accounts Receivable Balances (000) Moving Range
16 47.1 1.2
17 63.0 15.9
18 51.3 11.7
19 72.6 21.3
20 58.9 13.7
21 33.5 25.4
22 42.3 8.8
23 62.8 20.5
24 87.6 24.8
25 42.3 45.3

X = 55.14 R = 18.79

The moving range chart should be developed before the control

chart for individual measurement. Range must be in control
before developing control chart for individual measurements. 39
 Moving Range Chart for monitoring
moving balances of accounts receivable

40
Individual measurement chart for monitoring
moving balances of accounts receivable

41
 X & S Control Chart
A packing organization monitoring the performance of a packing machine,
each container should weigh 35 lb, during the Measure phase, the project
team performed the process capability study and identified that the
process is not capable(less than one sigma). In Analyze phase collected 12
sets of container weights with a subgroup size of 4.

42
43
44
45
From both the X bar and S charts, it is clearly evident that
most of the values are out of control. Hence the process is
not stable

46
•To correctly interpret the X bar S chart, always examine the S
chart first.

•The X bar chart control limits are derived from the values of S
bar (average standard deviation). If the values are out of
control in the S chart, the X bar chart control limits are
inaccurate.

•If the points are out of control in the S chart, then stop the
process. Identify the special cause and address the issue.
Remove those subgroups from the calculations.

•Once the S chart is in control, then review the X bar chart.

Afterward, you should interpret the points against the control
limits.

•All the points are to be interpreted against the control limits

but not specification limits.
47
•If any point is out of control in X bar chat. Identify the special
Control chart patterns and corrective
action
• A ‘non-random identifiable’ pattern in the plot of a chart might provide reason to look for
special cause in a process

• There are about 15 typical patterns identified by Western Electric company and 9 of them
have been discussed here

48
Control chart patterns and corrective
action
•
Natural Patterns
• No identifiable arrangement of the plotted point exists
• No point fall outside the control limits
• Majority of the points are near the centre line and few points close to control limits
• Demonstrates the presence of stable system of common causes

49
Control chart patterns and corrective
action
• Sudden shifts in the level
• Can occur because of changes intentional or otherwise in process settings (temperature, depth of
cut etc.)
• New operators, new equipment, new vendors, new methods are the reasons for sudden shift

50
Control chart patterns and corrective
action – Contd..
• Gradual shifts in the level
• Occurs when a process parameter changes gradually over a period of time
• X-bar chart might exhibit such a shift because of incoming quality of raw materials that would have
changed with time
• Change in style of supervision, maintenance program etc.
• R-chart might exhibit such a shift because of new operator, decrease in worker skill due to fatigue
etc.

51
Control chart patterns and corrective
action – Contd..
• Trending pattern
• Differs from gradual shift in level, that trends do not stabilize or settle down
• Represents changes that steadily increase or decrease
• For X-bar chart, can be due to tool wear, deterioration of equipment, build up of debris on jigs and
fixtures, change in temperature etc.
• For R-chart, it may be due to improvement in operator skill due to on job training, decrease due to
fatigue etc.

52
 Control chart patterns and corrective
action – Contd..
• Cyclic patterns
• Characterized by repetitive periodic
behaviour in the system
• Cycles of low and high points will appear on
the control chart
• X-bar chart may exhibit because of rotation
of operators, periodic change of
temperature or humidity, seasonal variation
in incoming components
• R-chart may exhibit this pattern because of
operator fatigue and getting energized in
subsequent breaks, a difference between
shifts, periodic maintenance of equipments
etc.
• If samples are taken so infrequently, only
the high and low points will be represented
53
 Control chart patterns and corrective action
– Contd..
Freak pattern
• Wild patterns
• Can be classified as Bunches and Freaks
• Cluster of several observation that are different from
other points and special causes are associated with
these points

• Freaks
• are caused by external disturbances that influence one or
more samples
• They are points that are too small or large with respect to
control limits and fall outside the control limits and hence
easy to identify
• Care should be taken that no measurement or recording Bunch pattern
error is associated with that freak point
• Some special causes may be sudden, very short-lived
power failures, use of new tool for a brief test period etc.

• Bunches
• Cluster of several observation that are different from other
points
• Possible causes may be use of new vendor, use of a
different machine, use of new operator etc., for a short
time period.

54
Control chart patterns and corrective
action – Contd..
• Mixture patterns
• Effect of two or more population in
the sample
• Characterized by points that fall
near the control limits, with
absence of points near the centre
line
• Might be due to material from two
different vendors, different
production method, two or more
machine being represented

55
Control chart patterns and corrective
action – Contd..
• Stratification patterns
• Is also due to presence of two or more
population distribution
• Output is combined or mixed and
samples are selected from it
• Majority of the points fall close to
centre line, with very few points near
the control limits
• Can be misinterpreted as indicating
unusually good control

56
 Control chart patterns and corrective
action – Contd..
• Interaction patterns
• Occurs when the level of one
variable affects the behaviour of
other variables associated with the
quality characteristic
• Interaction pattern can be detected
by changing the scheme for rational
sampling
• Example, low pressure and low
temperature may produce a
desirable effect on output
characteristic
• Effective sampling method would
involve controlling the temperature
at several high values and then
determining the effect of pressure
on output characteristic for each
temperature value 57