4.5.

The statistical process control (SPC)

The SPC is a methodology for monitoring a process to identify special

causes of variation and signal the need to take corrective action when
appropriate

It was Dr. Shewhart who first suggested the use of control charts. In the late
1920s, Shewhart suggested that every process exhibits some degree of
variation. Since no two things can be produced exactly alike, variation is
natural and should be expected.

However, Shewhart discovered that there were two types of variation, chance
cause variation (common causes) and assignable cause variation (special
causes).

SPC relies on control charts. SPC provides means by which a firm may
demonstrate its quality capability.

2GE15 – Quality Management / M. Rosário Moreira 95

4.5. The statistical process control (SPC)

There are three basic uses of control charts.

1st. they are used to monitor a given process. Because a control chart shows
the degree and nature of variation over time, it can be used to determine
whether a process is in a state of statistical control or is out-of-control.

2nd. control charts are used to estimate the parameters (mean, variation)
of a process. By knowing the parameters of a process, the output and the
variability of the output can be predicted.

3rd. control charts are used to improve a process. Once a process is in a

state of statistical control, efforts to reduce process variability can begin. By
reducing the variability of the process, the overall quality of the final product
increases, which reduces scrap and rework and increases profits.

2GE15 – Quality Management / M. Rosário Moreira 96

1
 4.5. The statistical process control (SPC)

But the primary purpose of control charts is to indicate at a glance when

production processes might have changed sufficiently to affect product
quality.

• If the indication is that product quality has deteriorated, or is likely to, then
corrective is taken.

• If the indication is that product quality is better than expected, then it is

important to find out why so that it can be maintained.

2GE15 – Quality Management / M. Rosário Moreira 97

4.5. The statistical process control (SPC)

Histograms do not take

into account changes
over time.

Control charts can tell

us when a process
changes

2GE15 – Quality Management / M. Rosário Moreira 98

2
4.5. The statistical process control (SPC)

Variation around us

It is generally accepted that as variation is reduced, quality is improved.

Sometimes that knowledge is intuitive:

If a train is always on time, schedules can be planned more precisely. If

clothing sizes are consistent, time can be saved by ordering from a catalog.
But rarely are such things thought about in terms of the value of low
variability.

With engineers, the knowledge is better defined. Pistons must fit cylinders,
doors must fit openings, boxes of cereal must have the right amount of raisins
– otherwise quality will be unacceptable and customers will be dissatisfied.

However, engineers also know that it is impossible to have zero variability.

2GE15 – Quality Management / M. Rosário Moreira 99

4.5. The statistical process control (SPC)

For this reason, designers establish specifications that define not only
the target value of something but also acceptable limits about the
target. These design limits are often referred to as the upper and lower
specification limits (or upper and lower tolerance limits).

A traditional way of interpreting such a specification is that any part that falls
within the allowed range is equally good, whereas any part falling outside the
range is totally bad.

2GE15 – Quality Management / M. Rosário Moreira 100

3
4.5. The statistical process control (SPC)

Taguchi pointed out that the traditional view is nonsense for two reasons:

1. From the customer’s view, there is often practically no difference between a

product just inside specifications and a product just outside. Conversely,
there is a far great difference in the quality of a product that is in the target
and the quality of one that is near a limit.
2. As customers get more demanding, there is pressure to reduce variability.

In this graph, the cost is

represented by a smooth curve.

In nearly anything that can be

measured, the customer sees
not a sharp line, but a gradation
of acceptability away from the
“Aim” specification.

2GE15 – Quality Management / M. Rosário Moreira 101

4.5. The statistical process control (SPC)

Process Capability

The Motorola Quality experts argue that the process used to produce a good
(or deliver a service) should be do good that the probability of generating a
defect should be very, very low. Motorola made process capability and product
design famous by adopting Six Sigma limits (we are going to see it later on).

We say that a process is capable when the mean and standard deviation of
the process are operating such that the upper and lower control limits are
acceptable relative to the upper and lower specification limits.

Process capability is defined as a statistical measure of the inherent process

variability for a given characteristics. In other words, process capability refers
to the ability of a process to produce products that meet the specifications set
by the customer or design engineer.

2GE15 – Quality Management / M. Rosário Moreira 102

4
4.5. The statistical process control (SPC)

Process Capability – Illustrating the concept

A company is designing a bearing for a rotating shaft (axle). There are many
variables involved for both the bearing and the axle – for example, the width of
the bearing, the size of the rollers, the size of the axle, the length of the axle,
how it is supported, etc.. The designer specifies tolerances for each of these
variables to ensure that the parts will fit properly.

Suppose that initially a design is selected and the diameter of the bearing is
set at 1,250 inches ± 0,005. This means that acceptable parts may have a
diameter that varies between 1,245 and 1,255 inches. These are the lower
and upper specification limits.

After running the production of 100 bearings and after measuring the diameter
of each one, we find that the average or mean diameter is 1,25 inches – the
process is “centered” right in the middle of the upper and lower specification
limits.
2GE15 – Quality Management / M. Rosário Moreira 103

4.5. The statistical process control (SPC)

Process Capability – Illustrating the concept (cont.)

Let’s say that the diameter values have a standard deviation or sigma equal to
0,002 inch. What this means is that our process does not make each bearing
exactly the same size.

Normally we monitor a process using control charts such that if the process
starts making bearings that are more than three standard deviations (± 0,006
inches) above or below 1,250 inches, we stop the process (we will see it later
on).

We will produce parts that vary between 1,244 and 1,256 inches. These are
the lower and upper CONTROL limits. (or lower and upper process limits)

Do not get confused with the terminology!

2GE15 – Quality Management / M. Rosário Moreira 104

5
4.5. The statistical process control (SPC)

Process Capability – Illustrating the concept (cont.)

The control limits relate to how consistent our process is for making the
bearing. Our goal in managing the process is to keep it within plus or minus 3
standard deviations of the process mean.

The specification limits are related to the design of the part. Recall that, from a
design view, acceptable parts have a diameter between 1,245 and 1,255
inches.

Our process control limits are slightly greater than the specification limits given
to us by the designer. This is not good because we will produce some parts
that do not meet specifications.

2GE15 – Quality Management / M. Rosário Moreira 105

4.5. The statistical process control (SPC)

Process Capability – Illustrating the concept (cont.)

Why doesn't our company, e.g., just check the diameter of each product and
throw out the ones with a diameter less than the lower specification limit or
greater than the upper specification limit?

This could certainly be done and for many, many parts 100 per cent testing is
done. The problem is for a company that is making thousands of parts each
hour, testing each critical dimension of each part made, can be very expensive.
And, for each product, there could be 10 or more additional critical dimensions
in addition to the diameter.

These would all need to be checked! This is why a company uses small
samples to periodically check that the process is in statistical control.

2GE15 – Quality Management / M. Rosário Moreira 106

6
4.5. The statistical process control (SPC)

Process Capability – Illustrating the concept (cont.)

This diagram represents the distribution of the bearing diameter dimension.

Notice that there is a probability (the red areas) of producing defective parts.

2GE15 – Quality Management / M. Rosário Moreira 107

4.5. The statistical process control (SPC)

Process Capability – Illustrating the concept (cont.)

Companies with 6-sigma processes insist that a process making a part be

capable of operating, so that the design specification limits are 6 standard
deviation away from the process mean. For our bearing process, how small
would the process standard deviation need to be for it to be 6-sigma capable?

Tolerance=0,005 inches. So, for a 6-sigma process, the standard deviation

should be 0,005/6 = 0,00083 inches.

So, for the process be 6-sigma capable, the mean diameter produced by the
process would need to be exactly 1,250 inches and the process standard
deviation would need to be less than or equal to 0,00083 inches.

2GE15 – Quality Management / M. Rosário Moreira 108

7
4.5. The statistical process control (SPC)

Process Capability – Illustrating the concept (cont.)

Even though we can not see it in the diagram, there is some probability that a
defect could be produced by this new process, but the probability is very, very
small.
2GE15 – Quality Management / M. Rosário Moreira 109

4.5. The statistical process control (SPC)

Difference between Process Control and Process Capability

Control and capability are two different concepts. A Process may be capable or
not, or in-control or out-of-control, independently of each other.

Clearly, we would like every process to be both capable and in control.

Is the in-control process capable of producing products that meet the

specifications of the customers? After a process has been stabilized, the
behavior of the process defines its capability.

It is important to understand that a process in statistical control will not

necessarily produce units that meet the specifications established by the
customer. Any attempt to improve a process must be preceded by a process
capability study.

2GE15 – Quality Management / M. Rosário Moreira 110

8
 4.5. The statistical process control (SPC)

Process capability is different from the state of process control which

is determined by control charts and based on the average performances of
subgroups against statistically computed control limits.

The difference between process capability and the state of process

control:
• When measurement of an individual item does not meet specification, the
item is called defective.
• When subgroup averages are compared with control limits and the
comparison shows some unpredictable amount of variation, an out-of-
control state is assumed.
• A process in statistical control does not necessarily follow specifications. A
capable process is not necessarily in control. A process may be out of
control and within specification or under control and out of specification.

2GE15 – Quality Management / M. Rosário Moreira 111

4.5. The statistical process control (SPC)

PROCESS

Natural
PROCESS

control Natural
Design control
specs limits
limits

Process cannot meet specifications Process can meet specifications

Natural
PROCESS

Design
control specs
limits

Process capability exceeds specifications

2GE15 – Quality Management / M. Rosário Moreira 112

9
 4.5. The statistical process control (SPC)

The Normal Distribution

s = Standard deviation

At +/- 3 sigma, the most Mean

common choice of -3 s -2 s -1 s +1 s +2 s +3 s
confidence/control limits in
68,27%
quality control application,
the area is 99,73%. 95,45%
99,73%

2GE15 – Quality Management / M. Rosário Moreira 113

4.5. The statistical process control (SPC)

The Normal Distribution

Suppose the process mean of the steel rod cutting process is 150 mm and
that the standard deviation is 5 mm, then from a knowledge of the curve’s
shape and the properties of the normal distribution, the following facts
would emerge:
• 68,27% of the steel rods produced will lie within ±5 mm of the mean,
i.e. μ ± 1s;
• 95,45% of the rods will lie within ±10 mm (μ ± 2s);
• 99,73% of the rods will lie within ±15 mm (μ ± 3s).

2GE15 – Quality Management / M. Rosário Moreira 114

10
4.5. The statistical process control (SPC)

s
Sigma = s = Deviation
( Square root of variance )

Axis graduated in Sigma

-7
-6
-5
-4
-3
-2
-1

1
2
3

5
6
7
0

4
between + / - 1s 68.27 % result: 317300 ppm outside
(deviation)
between + / - 2s 95.45 % 45500 ppm

between + / - 3s 99.73 % 2700 ppm

between + / - 4s 99.9937 % 63 ppm

between + / - 5s 99.999943 % 0.57 ppm

between + / - 6s 99.9999998 % 0.002 ppm

2GE15 – Quality Management / M. Rosário Moreira 115

4.5. The statistical process control (SPC)

Special causes likely
Control Charts

UCL

Nominal

LCL

1 2 3
Samples
The third sample plots outside the original distribution, indicating the likely
presence of an assignable cause.
2GE15 – Quality Management / M. Rosário Moreira 116

11
4.5. The statistical process control (SPC)

Control Limits and Errors

The control limits represent the range between which all points are
expected to fall if the process is in statistical control. If any point fall
outside the control limits or if any unusual patterns are observed, then some
special cause has probably affected the process. The process should be
studied to determine the cause.

If special causes are presented, then they are not representative of the true
state of statistical control and the calculations of the center line and control
limits will be biased. The corresponding data points should be eliminated, and
new values for XdoubleBar and Rbar, and the control limits should be
computed.

2GE15 – Quality Management / M. Rosário Moreira 117

4.5. The statistical process control (SPC)

Control Limits and Errors

sxBar = std dev of sample mean = s/SQRT(n)

(where s = process standard deviation)

Conventional approach uses +- 3 s/SQRT(n)

Limitations of control charts

• Type I Error: probability that an in-control value would appear as
out-of-control
• Type II Error: probability that a shift causing an out-of-control
situation would be mis-reported as in-control
• delays due to sampling interval
• charting without taking action on assignable causes
• over control actions

2GE15 – Quality Management / M. Rosário Moreira 118

12
4.5. The statistical process control (SPC)

Control Limits and Errors

Type 1 and Type 2 Error

Alarm No Alarm

Type 1
error No error
In Control

Out of Control Type 2

No error error

A Type I error occurs when it is concluded that a process is out of control

when in fact pure randomness is present.

2GE15 – Quality Management / M. Rosário Moreira 119

4.5. The statistical process control (SPC)

Control Limits and Errors - Type I error

Type I error:
Probability of
searching for
a cause when UCL
none exists

Process
average

LCL

(a) Three-sigma limits

As long as the area encompassed by the control limits is less than 100% of
the area under the distribution, there will be a probability of a Type I error.

2GE15 – Quality Management / M. Rosário Moreira 120

13
4.5. The statistical process control (SPC)

Control Limits and Errors - Type I error

Type I error:
Probability of
searching for UCL
a cause when
none exists
Process
average

LCL

(b) Two-sigma limits

When the control limits are changed to +/- 2 sigma, the probability of a Type
I error goes up considerably, from 0.0003 to 1 - 0.9544 = 0.0456.

2GE15 – Quality Management / M. Rosário Moreira 121

4.5. The statistical process control (SPC)

Control Limits and Errors - Type II error

Type II error:
Probability of concluding
that nothing has changed

UCL

Shift in process
average Process
average

LCL

(a) Three-sigma limits

2GE15 – Quality Management / M. Rosário Moreira 122

14
 4.5. The statistical process control (SPC)

Control Limits and Errors - Type II error

Type II error:
Probability of concluding
that nothing has changed

UCL

Shift in process
average Process
average

LCL

(b) Two-sigma limits

By reducing the control limits to +/- 2 sigma, we see the probability of failing
to detect the shift has been reduced.
2GE15 – Quality Management / M. Rosário Moreira 123

4.5. The statistical process control (SPC)

Commonly Used Control Charts

• Variables data
 Mean chart (Xbar chart)
 Range chart (R chart)
 Standard Deviations chart (s chart)

• Attribute data
 For “defectives”
− Fraction of nonconforming (p-chart)
− Number of nonconforming (np-chart)
 For “defects”
− Average number of defects (c-chart)
− Average number of defects per unit (u-chart)

2GE15 – Quality Management / M. Rosário Moreira 124

15
4.5. The SPC: Developing Control Charts

1. Prepare
• Choose measurement
• Determine how to collect data, sample size, and frequency of sampling
• Set up an initial control chart
2. Collect Data
• Record data
• Calculate appropriate statistics
• Plot statistics on chart
3. Determine trial control limits
• Center line (process average)
• Compute UCL, LCL
4. Analyze and interpret results
• Determine if in control
• Eliminate out-of-control points
• Recompute control limits as necessary
5. Use as a problem-solving tool
• Continue to collect and plot data
• Take corrective action when necessary
6. Compute process capability
2GE15 – Quality Management / M. Rosário Moreira 125

4.5. The SPC: Developing Control Charts

1. Prepare and 2. Collect Data

2GE15 – Quality Management / M. Rosário Moreira 126

16
4.5. The SPC: Developing Control Charts

3. Determine trial control limits

2GE15 – Quality Management / M. Rosário Moreira 127

4.5. The SPC: Developing Control Charts

4. Analyze and interpret results

2GE15 – Quality Management / M. Rosário Moreira 128

17
4.5. The SPC: Developing Control Charts

4. Analyze and interpret results and revise charts

2GE15 – Quality Management / M. Rosário Moreira 129

4.5. The SPC: Developing Control Charts

4. Analyze and interpret results and revise charts

A. Interpret the original charts

B. Isolate the cause
C. Take corrective action
D. Revise the chart: remove any points from the calculations that have been
corrected. Revise the control charts with the remaining points

2GE15 – Quality Management / M. Rosário Moreira 130

18
 4.5. The SPC: Developing Control Charts

4. Analyze and interpret results and 5. Use as a problem-solving tool

Decide if the variation is random (chance/common causes) or unusual

(assignable causes).

A process is considered to be in a state of control, or under control, when

the performance of the process falls within the statistically calculated control
limits and exhibits only chance, or common, causes.

2GE15 – Quality Management / M. Rosário Moreira 131

4.5. The SPC: Developing Control Charts

4. Analyze and interpret results and 5. Use as a problem-solving tool

A control chart exhibits a state of control when:

i. Two-thirds of the points are near the center value.
ii. A few of the points are close to the center value.
iii. The points float back and forth across the centerline.
iv. The points are balanced on both sides of the centerline.
v. There no points beyond the limits.
vi. There are no patterns or trends on the chart.
• Upward/downward, oscillating trend
• Change, jump, or shift in level
• Runs
• Recurring cycles

2GE15 – Quality Management / M. Rosário Moreira 132

19
 4.5. The SPC: Developing Control Charts

4. Analyze and interpret results and 5. Use as a problem-solving tool

Typical Out-of-Control Patterns

• Point outside control limits

• Sudden shift in process average
• Cycles
• Trends
• To stay close to the center line
• To stay close to the control limits
• Instability

2GE15 – Quality Management / M. Rosário Moreira 133

4.5. The SPC: Developing Control Charts

Typical Out-of-Control Patterns: Identifying potential shifts

2GE15 – Quality Management / M. Rosário Moreira 134

20
4.5. The SPC: Developing Control Charts

Typical Out-of-Control Patterns: Shift in process average

2GE15 – Quality Management / M. Rosário Moreira 135

4.5. The SPC: Developing Control Charts

Typical Out-of-Control Patterns: Cycles

2GE15 – Quality Management / M. Rosário Moreira 136

21
4.5. The SPC: Developing Control Charts

Typical Out-of-Control Patterns: Trend

2GE15 – Quality Management / M. Rosário Moreira 137

4.5. The SPC: Developing Control Charts

6. Compute Process Capability

• The range over which the natural variation of a process occurs as

determined by the system of common causes
• Measured by the proportion of output that can be produced within design
specifications

• Specifications are set by the customer. These are the “wishes”.

• Control limits are obtained by applying statistical rules on the data
generated by the process. These are the “reality.”

Process capability refers to the ability of a process to meet the

specifications set by the customer or designer (not the statistically
computed control limits)

2GE15 – Quality Management / M. Rosário Moreira 138

22
4.5.1. Constructing a Control Chart for Variables: Xbar and R

1. Define the problem

2. Select the quality characteristics to be measured
3. Choose a rational subgroup size to be sampled
4. Collect the data
5. Determine the trial centerline for the Xbar chart
6. Determine the trial control limits for the Xbar chart
7. Determine the trial control limits for the R chart
8. Examine the process: control chart interpretation
9. Revise the charts
10. Process capability calculus

2GE15 – Quality Management / M. Rosário Moreira 139

4.5.1. Constructing a Control Chart for Variables: Xbar and R

Mean chart (Xbar Chart)

Measures central tendency of a sample

Range chart (R Chart)

Measures amount of dispersion in a sample

Each chart measures the process differently. Both the process average and
process variability must be in control for the process to be in control.

2GE15 – Quality Management / M. Rosário Moreira 140

23
 4.5.1. Constructing a Control Chart for Variables: Xbar and R

5. Determine the trial centerline for the Xbar chart

• Compute the mean of sample means:

X i

X i 1
where m  number of subgroups
m

2GE15 – Quality Management / M. Rosário Moreira 141

4.5.1. Constructing a Control Chart for Variables: Xbar and R

6. Determine the trial control limits for the Xbar chart

• Compute the mean of sample ranges:

m

R i
R i 1
m
• Set 3-std.-dev. control limits for the sample means:

3s
UCLX  X  A2 R  X  3s x  X 
n
3s
LCL X  X  A2 R  X  3s x  X 
n

Where, A2 is obtained from the Table of Factors for Computing 3-sigma limits
(next slide)
2GE15 – Quality Management / M. Rosário Moreira 142

24
 4.5.1. Constructing a Control Chart for Variables: Xbar and R

2GE15 – Quality Management / M. Rosário Moreira 143

4.5.1. Constructing a Control Chart for Variables: Xbar and R

7. Determine the trial control limits for the R chart

• Set 3-std.-dev. control limits for the sample ranges:

UCLR  D4 R
LCL R  D3 R

R i
R i 1
m
Where, D3 and D4 are obtained from the Table of Factors for Computing 3-
sigma limits .

2GE15 – Quality Management / M. Rosário Moreira 144

25
4.5.1. Constructing a Control Chart for Variables: Xbar and R

9. Revise the charts

X new 
X X d

m  md

R new 
RR d

m  md

Where:

X d = discarded subgroup averages

md = number of discarded subgroups
Rd = discarded subgroup ranges

2GE15 – Quality Management / M. Rosário Moreira 145

4.5.1. Constructing a Control Chart for Variables: Xbar and R

9. Revise the charts

The formula for the revised limits are:

X 0  X new , R0  R new
R0
s0 
d2
UCL X  X 0  As 0
LCL X  X 0  As 0

UCLR  D2 s 0 where, A, D1, and D2 are obtained from

LCL R  D1s 0 the Table of Factors for Computing 3-sigma limits .

2GE15 – Quality Management / M. Rosário Moreira 146

26
4.5.1. Constructing a Control Chart for Variables: Xbar and R

10. Process capability analysis

1st Compute the mean of sample means (Xdouble_bar ).

2nd Compute the mean of sample ranges (Rbar ).

3rd Estimate the population standard deviation (σx):

σx = Rbar / d2

4th Estimate the natural tolerance of the process:

Natural tolerance = 6σx

5th Determine the specification limits:

USL = Upper specification limit
LSL = Lower specification limit

2GE15 – Quality Management / M. Rosário Moreira 147

4.5.1. Constructing a Control Chart for Variables: Xbar and R

10. Process capability analysis

6th Compute capability indices:

Process capability index
Cp = (USL – LSL) / 6σx

Upper capability index

CpU = (USL – Xdouble_bar) / 3σx

Lower capability index

CpL = ( Xdouble_bar – LSL) / 3σx

Process capability index (min)

Cpk = Minimum (CpU, CpL)

2GE15 – Quality Management / M. Rosário Moreira 148

27
4.5.1. Constructing a Control Chart for Variables: Xbar and R

10. Process capability analysis

Suppose that process mean = (USL+LSL)/2

Case 1: 6σ < USL - LSL
Ideal condition, the process will remain within the specifications even
after some shift.
Case 2: 6σ = USL - LSL
As long as the process remains in control and centered (so the process
does not shift), the parts produced will be within specification.
Case 3: 6σ > USL - LSL
Undesirable situation, the process is incapable of meeting the
specifications. Management intervention is needed in order to decrease
the variation

2GE15 – Quality Management / M. Rosário Moreira 149

4.5.1. Constructing a Control Chart for Variables: Xbar and R

10. Process capability analysis

Process capability index: Cp = (USL – LSL) / 6σx

If the capability index is larger than 1.00, a Case 1 situation exists. This
is desirable. The greater this value, the better. The process will remain
capable even after a slight shift of the process mean.
If the capability index is equal to 1.00, a Case 2 situation exists. This is
not the best. However, if the process is in control and the mean is
centered, the process is capable.
If the capability index is less than 1.00, a Case 3 situation exists. This is
undesirable and the process is not capable to meet the specifications.

2GE15 – Quality Management / M. Rosário Moreira 150

28
 4.5.1. Constructing a Control Chart for Variables: Xbar and R

10. Process capability analysis

6s
Capability ratio: Cr 
USL  LSL

The capability ratio is the inverse of the capability index and interpreted
similarly. A capability ratio of less than 1 is the most desirable situation.

2GE15 – Quality Management / M. Rosário Moreira 151

4.5.1. Constructing a Control Chart for Variables: Xbar and R

Example 1

Number of samples: 25

Sample size: 3

Data

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25

1 41 78 84 60 46 64 43 37 50 57 24 78 51 41 56 46 99 71 41 41 22 62 64 44 41

2 70 53 34 36 47 16 53 43 29 83 42 48 57 29 64 41 86 54 2 39 40 70 52 38 63

3 22 68 48 25 29 56 64 30 57 32 39 39 50 35 36 16 98 39 53 36 46 46 57 60 62

2GE15 – Quality Management / M. Rosário Moreira 152

29
4.5.1. Constructing a Control Chart for Variables: Xbar and R

Example 2
Slip Ring Diameter (cm)
Sample 1 2 3 4 5 Xbar R
1 5.02 5.01 4.94 4.99 4.96 4.98 0.08
2 5.01 5.03 5.07 4.95 4.96 5.00 0.12
3 4.99 5.00 4.93 4.92 4.99 4.97 0.08
4 5.03 4.91 5.01 4.98 4.89 4.96 0.14
5 4.95 4.92 5.03 5.05 5.01 4.99 0.13
6 4.97 5.06 5.06 4.96 5.03 5.01 0.10
7 5.05 5.01 5.10 4.96 4.99 5.02 0.14
8 5.09 5.10 5.00 4.99 5.08 5.05 0.11
9 5.14 5.10 4.99 5.08 5.09 5.08 0.15
10 5.01 4.98 5.08 5.07 4.99 5.03 0.10
50.09 1.15

2GE15 – Quality Management / M. Rosário Moreira 153

4.5.1. Constructing a Control Chart for Variables: Xbar and s

Xbar and s chart

The Xbar chart shows the center of the measurements and the R chart the
spread of the data.
An alternative combination is the Xbar and s chart. The Xbar chart shows the
central tendency and the s chart the dispersion of the data.

Why s chart instead of R chart?

• Range is computed with only two values, the maximum and the minimum.
However, s is computed using all the measurements corresponding to a
sample.
• So, an R chart is easier to compute, but s is a better estimator of standard
deviation specially for large subgroups

2GE15 – Quality Management / M. Rosário Moreira 154

30
4.5.1. Constructing a Control Chart for Variables: Xbar and s

Xbar and s chart

The control limits of Xbar chart are:

s
X 3
n

The above limits can also be written as:

m
UCLX  X  A3 s X i

X i 1
LCLX  X  A3 s m
m  the number of subgroups

Where A3 is obtained from the Table of Factors for Computing 3-sigma limits.

2GE15 – Quality Management / M. Rosário Moreira 155

4.5.1. Constructing a Control Chart for Variables: Xbar and s

Xbar and s chart: Trial Control Limits

The control limits of s chart are:

UCLS  B4 s
LCLs  B3 s

Where
m

s i
s i 1
m
m  the number of subgroups

Where B3 and B4 are obtained from the Table of Factors for 3-sigma limits .

2GE15 – Quality Management / M. Rosário Moreira 156

31
 4.5.1. Constructing a Control Chart for Variables: summary

2GE15 – Quality Management / M. Rosário Moreira 157

4.5.2. Constructing a Control Chart for Attributes

• In a control chart for variables, quality characteristic is expressed in

numbers. Many quality characteristics (e.g., clarity of glass) can be
observed only as attributes, i.e., by classifying into defectives and non-
defectives.

• If many quality characteristics are measured, a separate control chart for

variable will be needed for each quality characteristic. A control chart for
attribute is a cheaper alternative. It records an item defective if any
specification is not met and non-defective if all the specifications are met.

• The cost of collecting data for attributes is less than that for the variables

2GE15 – Quality Management / M. Rosário Moreira 158

32
 4.5.2. Constructing a Control Chart for Attributes

There are various types of control charts for attributes:

• The p chart for the fraction rejected
 based on fraction (percentage) of defective units in a varying sample
size

• The np chart for the total number rejected

 based on number of defective units in a fixed sample size

• The c chart for the number of defects

 based on the count of defects found in a fixed sample size

• The u chart for the number of defects per unit

 based on the counts of defects in a varying sample size

2GE15 – Quality Management / M. Rosário Moreira 159

4.5.2. Constructing a Control Chart for Attributes: p chart

The p Chart: Constant Sample Size

Steps
1. Gather data
2. Calculate p, the proportion of defectives
3. Plot the proportion of defectives on the control chart
4. Calculate the centerline and the control limits (trial)
5. Draw the centerline and control limits on the chart
6. Interpret the chart
7. Revise the chart

2GE15 – Quality Management / M. Rosário Moreira 160

33
 4.5.2. Constructing a Control Chart for Attributes: p chart

4. Calculate the centerline and the control limits (trial)

p
 np
n
p (1  p )
UCLp  p  3
n
p (1  p )
LCL p  p  3
n

2GE15 – Quality Management / M. Rosário Moreira 161

4.5.2. Constructing a Control Chart for Attributes: p chart

6. Interpret the chart

• The interpretation is similar to that of a variable control chart. There

should be no patterns in the data such as trends, runs, cycles, or sudden
shifts in level. All of the points should fall between the upper and lower
control limits.
• One difference is that for the p chart it is desirable that the points lie near
the lower control limit
• The process capability is pbar the centerline of the p chart

2GE15 – Quality Management / M. Rosário Moreira 162

34
4.5.2. Constructing a Control Chart for Attributes: p chart

7. Revise the chart

Revised centerline and control limits for the p chart .

p new 
 np  np d

nn d

p new (1  p new )
UCLp  p new  3
n
p new (1  p new )
LCLp  p new  3
n

2GE15 – Quality Management / M. Rosário Moreira 163

4.5.2. Constructing a Control Chart for Attributes: np chart

The np Chart

The np chart construction steps are similar to those of the p chart. The trial
centerline and control limits are as follows:

Centerline n p 
 np
m
where, m  number of subgroups
UCLnp  n p  3 n p(1  p)

LCLnp  n p  3 n p(1  p)

2GE15 – Quality Management / M. Rosário Moreira 164

35
 4.5.2. Constructing a Control Chart for Attributes: p / np chart

Choice Between the p and np Charts

• If the sample size varies, p chart is more appropriate

• If the sample size is constant, np chart may be used

• Smaller samples are

 less sensitive to the changes in the quality levels and
 less satisfactory as an indicator of the assignable causes of variation

2GE15 – Quality Management / M. Rosário Moreira 165

4.5.2. Constructing a Control Chart for Attributes: p / np chart

Example 1:

A manufacturer purchases small bolts in cartons that usually contain several

thousand bolts. Each shipment consists of a number of cartons. As part of the
acceptance procedure for these bolts, 400 bolts are selected at random from
each carton and are subjected to visual inspection for certain non-
conformities. In a shipment of 10 cartons, the respective percentages of
rejected bolts in the samples from each carton are 0, 0, 0.5, 0.75, 0,2.0,
0.25, 0, 0.25, and 1.25. Does this shipment of bolts appear to exhibit
statistical control with respect to the quality characteristics examined in this
inspection?

2GE15 – Quality Management / M. Rosário Moreira 166

36
4.5.2. Constructing a Control Chart for Attributes: p / np chart

Example 2:

An item is made in lots of 200 each. The lots are given 100% inspection. The
record sheet for the first 25 lots inspected showed that a total of 75 items did
not conform to specifications.
a. Determine the trial limits for an np chart.
b. Assume that all points fall within the control limits. What is your estimate
of the process average fraction nonconforming  p?

2GE15 – Quality Management / M. Rosário Moreira 167

4.5.2. Constructing a Control Chart for Attributes: p / np chart

Example 3:

A manufacturer wishes to maintain a process average of 0.5% nonconforming

product or less. 1500 units are produced per day, and 2 days’ runs are
combined to form a shipping lot. It is decided to sample 250 units each day
and use an np chart to control production.
(a) Find the 3-sigma control limits for this process.

2GE15 – Quality Management / M. Rosário Moreira 168

37
 4.5.2. Constructing a Control Chart for Attributes: p chart

The p Chart: Variable Sample Size

• When the number of items sampled varies, the p chart can be easily
adapted to varying sample sizes
• If the sample size varies
 the control limits must be calculated for each different sample size,
changing the n in the control-limit formulas each time a different
sample size is taken.
 calculating the centerline and interpreting the chart will be the same

2GE15 – Quality Management / M. Rosário Moreira 169

4.5.2. Constructing a Control Chart for Attributes: p chart

The p Chart: Variable Sample Size

Steps
1. Gather data
2. Calculate p, the proportion of defectives
3. Plot the proportion of defectives on the control chart
4. Calculate the centerline. For each sample calculate a separate pair of
control limits.
5. Draw the centerline and control limits on the chart
6. Interpret the chart

2GE15 – Quality Management / M. Rosário Moreira 170

38
 4.5.2. Constructing a Control Chart for Attributes: p chart

The p Chart: Variable Sample Size

4. Calculate the centerline. For each sample calculate a separate pair of

control limits.
Let m – number of samples.
m

n p i i
p i 1
m
(one centerline for all samples)
n
i 1
i

p(1  p )
UCLp  p  3 for the i  th sample
ni
p (1  p )
LCLp  p  3 for the i  th sample
ni

2GE15 – Quality Management / M. Rosário Moreira 171

4.5.2. Constructing a Control Chart for Attributes: p chart

The p Chart: Variable Sample Size

6. Interpret the chart

• The interpretation is similar to that of a variable control chart. there should

be no patterns in the data such as trends, runs, cycles, or sudden shifts in
level. All of the points should fall between the upper and lower control
limits.
• One difference is that for the p chart it is desirable that the points lie near
the lower control limit
• The process capability is pbar, the centerline of the p chart

2GE15 – Quality Management / M. Rosário Moreira 172

39
 4.5.2. Constructing a Control Chart for Attributes: c and u chart

The c and u Chart

Defective and defect

• A defective article is the one that fails to conform to some specification.
• Each instance of the article’s lack of conformity to specifications is a defect
• A defective article may have one or more defects

2GE15 – Quality Management / M. Rosário Moreira 173

4.5.2. Constructing a Control Chart for Attributes: c and u chart

The c and u Chart

The np and c charts

 Both the charts apply to total counts
 The np chart applies to the total number of defectives in samples
of constant size
 The c chart applies to the total number of defects in samples of
constant size

The p and u charts

 The p chart applies to the proportion of defectives
 The u chart applies to the number of defects per unit
 If the sample size varies, the p and u charts may be used

2GE15 – Quality Management / M. Rosário Moreira 174

40
4.5.2. Constructing a Control Chart for Attributes: c chart

The c Chart - Constant Sample Size

The number of nonconformities, or c, chart is used to track the count of

nonconformities observed in a single unit of product of constant size.

Steps
1. Gather the data
2. Count and plot c, the count of nonconformities, on the control chart.
3. Calculate the centerline and the control limits (trial)
4. Draw the centerline and control limits on the chart
5. Interpret the chart
6. Revise the chart

2GE15 – Quality Management / M. Rosário Moreira 175

4.5.2. Constructing a Control Chart for Attributes: c chart

The c Chart

3. Calculate the centerline and the control limits (trial)

Centerline c 
c
m
UCLc  c  3 c
LCLc  c  3 c
Where, m  number of samples

2GE15 – Quality Management / M. Rosário Moreira 176

41
 4.5.2. Constructing a Control Chart for Attributes: c chart

The c Chart

5. Interpret the chart

• The interpretation is similar to that of a variable control chart. there should

be no patterns in the data such as trends, runs, cycles, or sudden shifts in
level. All of the points should fall between the upper and lower control
limits.
• One difference is that for the c chart it is desirable that the points lie near
the lower control limit
• The process capability is cbar the centerline of the c chart

2GE15 – Quality Management / M. Rosário Moreira 177

4.5.2. Constructing a Control Chart for Attributes: c chart

The c Chart

6. Revise the chart (Revised centerline and control limits)

Centerline c new 
c  c d

m  md
UCLcnew  c new  3 c enw

LCLcnew  c new  3 c enw

2GE15 – Quality Management / M. Rosário Moreira 178

42
4.5.2. Constructing a Control Chart for Attributes: u chart

The u Chart - Constant Sample Size

The number of nonconformities per unit, or u chart is used to track the

number of nonconformities in a unit.

Steps
1. Gather the data
2. Count and plot u, the number of nonconformities per unit, on the control
chart.
3. Calculate the centerline and the control limits (trial)
4. Draw the centerline and control limits on the chart
5. Interpret the chart
6. Revise the chart

2GE15 – Quality Management / M. Rosário Moreira 179

4.5.2. Constructing a Control Chart for Attributes: u chart

The u Chart

2. Count and plot u, the number of nonconformities per unit, on the control
chart.

Let
n  number of inspecteditems in a sample
c  number of nonconformities in a sample
c
u
n

2GE15 – Quality Management / M. Rosário Moreira 180

43
4.5.2. Constructing a Control Chart for Attributes: u chart

The u Chart

3. Calculate the centerline and the control limits (trial)

Centerline u 
c
n
u
UCLu  u  3
n
u
LCLu  u  3
n

2GE15 – Quality Management / M. Rosário Moreira 181

4.5.2. Constructing a Control Chart for Attributes: u chart

The u Chart

6. Revise the chart

Centerline u 
c c d

nn d

u new
UCLunew  u new  3
n
u new
LCLunew  u new  3
n

2GE15 – Quality Management / M. Rosário Moreira 182

44
4.5.2. Constructing a Control Chart for Attributes: c and u chart

Example 4:
A c chart is used to monitor the number of surface imperfections on sheets of
photographic film. The chart presently is set up based on cbar of 2.6.
(a) Find the 3-sigma control limits for this process.

Example 5:
A shop uses a control chart on maintenance workers based on maintenance
errors per standard worker-hour. For each worker, a random sample of 5
items is taken daily and the statistic c/n is plotted on the worker’s control
chart where c is the count of errors found in 5 assemblies and n is the total
worker-hours required for the 5 assemblies.
(a) After the first 4 weeks, the record for one worker is Sum c=22 and Sum
n=54. Determine the central line and the 3-sigma control limits.
(b) On a certain day during the 4-week period, the worker makes 2 errors in
4,3 standard worker-hour. Determine if the point for this day falls within
control limits.

2GE15 – Quality Management / M. Rosário Moreira 183

4.5.2. Constructing a Control Chart for Attributes: summary

2GE15 – Quality Management / M. Rosário Moreira 184

45
4.5. When to use the different control charts

2GE15 – Quality Management / M. Rosário Moreira 185

4.6. Quality control in Services

In all services there is a continuing need to monitor quality.

Control charts are used extensively in services to monitor and control their
quality levels

• Identify important quality characteristics

Time, errors, behavior
• Measure and collect data using check sheets, check lists, or other means
• Take corrective action

2GE15 – Quality Management / M. Rosário Moreira 186

46
4.6. Quality control in Services

2GE15 – Quality Management / M. Rosário Moreira 187

Areas
A table entry is the proportion of the
area under the curve from a z of 0 to a
positive value of z. To find the area
under from a z of 0 to a negative z, subtract the
tabled value from 1.

Standardize z
0.0
.00
0.5000
.01
0.5040
.02
0.5080
.03
0.5120
.04
0.5160
.05
0.5199
.06
0.5239
.07
0.5279
.08
0.5319
.09
0.5359
0.1 0.5398 0.5438 0.5478 0.5517 0.5557 0.5596 0.5636 0.5675 0.5714 0.5753

d Normal
0.2 0.5793 0.5832 0.5871 0.5910 0.5948 0.5987 0.6026 0.6064 0.6103 0.6141
0.3 0.6179 0.6217 0.6255 0.6293 0.6331 0.6368 0.6406 0.6443 0.6480 0.6517
0.4 0.6554 0.6591 0.6628 0.6664 0.6700 0.6736 0.6772 0.6808 0.6844 0.6879

Distribution 0.5
0.6
0.6915
0.7257
0.6950
0.7291
0.6985
0.7324
0.7019
0.7357
0.7054
0.7389
0.7088
0.7422
0.7123
0.7454
0.7157
0.7486
0.7190
0.7517
0.7224
0.7549
0.7 0.7580 0.7611 0.7642 0.7673 0.7704 0.7734 0.7764 0.7794 0.7823 0.7852
0.8 0.7881 0.7910 0.7939 0.7967 0.7995 0.8023 0.8051 0.8078 0.8106 0.8133
0.9 0.8159 0.8186 0.8212 0.8238 0.8264 0.8289 0.8315 0.8340 0.8365 0.8389
1.0 0.8413 0.8438 0.8461 0.8485 0.8508 0.8531 0.8554 0.8577 0.8599 0.8621
1.1 0.8643 0.8665 0.8686 0.8708 0.8729 0.8749 0.8770 0.8790 0.8810 0.8830
1.2 0.8849 0.8869 0.8888 0.8907 0.8925 0.8944 0.8962 0.8980 0.8997 0.9015
1.3 0.9032 0.9049 0.9066 0.9082 0.9099 0.9115 0.9131 0.9147 0.9162 0.9177
1.4 0.9192 0.9207 0.9222 0.9236 0.9251 0.9265 0.9279 0.9292 0.9306 0.9319
1.5 0.9332 0.9345 0.9357 0.9370 0.9382 0.9394 0.9406 0.9418 0.9429 0.9441
1.6 0.9452 0.9463 0.9474 0.9484 0.9495 0.9505 0.9515 0.9525 0.9535 0.9545
1.7 0.9554 0.9564 0.9573 0.9582 0.9591 0.9599 0.9608 0.9616 0.9625 0.9633
1.8 0.9641 0.9649 0.9656 0.9664 0.9671 0.9678 0.9686 0.9693 0.9699 0.9706
1.9 0.9713 0.9719 0.9726 0.9732 0.9738 0.9744 0.9750 0.9756 0.9761 0.9767
2.0 0.9772 0.9778 0.9783 0.9788 0.9793 0.9798 0.9803 0.9808 0.9812 0.9817
2.1 0.9821 0.9826 0.9830 0.9834 0.9838 0.9842 0.9846 0.9850 0.9854 0.9857
2.2 0.9861 0.9864 0.9868 0.9871 0.9875 0.9878 0.9881 0.9884 0.9887 0.9890
2.3 0.9893 0.9896 0.9898 0.9901 0.9904 0.9906 0.9909 0.9911 0.9913 0.9916
2.4 0.9918 0.9920 0.9922 0.9925 0.9927 0.9929 0.9931 0.9932 0.9934 0.9936
2.5 0.9938 0.9940 0.9941 0.9943 0.9945 0.9946 0.9948 0.9949 0.9951 0.9952
2.6 0.9953 0.9955 0.9956 0.9957 0.9959 0.9960 0.9961 0.9962 0.9963 0.9964
2.7 0.9965 0.9966 0.9967 0.9968 0.9969 0.9970 0.9971 0.9972 0.9973 0.9974
2.8 0.9974 0.9975 0.9976 0.9977 0.9977 0.9978 0.9979 0.9979 0.9980 0.9981
2.9 0.9981 0.9982 0.9982 0.9983 0.9984 0.9984 0.9985 0.9985 0.9986 0.9986
3.0 0.9987 0.9987 0.9987 0.9988 0.9988 0.9989 0.9989 0.9989 0.9990 0.9990
3.1 0.9990 0.9991 0.9991 0.9991 0.9992 0.9992 0.9992 0.9992 0.9993 0.9993
3.2 0.9993 0.9993 0.9994 0.9994 0.9994 0.9994 0.9994 0.9995 0.9995 0.9995
3.3 0.9995 0.9995 0.9995 0.9996 0.9996 0.9996 0.9996 0.9996 0.9996 0.9997
3.4 0.9997 0.9997 0.9997 0.9997 0.9997 0.9997 0.9997 0.9997 0.9997 0.9998

2GE15 – Quality Management / M. Rosário Moreira 188

47