06-Mar-15

Outline
The Control Chart Techniques
State of Introduction
Control
Specifications
Process Capability
Six Sigma
Different Control Charts

Photo by - Creative Commons Attribution-NonCommercial-ShareAlike License https://www.flickr.com/photos/60648084@N00 Created with Haiku Deck

Learning Objectives Learning Objectives

Know the three categories of variation and their Be able to calculate the central value, trial control
sources. limits and the revised control limits for X bar and
Understand the concept of the control chart R chart.
method. Be able to explain what is meant by a process in
Know the purpose of variable control charts. control and the various out-of-control patterns.
Know how to select the quality characteristics, the Know the difference between individual
rational subgroup and the method of taking measurements and averages; control limits and
samples specifications.

Control Charts

Learning Objectives Variable Attribute

Control Charts Control Charts

Know the different situations between the x-bar and R p-chart

process spread and specifications and what can

be done to correct the undesirable situation. x-tilda and R np-chart

Be able to calculate process capability.

Know the statistical meaning of 6 X-bar and s c-chart

X and MR u-chart

Run Charts

1
 06-Mar-15

Variation Variation
The variation concept is a law of nature in
that no two natural items in any category The variation may be quite large and easily
are the same. noticeable
The variation may be very small. It may
appear that items are identical; however,
precision instruments will show difference
The ability to measure variation is necessary
before it can be controlled

Variation Variations
There are three categories of variation in Sources of Variation in production processes (Ishikawa
piece part production: Fishbone) or 7M:
Measurement
Methods
Materials Men Instruments
1. Within-piece variation: Surface
2. Piece-to-piece variation: Among pieces produced
OUTPUT
INPUTS PROCESS OUTPUTS
at the same time INFLUENCE

3. Time-to-time variation: Difference in product

Measuring Human
produced at different times of the day Tools Machines Inspection
Measuring
Performance
Environment

Variation Variation
Sources of variation are: Sources of variation are:
1. Equipment: 3. Environment
1. Tool wear 1. Temperature
2. Machine vibration 2. Light
3. Electrical fluctuations 3. Radiation
4. Calibration 4. Humidity etc.
2. Material 4. Operator
1. Tensile strength 1. Personal problem
2. Ductility 2. Physical problem etc.
3. Thickness
4. Porosity etc.

2
 06-Mar-15

Variation Variation
There is also a reported variation which is due to Variation may be due to chance causes
the inspection activity. (random causes) or assignable causes.
Variation due to inspection should account for When only chance causes are present, then
one tenth of the four other sources of variation.
the process is said to be in a state of
All variation contribute to errors and measurement
statistical control. The process is stable and
uncertainty, MU.
predictable.

Control Charts Control Charts

Continuous Categorical or Discrete
Variable data Numerical Data Control Numerical Data
Charts
x-bar and R-charts
x-bar and s-charts
Variables Attributes
Charts for individuals (x-charts) Charts Charts
Attribute data
For defectives (p-chart, np-chart) R X P C
Chart Chart Chart Chart
For defects (c-chart, u-chart)

Control Charts for Variables Control Charts

The control chart for variables is a means of

visualizing the variations that occur in the central
tendency and the mean of a set of observations. It
shows whether or not a process is in a stable state.

Figure 5-1 Example of a control chart

3
 06-Mar-15

Control Charts Variable Control Charts

The objectives of the variable control charts are:
1. For quality improvement
2. To determine the process capability
3. For decisions regarding product specifications
4. For current decisions on the production process
5. For current decisions on recently produced items

Figure 5-1 Example of a method of reporting inspection results

Control Chart Techniques Quality Characteristic

Procedure for establishing a pair of control charts for the The Quality characteristic must be measurable.
average Xbar and the range R: It can expressed in terms of the seven basic units:
1. Select the quality characteristic 1. Length (m)

2. Choose the rational subgroup 2. Mass (kg)

3. Collect the data 3. Time (sec.)
4. Determine the trial center line and control limits 4. Electrical current (A)

5. Temperature (K)
5. Establish the revised central line and control limits
6. Amount of Substance (mol)
6. Achieve the objective
7. Luminosity (candela/cd)
(others as appropriate).

Rational Subgroup Rational Subgroup

A rational subgroup is one in which the variation within There are two schemes for selecting the subgroup
a group is due only to chance causes. samples:
Within-subgroup variation is used to determine the 1. Select subgroup samples from product or service
control limits. produced at one instant of time or as close to that
Variation between subgroups is used to evaluate long- instant as possible
term stability. 2. Select from product or service produced over a
period of time that is representative of all the
products or services

4
 06-Mar-15

Rational Subgroup Subgroup Size

The first scheme will have a minimum variation within a As the subgroup size increases, the control
subgroup. limits become closer to the central value, which
The second scheme will have a minimum variation make the control chart more sensitive to small
among subgroups.
variations in the process average
The first scheme is the most commonly used since it
provides a particular time reference for determining As the subgroup size increases, the inspection
assignable causes. cost per subgroup increases
The second scheme provides better overall results and When destructive testing is used and the item
will provide a more accurate picture of the quality.
is expensive, a small subgroup size is required

Subgroup Size Data Collection

From a statistical basis a distribution of Data collection can be accomplished using the
subgroup averages are nearly normal for type of figure shown in Figure 5-2.
groups of 4 or more even when samples It can also be collected using the method in
are taken from a non-normal distribution Table 5-2.
When a subgroup size of 10 or more is It is necessary to collect a minimum of 25
used, the s chart should be used instead of subgroups of data.
the R chart. A run chart can be used to analyze the data in
See Table 5-1 for sample sizes the development stage of a product or prior to a
state of statistical control

Run Chart Trial Central Lines

Central Lines are obtained using:
g g

X i R i

X i 1
and R i 1
g g
where
X average of subgroup averages
X i average of the ith subgroup
g number of subgroups
R average of subgroup ranges
Ri range of the ith subgroup
Figure 5-4 Run Chart for data of Table 5-2

5
 06-Mar-15

Trial Control Limits Trial Control Limits

Trial control limits are established at 3 standard
deviations from the central value In practice calculations are simplified by using the
UCLX X 3 X UCLR R 3 R following equations where A2,D3 and D4 are factors
LCLX X 3 X LCLR R 3 R that vary with the subgroupsize and are found in
where
Table B of the Appendix.
UCL=upper control limit
LCL=lower control limit
UCLX X A2 R UCLR D4 R
X population standard deviation of the subgroup averages
LCLX X A2 R LCLR D3 R
R population standard deviation of the range

Trial Control Limits Revised Central Lines

X new
X X d
and R new
R R d

g gd g gd
where
X d discarded subgroup averages
g d number of discarded subgroups
Rd discarded subgroup ranges
Figure 5-5 Xbar and R chart for preliminary data with trial control limits

Standard Values
R0
X 0 X new R0 R new and 0
d2
UCLX X 0 A 0 UCLR D2 0
LCLX X 0 A 0 LCLR D1 0

Figure 5-6 Trial control limits and revised control limits for Xbar and R charts

6
 06-Mar-15

Achieve the Objective Revised Central Lines

X new
X X d
and R new
R R d

g gd g gd
where
X d discarded subgroup averages
g d number of discarded subgroups
Rd discarded subgroup ranges
Figure 5-7 Continuing use of control charts, showing improved quality

Sample Standard Deviation

Control Chart
Revised Limits for s chart
For subgroup sizes >=10, an s chart is more accurate
X0 X new
X Xd
than an R Chart. Trial control limits are given by: g gd

s0 snew
s s d
0
s0
g gd
i 1 si i 1 X i
g g c4
s X UCLX X 0 A 0 UCLs B6 0
g g LCLX X 0 A 0 LCLs B5 0
UCLX X A3 s UCLs B4 s where
sd discarded subgroup averages
LCLX X A3 s LCLs B3 s c4 , A, B5 , B6 factors found in Table B

State of Control State of Control

Process in Control
When special causes have been eliminated from
the process to the extent that the points
plotted on the control chart remain within the
control limits, the process is in a state of control
When a process is in control, there occurs a
natural pattern of variation

Figure 5-9 Natural pattern of variation of a control chart

7
 06-Mar-15

State of Control State of Control

Types of errors: When the process is in control:
Type I, occurs when looking for a special 1. Individual units of the product or service will
cause of variation when in reality a common be more uniform
cause is present 2. Since the product is more uniform, fewer
Type II, occurs when assuming that a samples are needed to judge the quality
common cause of variation is present when 3. The process capability or spread of the
in reality there is a special cause process is easily attained from 6
4. Trouble can be anticipated before it occurs

State of Control
When the process is in control:
Common
5. The % of product that falls within any pair of Causes
values is more predictable
6. It allows the consumer to use the producers
data
7. It is an indication that the operator is
Special
performing satisfactorily Causes

45

State of Control State of Control

When a point (subgroup value) falls outside its
control limits, the process is out of control.
Out of control means a change in the process
due to a special cause. A process can also be
considered out of control even when the points
fall inside the 3 limits

Figure 5-11 Frequency Distribution of subgroup averages with control limits

8
 06-Mar-15

State of Control
Out-of-Control Condition
It is not natural for seven or more consecutive
points to be above or below the central line. 1. Change or jump in level.
Also when 10 out of 11 points or 12 out of 14 2. Trend or steady change in level
points are located on one side of the central 3. Recurring cycles
line, it is unnatural. 4. Two populations (also called mixture)
Six points in a row are steadily increasing or 5. Mistakes
decreasing indicate an out of control situation

Patterns in Control Charts Patterns in Control Charts

Figure 5-13 Simplified rule for out-of-control pattern

Figure 5-12 Some unnatural runs-process out of control

Out-of-Control Patterns Specifications

Change or jump in level Trend or steady change in level

Recurring cycles Two populations

Figure 5-18 Comparison of individual values compared to averages

9
 06-Mar-15

Specifications Central Limit Theorem

If the population from which samples are taken
Calculations of the average for both the individual
is not normal, the distribution of sample
values and for the subgroup averages are the
averages will tend toward normality provided
same. However the sample standard deviation is

that the sample size, n, is at least 4. This
different.

X
n tendency gets better and better as the sample
where
X population standard deviation of subgroup averages
size gets larger. The standardized normal can be
population standard deviation of individual values used for the distribution averages with the
n=subgroup size
modification. X X
If we assume normality, then the population standard deviation Z
can be estimated from
s X n
c4

Central Limit Theorem Central Limit Theorem

Figure 5-19 Illustration of central limit theorem

Figure 5-20 Dice illustration of central limit theorem

Control Limits & Specifications Control Limits & Specifications

The control limits are established as a function of
the average
Specifications are the permissible variation in
the size of the part and are, therefore, for
individual values
The specifications or tolerance limits are
established by design engineers to meet a
particular function
Figure 5-21 Relationship of limits, specifications, and distributions

10
 06-Mar-15

Process Capability & Tolerance Process Capability & Tolerance

The process spread will be referred to as the Three situations are possible:
process capability and is equal to 6 Case I: When the process capability is less
The difference between specifications is called than the tolerance 6<USL-LSL
the tolerance Case II: When the process capability is equal
When the tolerance is established by the design to the tolerance 6=USL-LSL
engineer without regard to the spread of the Case III: When the process capability is
process, undesirable situations can result greater than the tolerance 6 >USL-LSL

Process Capability & Tolerance Process Capability & Tolerance

Case I: When the process capability is less than Case II: When the process capability is less
the tolerance 6<USL-LSL than the tolerance 6=USL-LSL

Figure 5-24 Case I 6<USL-LSL Figure 5-24 Case I 6=USL-LSL

Process Capability & Tolerance Process Capability

Case III: When the process capability is less The range over which the natural variation of
than the tolerance 6>USL-LSL a process occurs as determined by the system
of common causes
Measured by the proportion of output that
can be produced within design specifications

Figure 5-24 Case I 6>USL-LSL

11
 06-Mar-15

Process Capability Process Capability

This following method of calculating the process The process capability can also be obtained by using the
capability assumes that the process is stable or in standard deviation:
Take 25 (g) subgroups of size 4 for a total of 100
statistical control:
measurements
Take 25 (g) subgroups of size 4 for a total of Calculate the sample standard deviation, s, for each
100 measurements subgroup
Calculate the range, R, for each subgroup Calculate the average sample standard deviation, s
0
Calculate the average range, RBar= R/g sbar = s/g c4
Calculate the estimate of the population
Calculate the estimate of the population
standard deviation
standard deviation
0
R
Process capability will equal 6o
d2
Process capability will equal 60

Capability Index Capability Index

Process capability and tolerance are combined The capability index does not measure process
to form the capability index. performance in terms of the nominal or target
USL LSL value. This measure is accomplished by Cpk.
Cp
6 0 Min{(USL X ) or ( X LSL)
C pk
where C p capabilityindex 3
USL LSL tolerance where C p capabilityindex
6 0 process capability USL LSL tolerance
6 0 process capability

Capability Index Capability Index

1. The Cp value does not change as the process
Cp = USL - LSL center changes
6 o
2. Cp=Cpk when the process is centered
The Capability Index does not measure 3. Cpk is always equal to or less than Cp
process performance in terms of the 4. A Cpk = 1 indicates that the process is producing
nominal or target product that conforms to specifications
5. A Cpk < 1 indicates that the process is producing
Cpk = min{ (USL- X), (X-LSL)} product that does not conform to specifications

12
 06-Mar-15

Capability Index Cpk Measures

6. A Cp < 1 indicates that the process is not Cpk = negative number
capable
7. A Cp=0 indicates the average is equal to one Cpk = zero
of the specification limits
8. A negative Cpk value indicates that the Cpk = between 0 and 1

average is outside the specifications

Cpk = 1

Cpk > 1

Six Sigma Six Sigma

Six Sigma is both a quality management

philosophy and a methodology that focuses on
reducing variation, measuring defects, and
improving quality of products, processes and
services. Figure 5-27 Non-conformance
rate when process is centered

Figure 5-28 Non-conformance

rate when process is off center
1.5

Different Control Charts Different Control Charts

Charts for Better Operator Understanding: Charts for Variable Subgroup Size:
1. Placing individual values on the chart: This Used when the sample size is not the same
technique plots both the individual values and
Different control limits for each subgroup
the subgroup average. Not recommended since
it does not provide much information. As n increases, limits become narrower

2. Chart for subgroup sums: This technique plots As n decreases, limits become wider apart
the subgroup sum, X, rather than the group Difficult to interpret and explain
average, Xbar. To be avoided
UCL X n(UCLX )
UCL X n( LCLX )

13
 06-Mar-15

Different Control Charts Chart for Trends

Chart for Trends: Use Least Square Calculations
Used when the plotted points have an upward X a bG
( X )( G 2 ) ( G )( G X )
or downward trend that can be attributed to an a
g G 2 ( G )2
unnatural pattern of variation or a natural g G X ( G )( X )
b
g G 2 ( G )2
pattern such as tool wear.
where
The central line is on a slope, therefore its X subgroup average and represents the vertical axis
a= point of intercept on the vertical axis
equation must be determined.
b=slope of the line
G=subgroup number and represents the horizontal axis
g=number of subgroups

Chart for Moving Average and

Chart for Trends Moving Range
Value Xbar R

Used when we cannot have 44

multiple observations per 46

54 48.00 10

time period 38 46.00 16

49 47.00 16

X
46 44.33 11
45 46.67 4

X 31
55
40.67
43.67
15
24
n 37 41.00 24

R
42 44.67 18

R
43 40.67 6
47 44.00 5

n 51 47.00 8

Figure 5-32 Chart for Trend

NOTE: n here is equal to 12, NOT 14
An example

Chart for Moving Average and Moving

Range
Chart for Median and Range
This is a simplified variable control chart.
Extreme readings have a greater effect than in Minimizes calculations
conventional charts. An extreme value is used
Easier to understand
several times in the calculations, the number
of times depends on the averaging period. Can be easily maintained by operators
Recommended to use a subgroup of 3,
then all data is used.

14
 06-Mar-15

Chart for Individual values

Chart for Median and Range
UCLMD Md Md A5 RMd Used when only one measurement is taken
on quality characteristic
LCLMD Md Md A5 RMd Too expensive

UCLR D6 RMd Time consuming

Destructive
LCLR D5 RMd Very few items
For Table for A5, D5 and D6 see page 230

Chart for Individual Values Chart for Individual Values

Revised Limits:

X
X R
R
g g X 0 X new R0 R new
UCLx X 2.660 R UCLx X 0 3 0
LCLx X 2.660 R
LCLx X 0 3 0
UCLR 3.267 R
LCLR (0) R UCLR 3.686 R0
To use those equations, you have to use a moving range with n=2 LCLR (0) 0

Charts with Non-Acceptance Limits Charts with Non-Acceptance Limits

Non-Acceptance limits have the same

Relationship to averages as specifications
have to individual values. Control Limits tell what the
process is capable of doing, and reject limits tell
when the product is conforming to specifications.

Figure 5-35 Relationship of non-acceptance limits, control limits

and specifications.

15