STATISTICAL CONTROL OF

PROCESSES

ADVANCED ZIP CODE COURSE

HERBERT OLIVEIRA, Edited
Topics Covered

Common and Special Causes of Variation

Control Charts for variables
Statistical Base Control Charts
Limits of Control versus Limits of
Specification
Guidelines for graph planning
of control
Use and Interpretation of Control Charts

2
Causes of Variation
The causes of variation can be separated
em:

Common or random causes

Notable or special causes

3
Causes of Variation
Common or random causes
Inherent to the process itself
Relatively difficult to be identified
Consist of a very large amount
of small causes.
Typical Examples
Systematic purchase of raw materials with
low quality
Non-existence of training
Lack of depadronization of operations
4
Causes of Variation
Notable or special causes
They represent a temporary loss of control of the
process.
Possible to be identified and corrected.
The causes and the effects are easily
observable.
Typical Examples
Isolated batch of raw material with problem
Occasional deregulation of equipment
production
Measurement equipment failure
5
Causes of Variation

6
Causes of Variation

Extracted according to HERBERT OLIVEIRA,

2016

7
Causes of Variation

8
Causes of Variation

9
Causes of Variation
Most quality problems have the
your origin in common causes (problems)
of project) and not in special causes
(operational problems).

10
Causes of Variation

11
Causes of Variation

12
Variation and Predictability
There are no two objects in nature that
be absolutely equal. There is always
variation.
However, the variation due only to
common causes are predictable.

13
Variation and Predictability

14
Variation and Predictability

15
Improvement through the ZIP code

16
Improvement through the ZIP code

17
Control Charts
Objectives
Check if the process is stable
Keep the process stable
Improve the process performance
Types of control charts
Variables
Measures
Attributes
Counting
Classification

18
Control Charts
Ideal State
Process under control
100% Compliant Product
Control Letters
Keep the process under control
We alert to the occurrence of problems

Pseudo-Ideal State
Process under control
Some non-compliant product
Possibilities
Changes in the process or
Changes in specifications
Control Charts
Keep the process under control
Evaluate efforts for improvement

Pre-Chaotic State
Out of control process
100% Compliant Product
apparently good, but...
Special causes determine what is
produced by the process
Quality and compliance can
Chaotic State change at any time
Process out of control
Some non-compliant product
Special causes still dominate
Random Fluctuations due to
special causes will eventually
frustrate efforts in improvement of
quality
Quality and compliance can
change at any time
The only way to get out of the
chaos is to eliminate the special causes

Possible states of a process

19
Control Chart Planning
Important points:

Control Limits

Sampling Frequency

Sample Size

20
Control Chart Planning

Control Limits
Standard action limit: +/- 3
Justification: practical results.
Alert limit: +/- 2
Greater sensitivity
Confusing for the operations staff
Increase in the risk of 'false alarms'

21
Control Chart Planning

22
Control Charts for Variables
When variables are employed,
two graphs are needed:
to control the centralization of the process
(average)
another to control your variability
(dispersion)

Important: Each control chart has

a very specific purpose and a no
replaces the other.
23
Control Charts for Variables

24
Development and Use of Letters
X-Bar and R Control
In the previous section, the foundations for the
X-Bar and R control charts.

The following example demonstrates the practical use of these cards.

Example: Piston rings for an automotive engine are

produced by a forging process. It is desired to establish a
statistical control of the internal diameter of the produced rings
for this process using X-bar and R control charts. Twenty
and five samples, each with a size equal to 5, were
taken at a moment when the process is considered under
control. The measurement data of the internal diameter are the
shown in the table below:

25
Development and Use of the Letters of
X-Bar and R Control
Sample Observations Xi Ri
1 74,030 74,002 74,019 73,992 74,008 74,010 0.038
2 73,995 73,992 74,001 74,011 74,004 74,001 0.019
3 73,988 74,024 74,021 74,005 74,002 74,008 0.036
4 74,002 73,996 73,993 74,015 74,009 74,003 0.022
5 73,992 74,007 74,015 73,989 74,014 74,003 0.026
6 74,009 73,994 73,997 73,985 73,993 73,996 0.024
7 73,995 74,006 73,994 74,000 74,005 74,000 0.012
8 73,985 74,003 73,993 74,015 73,988 73,997 0.030
9 74,008 73,995 74,009 74,005 74,004 74,004 0.014
10 73,998 74,000 73,990 74,007 73,995 73,998 0.017
11 73,994 73,998 73,994 73,995 73,990 73,994 0.008
12 74,004 74,000 74,007 74,000 73,996 74,001 0.011
13 73,986 74,002 73,998 73,997 74,012 73,998 0.029
14 74,006 73,967 73,994 74,000 73,984 73,990 0.039
15 74,012 74,014 73,998 73,999 74,007 74,006 0.016
16 74,000 73,984 74,005 73,998 73,996 73,997 0.021
17 73,994 74,012 73,986 74,005 74,007 74,001 0.026
18 74,006 74,010 74,018 74,003 74,000 74,007 0.018
19 73,984 74,002 74,003 74,005 73,997 73,998 0.021
20 74,000 74,010 74,013 74,020 74,003 74,009 0.020
21 73,982 74,001 74,015 74,005 73,996 74,000 0.033
22 74,004 73,999 73,990 74,006 74,009 74,002 0.019
23 74,010 73,989 73,990 74,009 74,014 74,002 0.025
24 74,015 74,008 73,993 74,000 74,010 74,005 0.020
25 73,982 73,984 73,995 74,017 74,013 73,998 0.035

1850.028 0, 581
X 74,001 R 0.023
26
Development and Use of the Letters of
X-Bar and R Control
Recommendation: start using the letter of
amplitude (R), then the control limits of the chart
X-bar depends on the variability of the process and
without variability being under control, these
limits don't have much meaning.
Using the data from the table, the row is found
center of letter R:
25
Ri
0.581
R i 1
0.023
25 25

27
Development and Use of Letters
X-Bar and R Control
Once the R control chart indicates that the
the process variability is controlled, one can
to use the X-Bar chart. The central line is:
25
Xi
i 1 1850,028
X 74,001
25 25
From the table, we have A20.577
for the samples of
size n = 5. The control limits of the X chart
bars are:

28
Development and Use of the Letters of
X-Bar and R Control
The X-bar chart is shown in the following figure.
the chart indicates that the process is controlled.

LSC = 74,014
74,015

74,010

74,005

74,000

73,995

73,990 LIC = 73,988

Sample Number

29
Interpretation of Control Charts X-
Bar and R
Before analyzing the behavior of the control charts X-
bar, it becomes necessary to check if the R cards are in
control condition.

If both the X-bar and R control charts show

non-random behavior, the best strategy is
initially eliminate the special causes of letters R.

Em alguns casos, isso automaticamente elimina o

non-random behavior of X-bar cards.

One must avoid analyzing the X-bar letters at all costs.

in isolation, when the R cards indicate a condition of
control output.

30
Interpretation of Control Charts X-
Barra and R

31
Interpretation of Control Charts X-
Bar and R
Cyclical Behavior

X-bar letter:
Systematic changes in the process environment
Temperature, operator fatigue, regular exchange of
operators and equipment

Abnormal fluctuation of operating conditions

equipment
Voltage, vapor pressure, among others.

Letters R:
Maintenance procedures and operator fatigue,
among others.

32
Interpretation of Control Charts X-
Bar and R

LSC

LIC

Sample Number

33
Interpretation of Control Charts X-
Bar and R
Mixing Behavior

It is checked when the points tend to be

to locate very approximately or slightly
beyond the limits of control.
Main Causes:
About control, due to frequent adjustments
process.
When products come from various sources
(for example, parallel tanks) are fed
in a common sampling withdrawal stream for
the monitoring of the process.

34
Interpretation of Control Charts X-
Bar and R

LSC

LIC

Sample Number

35
Interpretation of Control Charts X-
Bar and R
Level Change Behavior

Main causes:

Use of new operators, methods,

materials or equipment.

Changes in inspection methods or standards

Change in the motivation of the operators

36
Interpretation of Control Charts X-
Bar and R

LSC

LIC

Sample Number

37
Interpretation of X Control Charts
Barra and R
Trend Behavior
Continuous movement in one direction.

Main causes:
Gradual wear or deterioration of parts of
some critical process equipment

Human causes:
Operator fatigue
Inefficiency of supervision

38
Interpretation of Control Charts X-
Bar and R

LSC

LIC

Sample Number

39
Interpretation of Control Charts X-
Bar and R
Stratification Behavior

Artificial distribution of a group of points in

returns from the center line
Evident absence of natural variability in
process.

Potential cause:
Inadequate determination of limits
control.

40
Interpretation of Control Charts X-
Barra and R

LSC

LIC

Sample Number

41
Guide for Implementing Letters of
Control

I. Determination of quality characteristics to

will be controlled.
II. Determination of the location in the process where the
control letters must be implemented.
III. Selection of the types of control charts to be
used.
IV. Taking corrective actions for improvement of the
process, based on the results obtained by the
analysis of control charts.

42
Reflection

In the midst of a race for advantage

competitive, we should not look for a
improvement, we must consider the
implementation of a process of
continuous improvement

(Eliyahu M. Goldratt)

43