Understanding Control Charts in Quality Control
Uploaded by
ScribdTranslationsUnderstanding Control Charts in Quality Control
Uploaded by
ScribdTranslationsCONTROL LETTERS
CONTROL LETTERS
Control charts are the most powerful tool for analyzing variation in
the majority of the processes.
They have been successfully disseminated in several countries within a wide variety.
of situations for process control.
Control charts focus attention on the special causes of variation.
when they appear and reflect the magnitude of the variation due to the causes
common.
Common or random causes are due to the natural variation of the process.
Special or attributed causes, for example: a bad machine adjustment,
operator errors, defects in raw materials.
It is said that a process is under Statistical Control when it presents causes.
common only. When this happens, we have a stable process and
predictable.
When there are special causes, the process is out of Statistical Control; the
control charts detect the existence of these causes at the moment when they
and, which allows us to take actions at the moment.
VARIABLES AND ATTRIBUTES
In Quality Control, the term variable refers to
any "measurable" quality characteristic such as a length,
weight, temperature, etc. While it is referred to as an attribute to the
quality characteristics that are not measurable and that present
different states such as compliant and non-compliant or defective and
not defective
Depending on the type of quality characteristic to be controlled, so will be the
corresponding Control Chart that, therefore, are classified into
Control Charts for Variables and Control Charts for Attributes.
TYPES OF CONTROL CHARTS
CONTROL CHARTS BY VARIABLES
CONTROL CHARTS X–R (VARIABLES)
Step 1: Collect the data.
The data are the result of measuring the characteristics of the
product, which must be registered and grouped from the
in the following way:
A sample (subgroup) of 2 to 10 consecutive pieces is taken and
note the measurement results (it is recommended to take 5). Also
they can be taken at time intervals of ½ - 2 hrs., for
detect if the process can show inconsistency in brief periods
of time.
Samples of 20 to 25 subgroups are taken.
X-R CONTROL CHARTS (VARIABLES)
X-R CONTROL CHARTS (VARIABLES)
CONTROL CHARTS X–R (VARIABLES)
CONTROL CHARTS X–R ( VARIABLES)
X-R CONTROL CHARTS (VARIABLES)
X–R CONTROL CHARTS (VARIABLES)
CONTROL CHARTS X-R ( VARIABLES)
CONTROL CHARTS. PARAMETERS OF
STATISTICAL CONTROL
A set of decision rules is suggested for detection
non-random patterns in control charts. When
detect one of the following patterns one can say that
Action must be taken to correct the problem now.
that the process may be out of control
These control parameters are valid for any type
control chart that is being worked on
CONTROL LETTERS. PARAMETERS OF
STATISTICAL CONTROL
CONTROL LETTERS. PARAMETERS OF
STATISTICAL CONTROL
CONTROL LETTERS. PARAMETERS OF
STATISTICAL CONTROL
CONTROL LETTERS. PARAMETERS OF
STATISTICAL CONTROL
CONTROL CHARTS X–S (VARIABLES)
CONTROL CHARTS X–S (VARIABLES)
CONTROL CHARTS X–S (VARIABLES)
CONTROL CHARTS X–S (VARIABLES)
CONTROL CHARTS X–S (VARIABLES)
CONTROL LETTERS OF READINGS
INDIVIDUALS I–MR (VARIABLE DATA)
READING CONTROL LETTERS
INDIVIDUAL I–MR (VARIABLE DATA)
READING CONTROL LETTERS
INDIVIDUALS I–MR (VARIABLE DATA)
CONTROL LETTERS OF READINGS
INDIVIDUALS I–MR (VARIABLE DATA)
CONTROL CHARTS BY ATTRIBUTES
CONTROL CHARTS BY ATTRIBUTES
Any quality characteristic that can be classified in a binary way: 'meets or
"does not comply", "works or does not work", "passes or does not pass", etc., for control purposes.
process, will be considered as an attribute and a graph will be used for its control
Control by Attributes.
The acceptance criteria when using control charts for attributes must be
clearly defined and the procedure to determine if those criteria are being met
is to produce consistent results over time. This procedure consists of
operationally define what you want to measure. An operational definition consists of:
1st. A criterion that is applied to an object or a group
2nd. A test of the object or the group and
3rd. A decision, yes or no: The object or group meets or does not meet the criteria.
P Chart for Defective Units Fraction
(attributes)
The graph p measures the defective fraction or the defective pieces in the
process. It can refer to samples of 75 pieces, taken twice a day;
100% of the production for one hour, etc. It is based on the evaluation of
a feature (was the required part installed?) or of many
characteristics (was something wrong found when checking the electrical installation?). It is
important that each verified component or product is registered as
acceptable or defective (even if a piece has several specific defects)
it will only be registered once as defective).
P Chart for Defective Units Fraction
(attributes)
Step 1 - Frequency and sample size:
Set the frequency at which the data will be taken (hourly, daily, weekly). Short intervals
Sampling will allow for quick feedback to the process in the presence of problems.
Large sample sizes allow for more stable assessments of process development and are more
sensitive to small changes in its average. It is advisable to take equal sample sizes
Although this situation does not necessarily have to occur, the sample size should be greater than 30.
the size of the subgroups will be 25 or more.
Step 2 - Calculation of the defective percentage (p) of the subgroup:
Record the following information for each subgroup:
The number of inspected parts–n
The number of defective parts - np
Calculate the defective fraction (p) using: np
p
n
P Chart for Fraction of Defective Units
(attributes)
P Chart for Defective Units Fraction
(attributes)
Example
An aluminum can manufacturer records the number of defective parts, taking samples every
n = 50, with 30 subgroups. Create the control chart for the following data series.
obtained during the sampling.
P Chart for Defective Units Fraction
(attributes)
P Chart for Fraction of Defective Units
attributes
P chart for defective units fraction
(attributes)
Graph np–Number of defective items
The np chart is based on the number of defective items instead of the proportion of defective items. The
limits are calculated using the following formulas.
Graph np–Number of defective items
Graph np–Number of defective items
Graph C – for number of defects
It is used to determine the occurrence of defects in the inspection of a product unit.
Esto es determinar cuantos defectos tiene un producto. Podemos tener un grupo de 5 unidades de
product, 10 units, etc.
The control limits are calculated using the following formulas:
Graph C – for number of defects
Chart C – for number of defects
U-Defects per Unit Chart
Example
A company that manufactures personal computers wishes to establish a
control diagram of the number of defects per unit. The sample size
It is about five computers. The table shows the number of defects in
20 samples of 5 computers each. Establish the control chart.
U-Defects per Unit Chart
U-Defects per Unit Chart
U-Defects per Unit Graph