|
See Average Run Length
The number of points that,
on average, will be plotted on a control
chart before an out of control condition
is indicated (for example a point plotting
outside the control limits).
If the process is in control:
If the process is out of control:
Where a
is the probability of a Type I error and
b the
probability of a Type II error.
A procedure used to track
a process against time for the purpose of
identifying 'special causes', or sporadic
or chronic problems in a timely fashion.
The control chart has control limits, usually
at three standard deviations from the mean.
Special causes are indicated by points plotting
outside the control limits, or various other
patterns of points:
An "In-Control" process is one
that is free of special causes of variation.
Control charts are used to monitor processes
to check that they remain in-control, and
to give timely notification of a special
cause.
Note that "In-Control" does not
necessarily mean that a process is performing
at an acceptable level. It is quite possible,
and common, for a process to be stable and
free of special causes but producing output
of unacceptable quality.
Samples should be collected so that, if
special causes are present, the chances
for variation within subgroups is minimized
but the chances for variation between subgroups
is maximized.
The items within a subgroup should be collected
closely together, and may be consecutive
items. The items within a subgroup should
not be collected across a shift change,
or across any other process change that
may be associated with a special cause.
Shewhart invented control charts and control
charts are often called Shewhart Control
Charts. See Control Chart.
Short Run SPC is used in batch production
when the runs of any one item are too short
to run regular SPC.
The basic idea is to measure the deviation
from the standard size. For example if bars
of varying diameter are machined the control
charts would monitor the deviations from
the target bar diameter, rather than the
bar diameter.
The snag with this may be that the variation
may vary from part to part. In this case
the deviation from the nominal may be standardized.
This is done by dividing by the mean range.
An alternative is to use the ZmR chart.
In this case the deviation from the target
value is divided by the Standard Deviation.
See Statistical Process Control
| Statistical
Process Control |
|
Statistical Process Control refers to use
of statistical methods to monitor processes
(production or business/transactional) to
achieve timely identification of changes
to the process.
All processes are subject to variation.
Statistical Process Control involves:
- making sure the process is stable, a
good indicator of this is that it approximates
a normal distribution
- using statistical methods to identify
'special causes', that indicate a departure
from stability
The main tool of SPC is the control chart.
The rules from the Western Electric Company
handbook are often used to signify 'special
causes' on control charts:
|
