Statistical Process Control (SPC)

• SPC – A methodology which uses the basic graphical and statistical

tools to analyze, control and reduce variability within the process.

• SPC – Uses in manufacturing and service.

• SPC – It combined many tools such as run chart, control chart,

histogram, distribution and confidence intervals to provide workers
with powerful method to participate in the quality improvement
process.

• The basics concept of using SPC for the process improvement can
be applied to any type of work where the key measurement exhibit
variation.

• SPC – to avoid waste by not providing unusable output in the first

place. It is a preventive strategy which seems sensible-even
obvious-to most people.

• Elements of SPC: Variation, cause of variation, control chart,

process control and process capability.
 Understanding Variation
• “Variability is like a virus. Each process can infect the one it touches,
including the management of the process. Infected process are
noisy and produce noisy meeting” from Myron Tribus.
• Process contains many sources of variability.
• Example manufacturing:
• Diameter of the machine shaft, susceptible to potential variation
from machine (clearances, bearing wear), tools (strength, rate
of wear), material (diameter, hardness), operator (part feed,
accuracy on centering).
• Example Service:
• Time require to process an invoice could vary according to
people performing various steps, the reliability of any equipment
they used, accuracy and legibility of the invoice itself,
procedures top followed and volume of the others work in the
office.
• Source of variation in the process can cause short and long term
variation. Change over period of time.
• Parts within specification tolerances are accepted and out of
specification are not accepted.
• To manage the process and reduce variation, the output must be
measured and variations traced back to their source.
• Cause of the variation:
• Common cause: refer to sources of variation that produce
chance or natural variation. Present of common cause, the
output of a process forms a distribution that is predictable and
stable over time.
• Special cause: refer to any factors causing variation that cannot
be adequately explained by any single distribution of the process
output. Present of the special cause, the output of the process is
in unpredictable ways.
•
 Process Control vs Process Capability
• Process Control
• A process is in state of statistical control when common causes are the
only source of variation
• But in natural process, most of the process are not in state of statistical
control.
• In order to achieve process in control, we have to eliminate all the
special causes from process until the mean and standard deviation are
not shifting and the process forms a stable distribution over time.
• Process Capability
• Is a measurement to tell how well the process respect to the output
specifications.
• Process capability is determined by the total variation that comes from
common causes.
• Capability represents the performance of the process itself as
demonstrated when the process is being operated in a state of
statistical control.
• Capability explain the proportion of output that will be within product
specification tolerances.
• Control and capability are often confused.
• Process control – voice of process.
• Process capability – voice of customer or term of specification from
customer.
• Process is in control does not mean that it is a good process.
• The goodness of a process is measured by its ” Process Capability”;
Cp and Cpk.
 Control Chart
• Control Chart is just a run chart which includes statistically
generated upper control limit (UCL) and lower control limit (LCL).
• These limits provide the user with bounds on the common cause
variability of the process output.
• Shewhart is the first person develop the control chart.
• Purpose of the control chart:
• To provide evidence of whether a process has been operating in
a state of statistical control and to signal the presence of special
causes of variation so that corrective action can be taken.
• Process improvement using control chart is an iterative procedure
repeating the fundamental phases of collection, control and
capacity.
• Generally to develop control chart:
• Data are gathered according to a careful plan.
• Data are used to calculate control limit.
• Once a process is in control, process capability can be measured.
• Collection
• The process is run and data for characteristic being studied are
gathered and converted to a form that can be plotted on a graph.
• These data might be the measured values of a dimension of a
machined piece, the number of flaws in a bolt of vinyl, railcar transit
times, number of bookkeeping error. Etc.
• Control
• Calculate the control limit based on data from the output of the
process.
• Draw the control limit UCL and LCL in the graph.
• The data are then compare with the control limit to see whether the
variation is stable and appears to come only from common causes.
• If special cause of variation are evident, operation of the process is
studied to determine what is affecting the process.
• After actions have been taken, further data are collect, control limit
are recalculate if necessary. This improvement will be continue until
all the special causes are eliminated.
•
• Capability
• After all special causes have been corrected and the process is
running in statistical control then process capability can be assessed.
• If variation from common causes is excessive, the process will not
produce output that consistently meets customer need.
• The process itself must be investigate and management action must
be taken to improve the system or reduce the variation of common
causes.