Statistical Process Control (SPC)

QUALITY CIRCLE FORUM OF INDIA

VADODARA CHAPTER

612, Bluechip Business Centre, Near Vadodara Stock Exchange,
Sayajigunj, Vadodara 390 005
Telefax:0265-2361184, Phone: 0265-6596140
E-mail : [email protected]
PRESENTED BY
QCFI/SPC-Eng/SCP/1 of 28/12-12-2006
Statistical Process Control (SPC)
1. OBJECTIVES OF SCP

1.1 General Objectives :
One or more of the following objectives
To increase knowledge about the process
To Steer a process to behave in the desired
way
To reduce variation in final product
To improve performance of a process

QCFI/SPC-Eng/SCP/2 of 28/12-12-2006
Statistical Process Control (SPC)
1.2 Specific objectives:
Considerations :
The market being served
Nature of the product
Process technology
Customer needs
Effective implementation of SPC reduces cost
and increase profile How?

QCFI/SPC-Eng/SCP/3 of 28/12-12-2006
Statistical Process Control (SPC)
1.2 Specific objectives: (Contd...)
By managing the process more economically aiming
at greater consistency and improvement.
By reducing variation around a target value in either
a final product or process output parameter.
By transferring variation in an in-process product
parameter to a controllable or manipulated process
variable.
By providing signals and evidence of how a process is
behaving and how it is likely to behave.
QCFI/SPC-Eng/SCP/4 of 28/12-12-2006
Statistical Process Control (SPC)
1.2 Specific objectives: (Contd...)
By assessing and quantifying what quality and
consistency level the process is currently capable of
producing.
By identifying when and where to look for special
causes and / or make preventive process adjustments
and when not to.
By pointing to potential root causes or failure modes
and their sources.
By identifying sources of poor yield and variability
and detecting special causes for the same.
QCFI/SPC-Eng/SCP/5 of 28/12-12-2006
Statistical Process Control (SPC)
1.2 Specific objectives: (Contd...)
By increasing speed of detection and reduce trouble
shooting costs.
By controlling and / or reducing common cause
variation through process design changes and other
system changes to procedures.
By increasing knowledge of how the system of causes
affects the process, to help make improvements to the
process.
SPC IS A STRATEGY FOR DEFECT PREVENTION
QCFI/SPC-Eng/SCP/6 of 28/12-12-2006
Statistical Process Control (SPC)
2. CONCEPTS :
In implementing SPC we need to understand following
concepts / terms
Process
Variation
Distribution
Stability
Common causes & special causes
Process control
Process capability
QCFI/SPC-Eng/SCP/7 of 28/12-12-2006
Statistical Process Control (SPC)
2.1 Process :
Transformation
Activities
People
Equipment
Materials
Methods
Documents /
Specifications
Products /
Services
A process is a set of interrelated or interacting activities
which transform inputs into outputs.
QCFI/SPC-Eng/SCP/8 of 28/12-12-2006
Statistical Process Control (SPC)
2.1 Process : (Contd...)
Process performance has to be monitored, to help
timely corrective actions to control the process.
Information about process performance can be had
from process characteristics e.g. Temperature, Cycle
time, Feed rate etc.
Target values for process characteristics are to be set
and monitoring is to be done to see whether actual
process confirms to these targets, to what extent.
Timely action is then taken if needed.
QCFI/SPC-Eng/SCP/9 of 28/12-12-2006
Statistical Process Control (SPC)
2.1 Process : (Contd...)
Action on the process Vs Action on the output

Action on the process is generally most economical
for defect prevention strategy. This helps in stability
and control of variation of process output within
acceptable limits.

Action on the output is generally least economical
when used to defect and to correct out of
specification product.
QCFI/SPC-Eng/SCP/10 of 28/12-12-2006
Statistical Process Control (SPC)
2.2 Variation :
Basic to process control is the concept of
variation.

Not two products or characteristics are exactly
alike because process contains many sources
of variability.

The difference among products may be large
or immeasurably small, but they always exist.
QCFI/SPC-Eng/SCP/11 of 28/12-12-2006
Statistical Process Control (SPC)
2.2 Variation: (Contd...)
Example: A machined shaft diameter; potential
sources of variation.
Machine : Clearances, Bearing wear
Tool : Strength, Rate of wear
Material : Diameter, Hardness
Operator : Part feed, Speed accuracy of centering
Maintenance : Lubrication, Replacement of worn out
: parts
Environment : Temperature consistency of power supply.
QCFI/SPC-Eng/SCP/12 of 28/12-12-2006
Statistical Process Control (SPC)
2.2 Variation : (Contd...)
Variation over time :
Some sources of variation in the process cause short
term piece to piece difference.
Other sources of variation tend to cause changes in
the output only over a long period of time either
gradually or stepwise or irregularly.
Hence time period and conditions over which
measurement are taken will affect total variation.
QCFI/SPC-Eng/SCP/13 of 28/12-12-2006
Statistical Process Control (SPC)
2.2 Variation : (Contd...)
View of Variation:
From requirements point of view, products
within specification limits are acceptable and
those beyond specification limits are not
acceptable.
To manage any process and reduce variation it
must be traced to sources. The causes of
variation are distinguished as COMMON
causes and SPECIAL causes.
QCFI/SPC-Eng/SCP/14 of 28/12-12-2006
Statistical Process Control (SPC)
2.3 Distribution :
While individual measured values may all be different, as a
group they tend to exhibit a pattern. This is called
DISTRIBUTION.
It can be desired by:
Location (Process level or centering)
Spread or dispersion (Range of values from smallest to
largest.)
Shape (Pattern of variation, whether symmetrical or
skewed etc.)
Any changes either in location or spread or dispersion or
shape indicates changes of process performance. No change
in these implies stability and therefore repeatability of the
process.
QCFI/SPC-Eng/SCP/15 of 28/12-12-2006
Statistical Process Control (SPC)
2.3 Distribution : (Contd...)

Changes In Distribution Location

A: Original Process B: Process Level Shifted


A B
Shift in Process Level
QCFI/SPC-Eng/SCP/16 of 28/12-12-2006
Statistical Process Control (SPC)
A
B
Change in Process Spread / Range
2.3 Distribution : (Contd...)

Changes In Distribution Spread
A: Original Process B: Increase in spread with
same location

QCFI/SPC-Eng/SCP/17 of 28/12-12-2006
Statistical Process Control (SPC)
A
B
Change in Shape Pattern of Variation
2.3 Distribution : (Contd...)

Changes In Distribution Shape
A: Original Symmetrical Pattern
B: Pattern is skewed
QCFI/SPC-Eng/SCP/18 of 28/12-12-2006
Statistical Process Control (SPC)
When distribution remains unchanged over a period of
time, then process is said to be stable and Repeatable
(When only common causes are present in the process).
2.4 Stability :
Time
Target
QCFI/SPC-Eng/SCP/19 of 28/12-12-2006
Statistical Process Control (SPC)
2.5 Common Causes :
They are minute and many and individually may not be
measurable.
The resulting pattern under the influence of common
causes is called State of Stastical Control or just
In Control.
The advantages of maintaining a state of statistical
control are:
Variation is restricted.
Process is repeatable.
Quality of future production can be predicted.
QCFI/SPC-Eng/SCP/20 of 28/12-12-2006
Statistical Process Control (SPC)
2.5 Special Causes :
When they occur, they make the overall process
distribution change. Unless they are arrested they will
continue to affect the process output in unpredictable
ways.

Changes in process pattern due to special causes can be
either detrimental or beneficial. When detrimental they
need to be identified and eliminated. When beneficial
they need to be continued by making them a permanent
part of the process.
QCFI/SPC-Eng/SCP/21 of 28/12-12-2006
Statistical Process Control (SPC)
2.6 Process Control :
This is the state where only common causes are present.
The proof of this is when the pattern of variation
conforms to the statistical normal distribution.

It involves continuous monitoring of the process for
special causes and eliminating them.

Evidence of special causes is provided by the changes in
process distribution.
QCFI/SPC-Eng/SCP/22 of 28/12-12-2006
Statistical Process Control (SPC)
2.7 Process Capability :
The ability of a process to produce within the variation
permitted by tolerance is called process capability.

The process control and process capability are
different.

Process control is basic and is required for stability
of a process.
Process capability is containment of common cause
variation substantially within the range provided by
the tolerance.
QCFI/SPC-Eng/SCP/23 of 28/12-12-2006
Statistical Process Control (SPC)
2.7 Process Capability : (Contd...)

LSL USL
Process is in Control (Stable) and Capable
QCFI/SPC-Eng/SCP/24 of 28/12-12-2006
Statistical Process Control (SPC)
2.7 Process Capability : (Contd...)
LSL USL
Process is in Control but not
satisfactory (Capable)
QCFI/SPC-Eng/SCP/25 of 28/12-12-2006
Statistical Process Control (SPC)
2.7 Process Capability : (Contd...)

LSL USL
Process is capable but not in control because
process level is not properly centered.
QCFI/SPC-Eng/SCP/26 of 28/12-12-2006
Statistical Process Control (SPC)
2.7 Process Capability : (Contd...)

LSL USL
Process not in control and not capable
QCFI/SPC-Eng/SCP/27 of 28/12-12-2006
Statistical Process Control (SPC)
Process
Capability
Process
In Control
Process
Out of Control
Capable Acceptable Not acceptable
Not capable Not acceptable Not acceptable
QCFI/SPC-Eng/SCP/28 of 28/12-12-2006