Statistical Process Control
L-28
Contents
1) Process Variability and Process Capability
a) Process Variations
b) Process Capability and Tolerances
2) Control Charts
a) Control Charts for Variables
b) Control Charts for Attributes
c) Interpreting the Control Charts
3) Other SPC Tools
a) Histogram
b) Pareto Charts
c) Check Sheets
d) Defect Concentration Diagrams
e) Scatter Diagrams
f) Cause and Effect Diagrams
4) Implementing Statistical Process Control
a) Elements of a successful SPC Program
b) Accomplishing a Quality Improvement Project
Statistical Process Control
SPC involves the use of various methods to measure and analyze a process.
The overall objectives of SPC are to (i) improve the quality of the process output (ii) reduce
process variability and achieve process stability (iii) Solve processing problems.
There are seven principal methods or tools used in SPC, they are called as magnificent
seven.
(i) Control charts
(ii) Histograms
(iii) Pareto charts
(iv) Check sheets
(v) Defect concentration diagram
(vi) Scatter diagrams
(vii) Cause and effect diagrams
1) Process variability and process capability
a) Process Variations
(i) Random: result from intrinsic variability in the process, no matter how well designed or
well controlled it is (ii) assignable: Something has occurred in the process that is not
accounted for by random variations. Reasons for assignable variations include operator
mistakes, defective raw materials, tool failures and equipment malfunctions
b) Process capability and Tolerances
Process capability relates to the normal variations inherent in the
output when the process is in statistical control.
PC = µ± 3σ
Where PC = process capability; µ = process mean ; σ = standard
deviation of the process
Assumptions underlying this definition are : (1) the output is normally
distributed (2) Steady state has been achieved and the process is in
statistical control. Under these assumptions 99.73% of the parts
produced will have output values that fall within ±3.0σ
Sample Standard Deviation
2) Control Charts
Control Chart is a graphical technique in which statistics computed
from measured values of a certain process characteristic are plotted
over time to determine if the process remains in statistical control
a) Control Charts for Variables
A process that is out of statistical control manifests this condition in the
form of significant changes in (i) Process mean (ii) Process variability.
Corresponding to these possibilities, there are two principal types of control
charts for variables (1) X –bar chart (2) R chart
The following procedure is used to construct the center, LCL and UCL for
each chart
(i) Compute the mean x-bar and range R for each of the m samples
(ii) Compute the grand mean x-bar, which is the mean of the x-bar values
for the m samples. This will be the center for the x-bar chart
(iii) Compute R-bar which is the mean of the R values for the m samples. This
will be the center for the R chart
(iv) Determine the UCL and LCL for the X-bar and R charts.
Contd.
Contd.
b) Control Charts for Attributes
Control charts for attributes monitor the number of defects present
in the sample or the fraction defect rate as the plotted statistic.
Examples of these kinds of attributes include: number of defects per
automobile, fraction of nonconforming parts in a sample, existence or
absence of flash in plastic molding, and number of flaws in a roll of
sheet steel.
The two principal types of control charts for attributes are (1) the
P-chart, which plots the fraction defect rate in successive samples (2)
the C-chart, which plots the number of defects, flaws or other
non-conformities per sample.
Contd.
Contd.
C) Interpreting Control Charts
Montgomery lists a set of specific indicators that a process is likely to be
out of statistical control and that corrective action should be taken. These
indicators are:
(i) One point that lies outside the UCL or LCL
(ii) Two out of three consecutive points that lie beyond ±2σ on one side of
the center line of the control chart
(iii) Four out of five consecutive points that lie beyond ±1σ on one side of
the center line of the control chart
(iv) Eight consecutive points that lie on one side of the center line
(v) Six consecutive points in which each point is always higher that its
predecessor or six consecutive points in which each point is always lower
than its predecessor.
Contd.
Control charts serve as the feedback loop in SPC
3) Other SPC Tools
a) Histogram
b)Pareto Charts
C) Check Sheets
Contd.
Following types of check sheets can be distinguished
(i) Process distribution check sheet
(ii) Defective item check sheet
(iii) Defect location check sheet
(iv) Defect factor check sheet
d) Defect Concentration diagrams
e) Scatter Diagrams
f) Cause and effect diagram
4) Implementing Statistical Process Control
a) Elements of a successful SPC Program
(i) Management commitment and leadership
(ii) Team approach to problem solving
(iii) SPC training for all employees
(iv) Emphasis on continuous improvement
(v) A recognition and communication system
b)Accomplishing a Quality Improvement Project