Unformatted text preview: STATISTICAL QUALITY MANAGEMENT STSTISTICAL PROCESS
CONTROL
KRIPA SHANKER Ph D (Cornell) FNAE, FIE(I), FITEE(I), LMISTE, LMIIIE, MORSI, MORSA, SMIIE Visiting Faculty Department of Mechanical Engineering
Indian Institute of Technology (BHU)
Varanasi
Former Emeritus Fellow
Industrial and Management Engineering Department
Indian Institute of Technology Kanpur
Former Vice Chancellor, Uttar Pradesh Technical University
Lucknow
Former Deputy Director, Indian Institute of Technology 1
Kripa Shanker
Kanpur 2
Prof Kripa Shanker Mechanical Engineering IIT (BHU) Varanasi STATISTICAL PROCESS CONTROL All processes are subject to a certain
degree of variability - (a fact about the
nature !)
• Causes of Variation : Common (natural) (chance)
(random) Special (assignable)
• Control Chart
(Walter Shewhart, 1920s, Bell
Laboratories) Separates the two causes of
variation.
Kripa Shanker 3 STATISTICAL PROCESS CONTROL • PROCESS CONTROL
A process is said to be operating in
statistical control when the source
of variation is only the common
(natural) (chance) (random)
causes.
Process must first be brought into
statistical control by detecting and
eliminating special (assignable)
causes.
Kripa Shanker 4 STATISTICAL PROCESS CONTROL • PROCESS CONTROL
Once the process is under control,
then
its performance is predictable,
its ability to meet the customer
expectations can be assessed.
The objective of a process control
system is to provide a statistical signal
when assignable causes of variation are
present.
Kripa Shanker 5 STATISTICAL PROCESS CONTROL
NATURAL VARIATIONS
(common) (chance) (random) variations affect almost every
• Natural
process and, are to be expected.
are many sources of natural
• There
variation that occur within a process (even when the process is under
statistical control !).
Natural variations form a pattern (can
be
described
as
distribution)
As
long
as
the
distribution
(output • measurements)
remains
within
specified
limits, the process is said
to be “in control“,6
Kripa Shanker STATISTICAL PROCESS CONTROL ASSIGNABLE VARIATIONS variations in a process can
• Assignable
be traced to a specific reason :
•
•
•
•
• machine wear,
mis-adjustment in equipment,
fatigued or untrained worker,
new batches of materials),
etc. Kripa Shanker 7 STATISTICAL PROCESS CONTROL QUALITY MANAGEMENT TASKS
Natural
and
Assignable
variations
distinguish two tasks :
(1) to ensure that the process is
capable of operating under control
(with only natural variations.)
(2) to
identify
and
eliminate
assignable variations so that the
process remains under control.
Kripa Shanker 8 STATISTICAL PROCESS CONTROL QUALITY MANAGEMENT TASKS
Measurements - Samples vs Individual
Because of natural and assignable variations,
statistical process control (SPC) uses statistics
(eg. average, range, etc.) of samples (often 4
to 8) as opposed to data on individual parts.
Individual pieces tend to be two erratic to
make trends quickly visible. Kripa Shanker 9 STATISTICAL PROCESS CONTROL CONTROL CHARTS
Control Chart (Walter Shewhart, 1920s, Bell
Laboratories) Separates the two causes of
variation.
→ A statistical tool to determine if a process is in
• Because
of
natural
and
assignable
control. variations, statistical process control (SPC)
uses statistics (eg. average, range, etc.) of
samples (often 4 to 8) as opposed to data on
individual parts.
• Individual pieces tend to be two erratic to
10
Kripa
Shanker
make trends quickly visible. STATISTICAL PROCESS CONTROL CONTROL CHARTS
Theory of Control Chart
X : Quality Characteristic (length, diameter, weight, surface finish, …..)
• UCL = Process Average + 3 Standard Deviations
• LCL = Process Average - 3 Standard Deviations
X
UCL = Upper Control Limit
+ 3X Process Average : Central Line - 3X
LCL = Lower Control Limit
TIME 11 Thank
You ...
View
Full Document
- Winter '15
- Faiz Hamid
