TM 720 Lecture 05: Variation Comparisons, Process Capability
Download
Report
Transcript TM 720 Lecture 05: Variation Comparisons, Process Capability
ENGM 720 - Lecture 05
Variation Comparisons,
Process Capability
4/8/2016
ENGM 720: Statistical Process Control
1
Assignment:
Reading:
•
•
Chapter 4 & 8
•
•
Finish reading through 4.3.4, 4.4 - 4.4.3, and CH 8 - 8.3
Begin reading 4.5
Chapter 5
•
Begin reading through 5.2, and 5.4
Assignments:
•
•
•
Obtain the Hypothesis Test (Chart &) Tables – Materials Page
Obtain the Exam Tables DRAFT – Materials Page
•
Verify accuracy as you work assignments
Access New Assignment and Previous Assignment Solutions:
•
Download Assignment 3 Instruction & Solutions
4/8/2016
ENGM 720: Statistical Process
Control
2
Comparison of Variances
• The second types of comparison are those that compare the
spread of two distributions. To do this:
• Compute the ratio of the two variances, and then compare the ratio
to one of two known distributions as a check to see if the magnitude
of that ratio is sufficiently unlikely for the distribution.
Definitely
Different
Probably
Different
Probably NOT
Different
Definitely NOT
Different
• The assumption that the data come from Normal distributions is very
important. Assess how normally data are distributed prior to
conducting either test.
4/8/2016
ENGM 720: Statistical Process
Control
3
Situation VII: Variance Test
With 0 Known
Used
when:
• existing comparison process has been operating without
much change in variation for a long time
Procedure:
• form ratio of a sample variance (t-distribution variable) to a
population variance (Normal distribution variable),
v = n - 1 degrees of freedom
2
0
n 1S 2
4/8/2016
2
0
ENGM 720: Statistical Process
Control
4
Situation VIII: Variance Test
With 0 Unknown
Use:
• worst case variation comparison process for when there is
not enough prior history
Procedure:
• form ratio of the sample variances (two 2-distributions),
v1 = n1 – 1 degrees freedom for numerator, and
v2 = n2 – 1 degrees freedom for the denominator
F0
4/8/2016
S 12
S
2
2
Note:
F1,v1,v 2
ENGM 720: Statistical Process
Control
1
F,v 2,v1
5
Table for Variance Comparisons
Decision on which test to use is based
on answering the following:
• Do we know a theoretical variance (2) or
•
should we estimate it by the sample variances
(s2) ?
What are we trying to decide (alternate
hypothesis)?
4/8/2016
ENGM 720: Statistical Process
Control
6
Table for Variance Comparisons
These questions tell us:
•
•
•
•
Four primary test statistics for variance
comparisons
•
•
•
What sampling distribution to use
What test statistic(s) to use
What criteria to use
How to construct the confidence interval
Two sampling distributions
Two confidence intervals
Six alternate hypotheses
Table construction
•
Note: F1-, v1, v2 = 1/ F, v2, v1
4/8/2016
ENGM 720: Statistical Process
Control
7
Grip Strength Example
True Corporate Training Example
•
How could grip strength vary among people in the SPC
training room?
Data collection to detect difference in dominant
hand mean between the left and right sides of
the training room
•
Expectations?
•
•
•
•
Significance Level?
Known parameters?
Best test?
Result?
• Direction of comparison?
4/8/2016
ENGM 720: Statistical Process
Control
8
Grip Strength Data Results
R-L Side, Equal Variance
Dominant Hand Means
Two-Sided Test at = .05
Comparison:
• HA: There is a difference
2
• L = x1 = 129.4, S1 = 2788,
Test: Is | t0 | > t.025, 52?
n1 = 34 people
• R = x2 = 104.0, S22 = 1225, • |1.91| > 2.009 - NO!
Keep the Null Hypothesis:
n2 = 20 people
• There is NOT a difference btwn
• Sp = 47.1, v = 52
L&R!
t0
x1 x 2 0
Sp
4/8/2016
1
1
n1 n 2
(n1 1) S12 (n2 1) S 22
Sp
n1 n2 2
ENGM 720: Statistical Process
Control
9
Grip Strength Data Results
R-L Side, No Assumptions
Dom. Hand Means
Comparison:
• L = x1 = 129.4, S1
t0
2
Two-Sided Test at = .05
Test: Is | t0 | > t.025, 51?
= 2788,
n1 = 34 people
• R = x2 = 104.0, S22 = 1225,
n2 = 20 people
• v = 51
2
S12 S 22
n n
2
1
v
x x2 0
2
2
1
S12
S 22
S12 S 22
n
n
1 2
n1 n 2
n1 1
n2 1
4/8/2016
• HA: There is a difference
• |2.12| > 2.009 - YES!
Reject the Null Hypothesis:
• There IS a difference btwn
L & R!
• Why is this wimpy test
significant when the other
wasn’t?
ANS: Check the equal
variance assumption!
ENGM 720: Statistical Process
Control
10
Grip Strength Data Results
Unknown σ0
Variances Comparison:
• S12 = 2788
• n1 = 34, v1 = 33
• S22 = 1225
• n2 = 20, v2 = 19
Two-Sided Test at = .10
Test: Is F0 > F.05, 33, 19?
• HA: There is a difference
• 2.276 > 2.07 - YES!
(Should also check F1– /2, 33, 19)
Reject the Null Hypothesis:
• There IS a difference in variance!
• At = .05, this test is just barely
F0
S
S
2
1
2
2
4/8/2016
F1 ,v1 ,v2
not significant
• (Should also have checked for
Normality with Normal Prob. Plot)
1
F ,v
2
,v1
ENGM 720: Statistical Process
Control
11
Statistical Quality Improvement
Goal: Control and Reduction of Variation
Causes of Variation:
•
Chance Causes / Common Causes
•
Assignable Causes / Special Causes
•
• In Statistical Control
• Natural variation / background noise
• Out of Statistical Control
• Things we can do something about - IF
we act quickly!
Both can cause defects – because specifications are
often set regardless of process capabilities!
4/8/2016
ENGM 720: Statistical Process
Control
12
Process Capability
Process
Capability Analysis (PCA)
•Is only done when the process is in a state of
Statistical Control
• Meaning:
NO SPECIAL CAUSES are present
•Process does not have to be centered to do
PCA
• Yield will improve if process is centered, but the value
is in knowing what / where to improve the process
•PCA is done periodically when the process has
been operating in a state of statistical control
• Allows for measuring improvement over time
• Allows for marketing your competitive edge
4/8/2016
ENGM 720: Statistical Process
Control
13
Process Capability - Timing
Process Capability Analysis is performed
when there are NO special causes of
variability present – ie. when the process is
in a state of statistical control, as
illustrated at this point.
Improving Process Capability and
Performance
Continually Improve the
System
Characterize Stable Process
Capability
Head Off Shifts in Location,
Spread
Time
Identify Special Causes - Bad
(Remove)
Identify Special Causes - Good
(Incorporate)
Reduce Variability
Center the Process
LSL
0
4/8/2016
USL
ENGM 720: Statistical Process
Control
14
Process Capability
Process Capability is INDEPENDENT of product
specifications
• Most specifications are set without regard for process
capability
• However, understanding process capability helps the engineer to
set more reasonable specifications
• PCA reflects only the Natural Tolerance Limits of the process
• PCA is done by examining the process
• Histogram
• Normal Probability Plot
4/8/2016
ENGM 720: Statistical Process
Control
15
Natural Tolerance Limits
The
natural tolerance limits assume:
• The process is well-modeled by the Normal Distribution
• Three sigma is an acceptable proportion of the process
to yield
The
Upper and Lower Natural Tolerance Limits
are derived from:
• The process mean () and
• The process standard deviation ()
Equations:
UNTL 3
LNTL 3
4/8/2016
ENGM 720: Statistical Process
Control
16
Natural Tolerance Limits
1 :68.26% of the total area
2 :95.46% of the total area
3 :99.73% of the total area
-3
or
LNTL
-2
-
+
+2
+3
or
UNTL
The Natural Tolerance Limits cover 99.73% of the process output
4/8/2016
ENGM 720: Statistical Process
Control
17
PCA: Histogram Construction
Verify rough shape and location of histogram
Quickly confirm applicability prior to statistical analysis
• Symmetric (roughly bell-shaped)
• Mean median mode
• Often hard to distinguish a Normal Distribution from a t-Distribution
• Sometimes even a Normal distribution doesn’t look normal
• More data and columns (bins) can make a difference
Verify location of process with respect to Specifications
• Quick inspection will show what to do to improve the process
4/8/2016
ENGM 720: Statistical Process
Control
18
PCA: Normal Probability Plot
A
Normal Plot better clarifies whether the
distribution is Normal by a visual inspection for:
• Non-random patterns (non-Normal)
• Fat Pencil Test (Normal if passes)
C
u
m
C
u
m
C
u
m
F
r
e
q
F
r
e
q
F
r
e
q
X
4/8/2016
X
ENGM 720: Statistical Process
Control
X
19
PCA: Parameter Estimation
The Normal Plot mid-point estimates the process mean
The slope of the “best fit” line for the Normal Plot estimates
the standard deviation
• Choose the 25th and 75th percentile points to calculate the slope
The Histogram mode should be close to the mean
The range/d2 (from Histogram) should be close to the
standard deviation
• Can also estimate
standard deviation by subtracting
th
50th percentile
from the 84 percentile of the Histogram
4/8/2016
ENGM 720: Statistical Process
Control
20
Process Capability Indices
Cp:
•Measures the potential capability of the current
process - if the process were centered within
the product specifications
•Two-sided Limits:
USL LSL
Cp
6
•One-sided Limit:
Cpu
4/8/2016
USL
3
ENGM 720: Statistical Process
Control
LSL
Cpl
3
21
Cp Relation to Process Fallout
Cp
Ratio
0.50
0.60
0.80
1.00
1.20
1.40
1.50
1.60
1.80
2.00
Two-Sided Specification Fallout
(ppm)
133614
71861
16395
2700
318
27
7
2
0.06
0.0018
One-sided Specification Fallout
(ppm)
66807
35931
8198
1350
159
14
4
1
0.03
0.0009
Recommended Minimum Ratios:
(D. C. Montgomery, 2001)
• Existing Process
1.25 (1-sided)
• Existing, Safety / Critical Parameter 1.45
• New Process
1.45
• New, Safety / Critical Parameter
1.60
4/8/2016
ENGM 720: Statistical Process
Control
1.33 (2-sided)
1.50
1.50
1.67
22
Process Capability Indices
Cpk:
•Measures actual capability of current process at its’ current location with respect to product
specifications
•Formula:
pk
pu
pl
C
min( C , C )
Where:
Cpu
4/8/2016
USL
3
LSL
Cpl
3
ENGM 720: Statistical Process
Control
23
Process Capability Indices
Regarding
Cp and Cpk:
•Both assume that the process is Normally
distributed
•Both assume that the process is in Statistical
Control
•When they are equal to each other, the
process is perfectly centered
•Both are pretty common reporting ratios
among vendors and purchasers
4/8/2016
ENGM 720: Statistical Process
Control
24
Process Capability Indices
Two
very different processes can have
identical Cpk values, though:
•because spread and location interact in Cpk!
LSL
4/8/2016
USL
ENGM 720: Statistical Process
Control
25
Process Capability Indices
Cpm:
•Measures the current capability of the
process - using the process target center
point within the product specifications in the
calculation
•Formula: Cpm USL LSL
6 2 ( T )2
Where target T is:
4/8/2016
1
T (USL LSL )
2
ENGM 720: Statistical Process
Control
26
Process Capability Indices
Cpkm:
•Similar to Cpm - just more sensitive to
departures from the process target center point
•Not really in very common use
•Formula:
C pkm
4/8/2016
C pk
T
1
2
ENGM 720: Statistical Process
Control
27
Questions & Issues
4/8/2016
ENGM 720: Statistical Process
Control
28