Professional Documents
Culture Documents
Control (SPC)
Praful Mehta
Agenda
• Control Charts
• Attribute Charts
• Case Study
2
Control Charts (SPC)
3
History of Control Charts
Developed in 1920’s
By Dr. Walter A. Shewhart
Shewhart worked for Bell Telephone Labs
4
What Is a Control Chart?
5
Why Use Control Charts?
6
Uses for Control Charts
Separate common cause variation from special cause
variation so that management can respond appropriately to
find root causes.
Understand and predict process capability.
Measure whether intentional changes had the desired result.
Find root cause(s).
Monitor key processes to maintain the gains.
7
R&S
Control chart functions
What’s causing variability?
Control charts are powerful aids to understanding the performance of a process over time.
8
Common-Cause Variation
Common causes:
Represent everyday, random variation in a process
Are a part of the process
Contribute to output variation because they themselves vary
Produce predictable levels of variation over time
Produce all of the variation in a stable process
9
R&S
Special-Cause Variation
Special causes:
Are not usually present
May come and go sporadically; may be temporary
or long-term
Appear under some particular circumstance
Are not predictable
Contribute to variation in an unstable process
10
Example: Special-Cause Variation
Equipment
Tool wear, machine vibration, …
Material
Raw material quality
Environment
Temperature, pressure, humidity
Operator
Operator performs- physical & emotional
12
Types Of Control Chart
There are two types of control charts that we deal with.
Variables Control Charts These charts are applied to data that follow a continuous
distribution.
Attributes Control Charts These charts are applied to data that follow a discrete
distribution.
Attributes Data:- Data that can be classified into one of several categories or
classifications is known as attribute data. Classifications such as conforming and
nonconforming are commonly used in quality control. Another example of
attributes data is the count of defects.
13
Types Of Attributes Control Chart
p chart :- This chart shows the fraction of nonconforming or defective
product produced by a process. It is also called the control chart for fraction
nonconforming.
np chart:- This chart shows the number of nonconforming. Almost the same
as the p chart.
14
Types Of Variables Control Chart
X-bar chart
In this chart the sample means are plotted in order to control the mean
value of a variable (e.g., size of piston rings, strength of materials, etc.).
R chart
In this chart, the sample ranges are plotted in order to control the
variability of a variable.
S chart
In this chart, the sample standard deviations are plotted in order to control
the variability of a variable.
15
Statistical Process Control
Measurement data
Attributes (yes/no)
Counts
? ? ?
Subgroup size Subgroups Subgroup size Subgroup size Subgroup size Subgroup size Subgroup size
constant and constant changes constant changes constant changes
equal to 1
Centerline
UCL X 3
shows where the process average is centered or the central
LCL X 3of the data
tendency
standard deviation
Upper control limit (UCL) and Lower control limit (LCL)
describes the process spread
19
Why 3 Sigma?
95%
99.74%
21
Some Terms
22
Control Limits V/S Specification Limits
Control Limits are used to determine if the process is in a
state of statistical control (i.e., is producing consistent
output). statistical
Customer
needs
1 7 13 19 25 31 37 43 49 55 61
Time Order
23
Number and Size of Sample
Shewhart gave the following rule of thumb:
"It has also been observed that a person would seldom if ever be
justified in concluding that a state of statistical control of a given
repetitive operation or production process has been reached until he
had obtained, under presumably the same essential conditions, a
sequence of not less than twenty five samples of size four."
24
The Control Chart Method
R Control Chart:
UCL = D4 x Rmean
LCL = D3 x Rmean
CL = Rmean
25
Elements of a Control Chart Step 1
26
Constructing an X-Bar & R chart Step 2
27
Constructing an X-Bar & R chart Step 3
28
Constructing an X-Bar & R chart Step 4
29
Constructing an X-Bar & R chart Step 5~10
30
Constructing an X-Bar & R chart Step 9
31
Example: Control Charts forfill Variable
Time to Data
customer data
Sample 1 2 3 4 5 X R
1 5.02 5.01 4.94 4.99 4.96 4.98 0.08
2 5.01 5.03 5.07 4.95 4.96 5.00 0.12
3 4.99 5.00 4.93 4.92 4.99 4.97 0.08
4 5.03 4.91 5.01 4.98 4.89 4.96 0.14
5 4.95 4.92 5.03 5.05 5.01 4.99 0.13
6 4.97 5.06 5.06 4.96 5.03 5.01 0.10
7 5.05 5.01 5.10 4.96 4.99 5.02 0.14
8 5.09 5.10 5.00 4.99 5.08 5.05 0.11
9 5.14 5.10 4.99 5.08 5.09 5.08 0.15
10 5.01 4.98 5.08 5.07 4.99 5.03 0.10
50.09 1.15
32
Calculation
From Table :
Sigma X-bar = 50.09
Sigma R = 1.15
m = 10
Thus;
X-Double bar = 50.09/10 = 5.009 cm
R-bar = 1.15/10 = 0.115 cm
33
Trial control limit
5.10
UCL
5.08
5.06
5.04
X bar
5.02
5.00 CL
4.98
4.96 LCL
4.94
0 1 2 3 4 5 6 7 8 9 10 11
Subgroup
35
R Chart
0.25 UCL
0.20
Range
0.15
CL
0.10
0.05
LCL
0.00
0 1 2 3 4 5 6 7 8 9 10 11
Subgroup
36
Run Chart 6.70
6.65
M e a n , X -b a r
6.60
6.55
6.50
6.45
6.40
6.35
6.30
0 5 10 15 20 25
Subgroup number
0.35
0.3
0.25
R a nge , R
0.2
0.15
0.1
0.05
0
0 5 10 15 20 25
Subgroup number
37
Reading and interpreting of Control
Charts
Chart Description Example #1 Example #2 Interpretation
The process is stable, not changing.
UCL 20 UCL 20
Chart points do not form a particular pattern Doesn’t necessarily mean to leave the
Process
AND lie within the upper and lower chart process alone. May be opportunities to
In X 15 X 15
39
Interpreting X-bar & R Charts Rule 1
40
Interpreting X-bar & R Charts Rule 2
41
Interpreting X-bar & R Charts Rule 3
42
Interpreting X-bar & R Charts Rule 4
43
Control Chart Rules – examples of the rationale
44
Variable control chart alarms
• 2 out of 3 points > 2 standard deviations from center line (same side)
• 4 out of 5 points > 1 standard deviation from center line (same side)
• 8 points in a row > 1 standard deviation from center line (either side)
45
Statistically Analysis- Hourly Output
Data Points: 315
Findings:
• Average Hourly output
improved from 124 to 156
numbers.
46
Exercise
47
Answer
48
Factors
Appendix: Determining Control Limits for x-bar and R-Charts
SAMPLE SIZE FACTOR FOR x-CHART FACTORS FOR R-CHART
n A2 D3 D4
70 UCL=70.78
Degrees (F)
_
60 X=60.36
50 LCL=49.94
9:00 9:30 10:00 10:30 11:00 11:30 12:00 12:30 13:00 13:30 14:00 14:30 15:00
Time
UCL=12.80
12
Moving Range
6
__
MR=3.92
3
0 LCL=0
9:00 9:30 10:00 10:30 11:00 11:30 12:00 12:30 13:00 13:30 14:00 14:30 15:00
Time
50
Exercise: Interpreting I-MR Charts
Case 2: QC Lab TAT
A DMAIIC team was chartered to reduce QC lab reduce turn-around-time (TAT) to less than 10 days. As a first
step, in late August the team removed the non-value-adding activities.
14 _
X=13.384
Days
12
LCL=11.488
10
3/8/05 3/29/05 4/19/05 5/10/05 5/31/05 6/21/05 7/12/05 8/2/05 8/23/05 9/13/05 10/4/05
Week
UCL=2.329
2.0
Moving Range
1.5
1.0
__
MR=0.713
0.5
0.0 LCL=0
3/8/05 3/29/05 4/19/05 5/10/05 5/31/05 6/21/05 7/12/05 8/2/05 8/23/05 9/13/05 10/4/05
Week
51
Exercise: Interpreting I-MR Charts
Case 3: Test Volume
One skeptic on the QC DMAIIC team thought that the improvement in TAT was a result of a significant reduction
in volume.
26
# Tests
_
X=24.469
24
22
LCL=21.723
3/8/05 3/29/05 4/19/05 5/10/05 5/31/05 6/21/05 7/12/05 8/2/05 8/23/05 9/13/05 10/4/05
Week
UCL=3.373
3
Moving Range
__
1 MR=1.032
0 LCL=0
3/8/05 3/29/05 4/19/05 5/10/05 5/31/05 6/21/05 7/12/05 8/2/05 8/23/05 9/13/05 10/4/05
Week
52
Exercise: Interpreting I-MR Charts
Case 4:Marketing Costs
The director of marketing was reviewing her expenses for the past two years. Based on the average, she had
budgeted the cost of $105,000 each month, but last month expenses were $117,000. She asked the staff to
explain why the costs were so much higher last month.
120000
Dollars
_
X=104636
100000
80000 LCL=81542
r r l ct b ar r l ct
04 Fe b Ma Ap Ma
y
Ju
n Ju g
Au Se
p
O No
v c 5
De n 0 F e M Ap Ma
y
Ju
n Ju g p
Au Se O No
v
De
c
n
Ja Ja
Month
30000
UCL=28371
Moving Range
20000
10000 __
MR=8683
0 LCL=0
Jan 04 Mar May Jul Sep Nov Jan 05 Mar May Jul Sep Nov
Month
53
Attribute Charts
Control Charts for Attributes
p-charts
uses portion defective in a sample
c-charts
uses number of defects in an item
55
p-Chart
UCL = p + zp
LCL = p - zp
z = number of standard deviations from process
average = 3
p = sample proportion defective; an estimate of
process average
p = standard deviation of sample proportion
p(1 - p)
p =
n
56
p-Chart Example
NUMBER OF PROPORTION
SAMPLE DEFECTIVES DEFECTIVE
1 6 .06
2 0 .00
3 4 .04
: : :
: : :
20 18 .18
200
57
p-Chart Example (cont.)
total defectives
p= = 200 / 20(100) = 0.10
total sample observations
58
p-Chart Example (cont.)
0.20
0.16
0.12
p = 0.10
0.10
0.08
0.06
0.04
2 4 6 8 10 12 14 16 18 20
Sample number
59
P-Charts
60
P control chart
61
MINITAB FOLLOW ALONG: CONSTRUCT P AND NP
CHARTS (EQUAL SAMPLE SIZES)
Background: A manufacturer of cough syrup sampled 50 bottles at a time
to see if they were filled correctly (not over or under filled).
62
R&S
C-CHART
UCL = c + zc
c = c
LCL = c - zc
where
63
c-Chart (cont.)
Number of defects in 15 sample rooms
NUMBER
SAMPLE OF
DEFECTS 190
1 12 c = 15 = 12.67
2 8
3 16 UCL = c + zc
: : = 12.67 + 3 12.67
= 23.35
: :
15 15 LCL= c + zc
190 = 12.67 - 3 12.67
= 1.99
64
c-Chart (cont.)
24
UCL = 23.35
21
18
Number of defects
c = 12.67
15
12
3 LCL = 1.99
2 4 6 8 10 12 14 16
Sample number
65
C Chart
66