Chap 4 SPC

Chapter 4
Statistical Process Control
4.1
Introduction
Weights of the cereal packets:
498 gms, 505 gms, 443 gms, 503 gms, 501 gms
Natural Cause Assignable cause
Possible Factors behind Assignable Causes

Man Inexperience, fatigue, personal problem Machine Faulty machine, tool wear, aging machine, broken tool Materials Defective raw materials supplied by new supplier Method Work instruction Measurement Problem in the measuring instrument itself Environment - Temperature, humidity, sudden power surge
Definition of SPC
Statistical process control is a statistical method of separating variation resulting from special causes from variation resulting from natural causes to eliminate the special causes and to establish and maintain consistency in quality of products/services.
Various SPC Tools

Pareto charts Flow diagram Cause and effect diagram Histogram Check sheet Scatter diagram Run chart
Control chart
Origin of SPC
Walter Shewhart Father of SPC -- The Economic Control of Quality of Manufactured Products (published in 1931)
Variable Attribute
Types of Control Charts
Control Charts Variable Charts

x-bar chart Median chart R chart s chart p chart
Attribute Charts
np chart c chart u chart
Overall look of a Control Chart

Upper control limit
Centre line Lower control limit

1 2 3 4 5 6 7
Sample number
Overall Look of a Control Chart (Contd)

Upper control limit
Centre line
Lower control limit

1 2 3 4 5 6 7 8 9 10
Sample number
4.2Variable Control Charts
Examples of Variables
Length Weight Time Volume Temperature Density Pressure Energy Force velocity voltage
Specific Examples of Variables

Time waiting for service Time to process insurance claims Time from order receipt to shipment Diameter of a piston Weight of fertilizer packets Volume of shampoo bottles
X-bar and R charts should be used together
R chart
LCL R D3 R UCL R D4 R
For x-bar chart n

2
For R-chart D3
0
A2
1.880
D4
3.267
3
4 5
1.023
0.729 0.577
0
0 0
2.574
2.282 2.114
6
7 8 9 10 11 12 13
0.483
0.419 0.373 0.337 0.308 0.285 0.266 0.249
0
0.076 0.136 0.184 0.223 0.256 0.283 0.307
2.004
1.924 1.864 1.816 1.777 1.744 1.717 1.693
14
15
0.235
0.223
0.328
0.347
1.672
1.653
Example
Bukit Battery Corp. has recently been receiving complaints from retailers that its 9-volt batteries are not lasting as long as other name brands. James West, head of the TQM program at Wests Austin plant, believes there is no problem because his batteries have had an average life of 50 hours, about 10% longer than competitors models. To raise the life time above this level would require a new level of technology not available to West. Nevertheless, he is concerned enough to set up hourly assembly line checks. He decides to take size-5 samples of 9-volt batteries for each of the next 10 hours to create the standards for control chart limits (see the following table):
Bukit Battery data Battery Lifetimes (in hours) Sample Hour 1 2 3 1 51 45 50 2 50 47 35 3 49 70 48 4 50 46 39 5 50 36 47
4
5 6 7
55
49 59 36
70
38 62 33
50
64 40 49
30
36 54 48
51
47 64 56
8
9 10
50
43 70
67
22 45
53
36 50
43
27 47
40
44 41
Is the manufacturing process under control?
Bukit Battery data Battery Lifetimes (in hours) Sample Hour 1 2 3 4 5 R
1
2 3 4
51
45 50 55
50
47 35 70
49
70 48 50
50
46 39 30
50
36 47 51
2
34 15 40
5
6 7 8
49
59 36 50
38
62 33 67
64
40 49 53
36
54 48 43
47
64 56 40
28
24 23 27
9
10
30
70
22
45
36
50
27
47
44
41
22
29
51.58
40
30
24.4
20
10
10
X-bar Chart
LCLx x A2 R UCLx x A2 R
Bukit Battery data Battery Lifetimes (in hours)

Sample Hour 1 2 3 4 5 6 7 8 9 10 1 51 45 50 55 49 59 36 50 30 70 2 50 47 35 70 38 62 33 67 22 45 3 49 70 48 50 64 40 49 53 36 50 4 50 46 39 30 36 54 48 43 27 47 5 50 36 47 51 47 64 56 40 44 41 R 2 34 15 40 28 24 23 27 22 29
x
50.0 48.8 43.8 51.2 46.8 55.8 44.4 50.6 31.8 50.6
61.46 60
50
47.38
40
33.30
1 2 3
10
When can we say that a process is under control?

There are no sample points that lie outside the control limits. Most of the points are clustered around the center line; not too many points close to the control limits. Approximately half of the sample points lie above the center line and approximately half of the points lie below the center line.
Properties of Normal Distribution

= Standard deviation
Mean -3 -2 -1 +1 +2 +3 68.26% 95.44% 99.74%
Special cases
Checklist of questions for out-of-control process

(i) Are there any differences in the measurement accuracy of the instruments used? Are there differences in the methods used by different operators? Does the environment affect the process? Is the process affected by tool-wear?
(ii) (iii) (iv)
Checklist of questions for out-ofcontrol process (Contd)

Were any untrained workers involved in the process? (vi) Has there been any change in the source of the raw materials? (vii) Is the process affected by operators fatigue? (viii) Did samples come from different shifts, operators, or machines?
(v)
Few Points for solving real world problems

Revised control limit Sample size Sample number
4.3 Attribute Control Charts
Defects and Defective

Defects are products/services non-conformities in the
A product or service is said to be defective if it has one or more than one defects.
p chart
p (1 p ) LCL p p 3 n p (1 p ) UCL p p 3 n
Example
A quality engineer takes 20 samples of 100 transistors each from the output of an assembly line. Each transistor is tested and the number of defectives in each sample is recorded which are shown in the next slide.
Draw the p chart and conclude.
Sample No.
1
Number of defectives
3
Sample No.
11
Number of defectives
9
2 3 4 5 6 7 8 9 10
2 1 4 0 2 0 5 1 2
12 13 14 15 16 17 18 19 20
1 3 3 4 5 3 3 1 6
Sample No.
No. of Defectives
Percentage of defectives
Sample No.
No. of Defectives
Percentage of defectives
1 2 3 4
3 2 1 4
0.03 0.02 0.01 0.04
11 12 13 14
9 1 3 3
0.09 0.01 0.03 0.03
5
6 7 8 9 10
0
2 0 5 1 2
0.00
0.02 0.00 0.05 0.01 0.02
15
16 17 18 19 20
4
5 3 3 1 6
0.04
0.05 0.03 0.03 0.01 0.06
0.0793
0.029
7 8 9 10 11 12 13 14 15 16 17 18 19 20
C chart
LCLc c 3 c UCLc c 3 c
Example
Coils of synthetic rope are monitored for defects. Each coil contains 1,200 feet of rope. Eighteen rolls have been selected and the number of defects per roll is presented in the next slide. Identify an appropriate control chart and determine the charts control limits. Draw the control chart and conclude.
Sample No.
1
Number of defects
1
Sample No.
10
Number of defects
2
2 3 4 5 6 7 8 9
3 4 2 4 2 0 3 2
11 12 13 14 15 16 17 18
1 4 3 0 5 4 2 3
7.24
6 5 4
1 2 3 4 5 6 7 8 9
3 2.5 2
1
10 11 12 13 14 15 16 17 18
Start
Collect data. Determine sample size. Select 25 to 30 samples
After tabulating data compute the necessary measures Compute the center line and control limits for the control chart using the appropriate formulas.
Draw the control chart and plot the sample data on it.
Are any data points outside the Control limits?
Yes
No
The center line and control limits become permanent, and the control chart is used for monitoring the future behavior of the process.
Special causes are sought for out-of-control conditions and the out-ofcontrol points are removed. New centerline and control limits are computed from the remaining points. The remaining points are plotted ob the new chart.
End
Control Chart Selection

Quality Characteristic
variable
defective n>1? no x and MR
attribute
defect
yes
n>=10 or no computer? x and R
constant sample size?

no
yes
p or np
constant sampling unit? yes c no u
yes
x and s
p-chart with variable sample size
4.4 Applications
Quality Characteristics: Selected Industries

Hospital: Timeliness and quickness of care, staff responses to request, accuracy of lab test, cleanliness, checkout time. Grocery Store: Waiting time to check out, frequency of stockout items, cleanliness, customer complaints Airlines: Flight delays, lost luggage, check-in-time, flight attendants courtesy, cabin cleanliness. Fast Food Restaurant: Waiting time for service, customer complaints, cleanliness, order accuracy, employee courtesy
1.
KFCs South Central division (Texas and Oklahoma) uses SPC to monitor service times in its drive-through windows. It uses x-bar and R charts. Kurt Manufacturing Company (based Minneapolis) produces precision machines. in
2.
SPC is used by machine operators to monitor their processes by measuring variability, reliability of the machines. SPC charts are displayed on the shop floor so that everybody can see it. If there is any problem in the production, then the operators are empowered to stop production and find out out the cause and remove it.
3. P*I*E* Nationwide is Americas fourth largest trucking company.One part of the companys quality progrom is to extensive use of SPC.
A p-chart was initially used to monitor the production of daily defective freight bills. This resulted in reduction in the error rate from 10% to 0.8% within one year and the subsequent reduction in inspection time increased productivity by 30%. It is also estimated that use of SPC results savings of $38,000 at a single trucking terminal.
Motorola Malaysia Sdn. Bhd.

Chap 4 SPC

Uploaded by

Copyright:

Available Formats

Chap 4 SPC

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chap 4 SPC

Uploaded by

Copyright:

Available Formats

Chapter 4

Statistical Process Control

Weights of the cereal packets:

Possible Factors behind Assignable Causes

Various SPC Tools

Types of Control Charts

Control Charts Variable Charts

Overall look of a Control Chart

Centre line Lower control limit

Overall Look of a Control Chart (Contd)

Lower control limit

4.2Variable Control Charts

Specific Examples of Variables

For x-bar chart n

Bukit Battery data Battery Lifetimes (in hours) Sample Hour 1 2 3 1 51 45 50 2 50 47 35 3 49 70 48 4 50 46 39 5 50 36 47

Is the manufacturing process under control?

Bukit Battery data Battery Lifetimes (in hours) Sample Hour 1 2 3 4 5 R

Bukit Battery data Battery Lifetimes (in hours)

When can we say that a process is under control?

Properties of Normal Distribution

Mean -3 -2 -1 +1 +2 +3 68.26% 95.44% 99.74%

Checklist of questions for out-of-control process

(ii) (iii) (iv)

Checklist of questions for out-ofcontrol process (Contd)

Few Points for solving real world problems

4.3 Attribute Control Charts

Defects and Defective

0.03 0.02 0.01 0.04

0.09 0.01 0.03 0.03

Collect data. Determine sample size. Select 25 to 30 samples

Control Chart Selection

constant sample size?

constant sampling unit? yes c no u

p-chart with variable sample size

Quality Characteristics: Selected Industries

Motorola Malaysia Sdn. Bhd.

You might also like