00 upvotes00 downvotes

165 views79 pagesJan 25, 2011

© Attribution Non-Commercial (BY-NC)

PPTX, PDF, TXT or read online from Scribd

Attribution Non-Commercial (BY-NC)

165 views

00 upvotes00 downvotes

Attribution Non-Commercial (BY-NC)

You are on page 1of 79

2

3

CONTENTS

4

The theme of SPC is the Process: Everything we do in any

üRequires UNDERSTANDING

ühas VARIATION

ühas a CAPABILITY

üneeds IMPROVEMENT

5

Control: What & Why

qOptimization of performance

6

Taken from Presentation: How to Streamline, Integrate and Synergize Project Management and

Six Sigma Techniques for Optimal Results, by Candace G. Medina, CGM Associates, LLC, 2006 7

8

SPC is not really about statistics or control, it is about

three issues:

qQuality

qDelivery

qprice

9

What is quality?

customer.

10

Taken from Statistical Process Control, 5th Ed by John S. Oakland

11

Some basic SPC tools are: process flowcharting (what is done);

12

Taken from Six Sigma for Dummies , by Craig Gygi At All, Wiley Publishers, 2005 13

14

15

Histogram

Purpose

A histogram is used to graphically summarize and display the

distribution of a process data set.

- What is the systems most common response?

- What distribution (center, variation & shape) does the data have?

- Does the data look symmetric or is it skewed to the left or right?

- Does the data contain outliers?

16

Pareto Chart: (80:20 rule)

It’s a vertical bar chart that relates non-

numerical or qualitative categories to their

respective frequency or cost.

It charts the causes in descending order of

frequency or cost from left to right.

A Pareto chart is used to graphically

summarize and display the relative

importance of the differences between

groups of data.

17

Land Owners In Pakistan

120

100

80

60

40

20

0

Private

Armed Forces / Business

Feudals Industrial Employed Labour

Government Tycoons

Sector

Series2 55 25 9 6 4 1

Series1 55 80 89 95 99 100

18

Cause & Effect Diagram

Measurements Material Personnel

Safety

Related

Incidents

19

Taken from Six Sigma for Dummies , by Craig Gygi At All, Wiley Publishers, 2005 20

And of course:

Taken from Statistical Process

Control, 5th Ed by John S. Oakland

1.Bar Charts

2.Graphs 21

Control charts are simply run charts or plot of data in real time.

This plot enables us to see wood rather than tree which is not the

22

Taken from Statistical Process Control, 5th Ed by John S. Oakland

23

24

This is

rather

like a set

of traffic

lights

which

signal:

‘stop’,

‘caution’

or ‘go’.

25

The control chart has 3 zones and the action required depends on the

26

Questions to be asked

27

This approach may be applied to both variables

investigation.

28

Taken from Statistical Process Control, 5th Ed by John S. Oakland

29

Taken from Statistical Process Control, 5th Ed by John S. Oakland

30

Taken from Statistical Process Control, 5th Ed by John S. Oakland

31

The run chart becomes a control chart if decision lines are added

32

Some variation is just natural and cannot be eliminated. The

natural forces of nature naturally work to mix up

things for us because human created processes are

nature manipulation. For example heads and tails variation

in coin toss is natural. If the mail man arrival is 11:30 am

daily. But he comes (5 days of week) at 11:30:21, 11:29:45,

11:31:00, 11:30:10 and 11:29:59 am. This is perfectly

natural. But he some day comes at 12:30 pm or 10:30 it is

special cause variation. It may be due to schedule change or

flat tire.

eliminate it but we can eliminate special cause variation.

variation certainly repays later.

33

Long Term Variation (Shift)

Taken from Six Sigma for Dummies , by Craig Gygi At All, Wiley Publishers, 2005 34

Short Term Variation

Taken from Six Sigma for Dummies , by Craig Gygi At All, Wiley Publishers, 2005 35

36

Taken from Six Sigma for Dummies , by Craig Gygi At All, Wiley Publishers, 2005

37

About Sampling:

only link to the entire target population, you want that sample to be

really good.

38

A good sample represents the target population. The sample doesn't

being included in the sample. In other words, the process you use

39

"Less good information is better than more bad information, but

the sample size, but the bigger the sample size, the more accurate

40

A quick and dirty formula to calculate the accuracy of a survey is to

divide by the square root of the sample size. For example, a survey

margin of error.

41

Normal Distribution Curve

42

The Central Limit Theorem

size, the greater will be this tendency. Also, Grand or Process Mean

43

Taken from Statistical Process Control, 5th Ed by John S. Oakland

44

45

Taken from Statistical Process Control, 5th Ed by John S. Oakland

46

47

Steps in assessing process stability

100.

4 Calculate the Process Mean X – the average value of X and the Mean

48

5 Plot all the values of X and R and examine the charts for any

possible miscalculations.

6 Calculate the values for the action and warning lines for the

49

Some Terms

vGrand Mean

vState of Control

vProcess Capability

50

Total number of readings should be at least 50 and:

üNO Mean or Range values which lie outside the Action Limits

üNO more than about 1 in 40 values between the Warning and Action

Limits (Zone 2)

üNO incidence of two consecutive Mean or Range values which lie

outside the same Warning Limit on either the mean or the range chart

(Zone 2)

üNO runs of more than six sample Means which lie either above or

below the Grand Mean (Zone 1)

üNO trends of more than six values of the sample Means which are

either rising or falling (Zone 1)

51

Charts for individuals or run charts

and the individual values, not means of samples, are plotted. The

target value. The action lines (UAL and LAL) or control limits (UCL

52

When plotting the individual results on the i-chart, the rules for

either side of the mean are usually drawn, and action taken if four

53

Taken from Statistical Process Control, 5th Ed by John S. Oakland

54

The zone control chart

55

Taken from Statistical Process Control, 5th Ed by John S. Oakland

56

Pre-control

57

Other Control Charts

vDifference Charts

vZ Charts

58

Going

Practical

59

Taken from Statistical Methods for Industrial Process Control by David Drain, CRC Press

60

Taken from Modern Control Engineering, 2003, Lecture Notes

61

Taken from Modern Control Engineering, 2003, Lecture Notes

62

63

64

Taken from Statistical Methods for Industrial Process Control by David Drain, CRC Press

65

Taken from Statistical Methods for Industrial Process Control by David Drain, CRC Press

66

All following figures, tables and

graphs are based on Practical

industrial data. Either reproduced

or copied as such from industrial

presentations.

67

68

69

% oxygen in furnace flue gases

70

Fuel Gas Flow in KSCFH to two cells of

furnace

71

Pressure Variability Limits

TAG Norma UPPER LOWER UPPER LIMIT CONTROL LOWER LIMIT CONTROL

l LIMIT LIMIT PHILOSPHY PHILOSPHY

PRC - 101 505# 595# 490% On low load or other plant tripping to To control load

control battery limit press. According

to load

PRC - 102 589# 650/650 550 To avoid design limits of vessel D-2503

A/B

PRC - 103 479# 488/490 - To avoid design limit of vessel D-2510

PRC - 132 On high pressure P-2503 may trip on O/L Anti surge of K-2502 may open

amps.

PRC - 319 94# 100# 60# May damage burners, coil temp increase May burners Extinguish.

PRC - 306 240# 250# 160# May pop PSV - 929 Process air reduce.

KGT rpm may decrease

Six Sigma and ( continued )

Case Study – 3

Before PI

Thermo-compressor would have been sent to workshop for its opening for

internals’ inspection. Activity would have taken a minimum of two days,

resulting in low-pressure steam venting.

After PI

Motive steam flow trends from PI showed that the drop in efficiency

happened over a period of two months, contributed by scaling on the

motive steam path. This was confirmed by boroscopy. Thermo-compressor

was back in service after four hours of downtime required for cleaning.

decisions ”

74

SPC is a set of tools for managing processes, and

75

BIBLIOGRAPHY :

76

BIBLIOGRAPHY :

schungsbereich in 2002

77

BIBLIOGRAPHY :

ØStatistical Methods for Industrial Process Control by David Drain, ,

ØSix Sigma for Dummies, by Craig Gygi, Neil De Carlo and Bruce Williams

Sons in 2003

78

Thank You

79

## Much more than documents.

Discover everything Scribd has to offer, including books and audiobooks from major publishers.

Cancel anytime.