Introduction to Quality and Statistical Process Control

Chapter Goals
After completing this chapter, you should be able to:

   

Use the seven basic tools of quality Construct and interpret x-bar and R-charts Construct and interpret p-charts Construct and interpret c-charts
2

Chapter Overview
Quality Management and Tools for Improvement Philosophy of Quality Demings 14 Points Jurans 10 Steps to Quality Improvement Tools for Quality Improvement The Basic 7 Tools Control Charts X-bar/R-charts p-charts c-charts
3

Themes of Quality Management

      

Primary focus is on process improvement Most variations in process are due to systems Teamwork is integral to quality management Customer satisfaction is a primary goal Organization transformation is necessary It is important to remove fear Higher quality costs less

Demings 14 Points


1. Create a constancy of purpose toward improvement

 become more competitive, stay in business, and provide jobs

2. Adopt the new philosophy

 Better to improve now than to react to problems later

 3. Stop depending on inspection to achieve quality -- build in quality from the start
 Inspection to find defects at the end of production is too late

 4. Stop awarding contracts on the basis of low bids 5

 Better to build long-run purchaser/supplier relationships

Demings 14 Points

(continued)

 5. Improve the system continuously to improve quality and thus constantly reduce costs  6. Institute training on the job
 Workers and managers must know the difference between common cause and special cause variation

 7. Institute leadership
 Know the difference between leadership and supervision

 8. Drive out fear so that everyone may work effectively. 6  9. Break down barriers between departments

Demings 14 Points
 10. Eliminate slogans and targets for the workforce

They can create adversarial relationships

(continued)

 11. Eliminate quotas and management by objectives  12. Remove barriers to pride of workmanship  13. Institute a vigorous program of education and self-improvement  14. Make the transformation everyones job

Jurans 10 Steps to Quality Improvement

 1. Build awareness of both the need for improvement and the opportunity for improvement  2. Set goals for improvement  3. Organize to meet the goals that have been set  4. Provide training  5. Implement projects aimed at solving problems

Jurans 10 Steps to Quality Improvement (continued)

     6. Report progress 7. Give recognition 8. Communicate the results 9. Keep score 10. Maintain momentum by building improvement into the companys regular systems
9

The Deming Cycle

Plan

Act

The Deming Cycle

Do
The key is a continuous cycle of improvement 10

Study

The Basic 7 Tools

1. Process Flowcharts 2. Brainstorming 3. Fishbone Diagram 4. Histogram 5. Trend Charts 6. Scatter Plots 7. Statistical Process Control Charts
11

The Basic 7 Tools

1. 2. 3. 4. 5. 6. 7. Process Flowcharts Brainstorming Fishbone Diagram Histogram Trend Charts Scatter Plots Statistical Process Control Charts

(continued)

Map out the process to better visualize and understand opportunities for improvement

12

The Basic 7 Tools

1. 2. 3. 4. 5. 6. 7.

(continued)

Process Flowcharts Fishbone (cause-and-effect) diagram: Brainstorming Fishbone Diagram Cause 1 Cause 2 Histogram Trend Charts Sub-causes Scatter Plots Problem Statistical Process Control Charts
Sub-causes

Show patterns of variation

Cause 3

Cause 4

13

The Basic 7 Tools

1. 2. 3. 4. 5. 6. 7. Process Flowcharts Brainstorming Fishbone Diagram Histogram Trend Charts Scatter Plots Statistical Process Control Charts

(continued)

Identify trend y

time y Examine relationships

14

The Basic 7 Tools

1. 2. 3. 4. 5. 6. 7.

(continued)

Process Flowcharts Brainstorming Examine the performance Fishbone Diagram of a process over time Histogram Trend Charts X Scatter Plots Statistical Process Control Charts

time

15

Introduction to Control Charts

 Control Charts are used to monitor variation in a measured value from a process
 Exhibits trend  Can make correction before process is out of control

 A process is a repeatable series of steps leading to a specific goal  Inherent variation refers to process variation that exists naturally. This variation can be reduced but not eliminated

16

Process Variation
Total Process Common Cause Special Cause = + Variation Variation Variation
Variation is natural; inherent in the world around us No two products or service experiences are exactly the same With a fine enough gauge, all things can be seen to differ
17

Sources of Variation
Total Process Common Cause Special Cause = + Variation Variation Variation
Variation is often due to differences in: People Machines Materials Methods Measurement Environment

18

Common Cause Variation

Total Process Common Cause Special Cause = + Variation Variation Variation
Common cause variation naturally occurring and expected the result of normal variation in materials, tools, machines, operators, and the environment

19

Special Cause Variation

Total Process Common Cause Special Cause = + Variation Variation Variation
Special cause variation abnormal or unexpected variation has an assignable cause variation beyond what is considered inherent to the process
20

Statistical Process Control Charts

 Show when changes in data are due to:
 Special or assignable causes
 Fluctuations not inherent to a process  Represents problems to be corrected  Data outside control limits or trend

 Common causes or chance

 Inherent random variations  Consist of numerous small causes of random variability

21

Control Chart Basics

Special Cause Variation: Range of unexpected variability

UCL
Common Cause Variation: range of expected variability

+3 -3

Process Average LCL time

UCL = Process Average + 3 Standard Deviations LCL = Process Average 3 Standard Deviations
22

Process Variability
Special Cause of Variation: A measurement this far from the process average is very unlikely if only expected variation is present

UCL
3 99.7% of process values should be in this range

Process Average LCL time

UCL = Process Average + 3 Standard Deviations LCL = Process Average 3 Standard Deviations
23

Statistical Process Control Charts

Statistical Process Control Charts X-bar charts and R-charts
Used for measured numeric data

p-charts

c-charts

Used for proportions (attribute data)

Used for number of attributes per sampling unit

24

x-bar chart and R-chart

 Used for measured numeric data from a process  Start with at least 20 subgroups of observed values  Subgroups usually contain 3 to 6 observations each

25

Steps to create an x-chart and an R-chart

 Calculate subgroup means and ranges  Compute the average of the subgroup means and the average range value  Prepare graphs of the subgroup means and ranges as a line chart

26

Steps to create an x-chart and an R-chart

(continued)

 Compute the upper and lower control limits for the x-bar chart  Compute the upper and lower control limits for the R-chart  Use lines to show the control limits on the xbar and R-charts

27

Example: x-chart
 Process measurements:
Subgroup measures
Subgroup number

Individual measurements

Mean, x 14.5 13.0 19.0

Average subgroup mean = x

Range, R 6 7 4
Average subgroup range = R
28

1 2 3

15 12 17

17 16 21

15 9 18

11 15 20

Average of Subgroup Means and Ranges

Average of subgroup means: Average of subgroup ranges:

x x!
k
where:

R R!
k
where:

xi = ith subgroup average k = number of subgroups

Ri = ith subgroup range k = number of subgroups

29

Computing Control Limits

 The upper and lower control limits for an xchart are generally defined as
UCL = Process Average + 3 Standard Deviations LCL = Process Average 3 Standard Deviations

 or

UCL ! x  3W LCL ! x  3W
30

Computing Control Limits(continued)

 Since control charts were developed before it was easy to calculate , the interval was formed using R instead  The value A2R is used to estimate 3 , where A2 is from Appendix Q  The upper and lower control limits are UCL ! x  A 2 (R ) where A2 = Shewhart
LCL ! x  A 2 (R )
factor for subgroup size n from appendix Q
31

Example: R-chart
The upper and lower control limits for an R-chart are

UCL ! D 4 ( R ) LCL ! D3 ( R )
where: D4 and D3 are taken from the Shewhart table (appendix Q) for subgroup size = n
32

x-chart and R-chart

UCL

x-chart

x
LCL time UCL

R-chart

R
LCL time

33

Using Control Charts

 Control Charts are used to check for process control H0: The process is in control
i.e., variation is only due to common causes

HA: The process is out of control

i.e., special cause variation exists

 If the process is found to be out of control, steps should be taken to find and eliminate the special causes of variation
34

Process In Control
 Process in control: points are randomly distributed around the center line and all points are within the control limits
x
UCL

x
LCL

time
35

Process Not in Control

Out of control conditions:  One or more points outside control limits  Nine or more points in a row on one side of the center line  Six or more points moving in the same direction  14 or more points alternating above and below the center line 36

Process Not in Control

 One or more points outside control limits or more points in a row on one Nine side of the center line UCL UCL

x
LCL Six or more points moving in the same direction UCL

x
LCL 14 or more points alternating above and below the center line UCL

x
LCL

x
LCL 37

Out-of-control Processes
 When the control chart indicates an out-ofcontrol condition (a point outside the control limits or exhibiting trend, for example)
 Contains both common causes of variation and assignable causes of variation  The assignable causes of variation must be identified
 If detrimental to the quality, assignable causes of variation must be removed  If increases quality, assignable causes must be incorporated into the process design

38

p-Chart
 Control chart for proportions
 Is an attribute chart

 Shows proportion of nonconforming items

 Example -- Computer chips: Count the number of defective chips and divide by total chips inspected
 Chip is either defective or not defective  Finding a defective chip can be classified a success

39

p-Chart

(continued)

 Used with equal or unequal sample sizes (subgroups) over time

 Unequal sizes should not differ by more than 25% from average sample sizes  Easier to develop with equal sample sizes

 Should have np > 5 and n(1-p) > 5

40

Creating a p-Chart
 Calculate subgroup proportions  Compute the average of the subgroup proportions  Prepare graphs of the subgroup proportions as a line chart  Compute the upper and lower control limits  Use lines to show the control limits on the pchart
41

p-Chart Example
Subgroup number Sample size Number of successes Proportion, p

1 2 3

150 150 150

15 12 17

10.00 8.00 11.33

Average subgroup proportion = p

42

Average of Subgroup Proportions

The average of subgroup proportions = p
If equal sample sizes: If unequal sample sizes:

p p!
k

n p p! n
i i

where: pi = sample proportion for subgroup i k = number of subgroups of size n

where: ni = number of items in sample i 7ni = total number of items sampled in k samples
43

Computing Control Limits

 The upper and lower control limits for an pchart are
UCL = Average Proportion + 3 Standard Deviations LCL = Average Proportion 3 Standard Deviations

 or

UCL ! p  3W LCL ! p  3W
44

Standard Deviation of Subgroup Proportions

The estimate of the standard deviation for the subgroup proportions is
If equal sample sizes: If unequal sample sizes: Generally, s p is computed separately for each different sample size

( p )(1  p ) sp ! n
where:

p = mean subgroup proportion

n = common sample size
45

Computing Control Limits(continued)

 The upper and lower control limits for the pchart are
UCL ! p  3( s p ) LCL ! p  3(s p )
If sample sizes are equal, this becomes
Proportions are never negative, so if the calculated lower control limit is negative, set LCL = 0

( p )(1  p ) UCL ! p  3 n ( p )(1  p ) LCL ! p  3 n

46

p-Chart Examples
 For equal sample sizes
UCL p LCL

For unequal sample sizes

UCL p LCL

s p is constant since
n is the same for all subgroups

s p varies for each

subgroup since ni varies

47

c-Chart
 Control chart for number of nonconformities (occurrences) per sampling unit (an area of opportunity)
 Also a type of attribute chart

 Shows total number of nonconforming items per unit

 examples: number of flaws per pane of glass number of errors per page of code
48

 Assume that the size of each sampling unit remains constant

Mean and Standard Deviation for a c-Chart

 The mean for a c-chart is The standard deviation for a c-chart is

x c!
k

s! c

where: xi = number of successes per sampling unit k = number of sampling units

49

c-Chart Control Limits

The control limits for a c-chart are

UCL ! c  3 c LCL ! c  3 c

50

Process Control
Determine process control for p-chars and ccharts using the same rules as for x-bar and Rcharts Out of control conditions:
    One or more points outside control limits Nine or more points in a row on one side of the center line Six or more points moving in the same direction 14 or more points alternating above and below the center line
51

c-Chart Example
 A weaving machine makes cloth in a standard width. Random samples of 10 meters of cloth are examined for flaws. Is the process in control?

Sample number Flaws found

1 2

2 1

3 3

4 0

5 5

6 1

7 0
52

Constructing the c-Chart

 The mean and standard deviation are:

x c!
k

2  1 3  0  5  1 0 ! ! 1.7143 7

s ! c ! 1.7143 ! 1.3093

The control limits are:

UCL ! c  3 c ! 1.7143  3(1.3093) ! 5.642 LCL ! c  3 c ! 1.7143  3(1.3093) ! 2.214
Note: LCL < 0 so set LCL = 0
53

The completed c-Chart

The process is in control. Individual points are distributed around the center 6 UCL = 5.642 line without any pattern. Any improvement in the process must come 5 from reduction in common-cause variation
4 3 2 1 0 1 2 3 4 5 6 7

c = 1.714 LCL = 0 Sample number

54