0% found this document useful (0 votes)
127 views

STAT - ANOVA and Control Chart

Uploaded by

DebashishDolon
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
127 views

STAT - ANOVA and Control Chart

Uploaded by

DebashishDolon
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 54

LEAN SIX SIGMA

CERTIFICATION COURSE
STAT-06 ANOVA & Control Chart

stc bangladesh
ANOVA
Analysis Of Variance

Copyright©stc bangladesh 2
Module Objectives

By the end of this module, the participant will be able to:

– Understand how ANOVA works


– Interpret an ANOVA table
– Determine significant effects
– Test groups of data for equal variances
– Run main effects plots

Copyright©stc bangladesh 3
Why Learn ANOVA?

– Performs hypothesis testing for two or more means

– Handles multiple levels

– Shows sources of process variation

– Generates an underlying variability estimate

Copyright©stc bangladesh 4
ANOVA

• Hypothesis Test for MEANS

– Uses two components of variance

• within variance (no change)

• between variance (after a change)

• Uses the F-distribution to test the variance components

Copyright©stc bangladesh 5
When To Use ANOVA
Variables Road Map
Variables Data

1 Sample 2 Samples More Than 2 Samples

1 Sample 2 Sample Paired Tukey's


ANOVA
t-test t-test Comparisons Quick Test

1 Mean 2 Means 2+ Means

ANOVA is used to test two or more means

Copyright©stc bangladesh 6
Process Variation

• All processes are influenced by other factors

• Is variation due to a real factor effect or are the differences just


random variation?

• t-tests are tools that offer some help, but are limited to testing
two means

• Finding factors that are sources of variation are key to process


improvement

ANOVA allows concurrent testing of several means

Copyright©stc bangladesh 7
The ANOVA Hypothesis Test

• Ho: m1=m2=…=mn

• Ha: at least one mean is different


• Reject Ho if Fobs > F a-1, N-a, 
• p-value will be used to evaluate the hypothesis

The null is no difference in means. The alternative is one or more is different

Copyright©stc bangladesh 8
Sources Of Variation
Y

Total
Mean
Variation

ANOVA reveals significant mean shifts by analyzing within,


between and total process variance

Copyright©stc bangladesh 9
Sources Of Variation

SS Between is the variation due to factor change

SS Within is the variation (Noise) within each subgroup

SS Total is all the variation (Noise) around the mean of the


process.

Copyright©stc bangladesh 10
The ANOVA Table
Sum of Degree of
Mean square
Source of variation square freedom F-statistic

SSB
Between treatments
SSB a-1 MSB= a-1
MSB
SSE
Fo= MS E
Within treatments (Error) N-a
SSE MSE= N-a

Total SST N-1

a: Level of factors
N: Total number of observations
n = sample size

Copyright©stc bangladesh 11
The ANOVA Table

SSTotal  SS Between  SSWithin

All process variation is accounted for


Copyright©stc bangladesh 12
Example practice……..
A process can be run at any of three temperature: 180 F, 200 F and 220 F.
Using the data in the table and assuming  = 0.05, determine whether
temperature significantly affects the moisture content.

Level 180 200 220


1 10.8 11.4 14.3
2 10.4 11.9 12.6
3 11.2 11.6 13.0
4 9.9 12.0 14.2
Total 42.3 46.9 54.1

a: Level of factors=3
N: Total number of observations=12
n = sample size=4

Copyright©stc bangladesh 13
Example-Solution

SS Between = [(42.32+46.92+54.12)/4]-[(42.3+46.9+54.1) 2/12]=17.69

SS within = (10.82+10.42+11.22+9.92+….+13.02+14.22)-[(42.32+46.92+54.12)/4]=3.34

SS Total = (10.82+10.42+11.22+9.92+….+13.02+14.22)-[(42.3+46.9+54.1) 2/12]=21.03

Copyright©stc bangladesh 14
Analyzing The Variation

SSTotal  SSBetween  SSWithin


21.03  17.69  3.34

Source SS %
SSBetween 17.69 84.12%
SSWithin 3.34 15.88%
SSTotal 21.03

The within and between variation must be properly allocated


across the sources

Copyright©stc bangladesh 15
The ANOVA Table
Sum of Degree of
Source of variation Mean square F-statistic
square freedom
Between treatments 17.69 2 8.85 23.92

Within treatments (Error) 3.34 9 0.37

Total 21.03 11

Fobs=23.92
F0.05,2,9=4.26
As Fobs > F table, reject the null.

Copyright©stc bangladesh 16
The ANOVA Table- Exercise

Analysis of Variance for Invoices


Source DF SS MS F P
Consumer 3 13.00 4.33
Commercial 20.00 10.00 0.0012
Error 75
Total 80 183.00

Copyright©stc bangladesh 17
The ANOVA Table-Solution
Analysis of Variance for Invoices
Source DF SS MS F P
Consumer 3 13.00 4.33 2.17 0.0986
Commercial 2 20.00 10.00 5.00 0.0012
Error 75 150.00 2.00
Total 80 183.00

Copyright©stc bangladesh 18
Exercise……..
Researchers want to test a new anti-anxiety medication. They split participants
into three conditions (0mg, 50mg, and 100mg), then ask them to rate their
anxiety level on a scale of 1-10. Are there any differences between the three
conditions using alpha = 0.05?

Copyright©stc bangladesh 19
Solution……..
SS Between = [(572+472+212)/7]-[(57+47+21)/21]= 98.67

SS within = (92+82+72+82+….+32+22)-[(572+472+212)/7]= 10.29

SS Total = 98.67+10.29= 108.96

Here, a=3; N=21 & n=7

Sum of Degree of
Source of variation Mean square F-statistic
square freedom
Between treatments 98.67 2 49.34 86.56

Within treatments (Error) 10.29 18 0.57

Total 108.96 20

Copyright©stc bangladesh 20
Solution……..

Fobs=86.56
F0.05,2,18=3.55
As Fobs > F table, reject the null.

Copyright©stc bangladesh 21
Introduction to Control Chart

Copyright©stc bangladesh 22
Module Objectives

By the end of this module participant will be able to:

– Describe the different types of control chart


– Recognize upper and lower control limit
– Use control charts for decision making
– Relate the use of control charts to real world examples

Copyright©stc bangladesh 23
Control chart
1. A statistical tool used to distinguish between process variation
resulting from common causes and variation resulting from special
causes.

2. A graph that has time or order of submission of sequential lots on X


axis and the average test results on the Y axis.

Copyright©stc bangladesh 24
Common Vs Special Cause

1. Common causes of variation


─ Random causes that we cannot identify
─ Unavoidable
─ Cause slight differences in process variables like diameter, weight,
service time, temperature, etc.

2. Assignable causes of variation


─ Causes can be identified and eliminated
─ Typical causes are poor employee training, worn tool, machine
needing repair, etc.

Copyright©stc bangladesh 25
Common Vs Special Cause
• Process: Driving to Work
• Average Time: 12 minutes
• Standard Deviation: 2.5 minutes

• Common Causes – Wind speed, miss one green light, driving speed,
number of cars on road, time when leaving house.

• Special Causes – Stop for school bus crossing, traffic accident, pulled
over for speeding, poor weather conditions, car mechanical problems,
construction detour, stoplights not working properly, train crossing.

Copyright©stc bangladesh 26
What Control Chart Does?

• It graphically, represents the output of the process.

• Uses statistical limits and patterns of plot, for decision making

Copyright©stc bangladesh 27
Why Use Control Charts?
• To monitor process variation over time

• To differentiate between special cause and common cause variation

• To assess effectiveness of changes

• To communicate process performance

• In controlling and analyzing physical, chemical, analytical or biological


parameters of a product, such as: Weight, thickness variation of
products, Volume filling in a container the percentage of defects in
parenteral products etc.

Copyright©stc bangladesh 28
When to use a control chart?
─ Controlling ongoing processes by finding and correcting problems as
they occur.

─ Predicting the expected range of outcomes from a process.

─ Determining whether a process is stable (in statistical control).

─ Analyzing patterns of process variation from special causes (non-routine


events) or common causes (built into the process).

─ Determining whether the quality improvement project should aim to


prevent specific problems or to make fundamental changes to the
process

Copyright©stc bangladesh 29
Types of Control Charts
1. Variables control charts:

─ Variable data are measured on a continuous scale.


Example: time, weight, distance or temperature etc.

─ Applied to data with continuous distribution

2. Attributes control charts:

─ Attribute data are counted and cannot have fractions or decimals.


Attribute data arise when you are determining only the presence or
absence of something: success or failure, accept or reject, correct or
not correct.
Example: a report can have four errors or five errors, but it cannot have four
and a half errors.

─ Applied to data following discrete distribution

Copyright©stc bangladesh 30
Types & Selection of Control Chart
np chart
Constant Sample Size
(number of defectives)
Defectives
Attribute data

p chart
Variable Sample Size
(proportion defectives)

c chart
Constant Sample Size
(defects per sample )
Control Charts

Defects
U chart
Variable Sample Size
(defects per unit)

n>10 x̅ and S chart


Variable data

2<n<10 x̅ and R chart

n=1 I-MR chart

Copyright©stc bangladesh 31
Out of Control Condition

1. One point more than 3σ from the center line (either side)

Copyright©stc bangladesh 32
Out of Control Condition

2. Nine points in a row on the same side of the center line

Copyright©stc bangladesh 33
Out of Control Condition

3. Six points in a row, all increasing or all decreasing

Copyright©stc bangladesh 34
Out of Control Condition

4. Fourteen points in a row, alternating up and down

Copyright©stc bangladesh 35
Out of Control Condition

5. Two out of three points more than 2σ from the center line (same side)

Copyright©stc bangladesh 36
Out of Control Condition

6. Four out of five points more than 1σ from the center line (same side)

Copyright©stc bangladesh 37
Out of Control Condition

7. Fifteen points in a row within 1σ of the center line (either side)

Copyright©stc bangladesh 38
Out of Control Condition

8. Eight points in a row more than 1σ from the center line (either side)

Copyright©stc bangladesh 39
R-chart
─ Always look at the Range chart first.
─ Indicates how the range of the subgroups changes over times.
─ Look for out of control points. If there are any, then the special
causes must be eliminated.
─ Once the effect of the out of control points from the Range chart is
removed, look at the X-bar Chart.

Copyright©stc bangladesh 40
x̅-chart
─ X-bar charts are used to monitor the mean of a process based on the
average of a series of observations, called a subgroup.

Copyright©stc bangladesh 41
x̅-R chart
─ An X-bar and R (range) chart is used to monitor mean and variance of a
process when quality characteristic is continuous.
─ The standard chart for variables data, X-bar and R charts help
determine if a process is stable and predictable.
─ X-bar and R charts can be used to any process with a subgroup size 2 to
10.

x̅-R chart

Copyright©stc bangladesh 42
Example practice……..
A quality control inspector at the Cocoa Fizz soft drink company has taken
three samples with four observations each of the volume of bottles filled.
use the below data to develop x̅-R control charts.
0.2  0.3  0.2
Time 1 Time 2 Time 3 R  .233
3
Observation 1 15.8 16.1 16.0

Observation 2 16.0 16.0 15.9


UCLR  D4 R  2.28(.233) .53
Observation 3 15.8 15.8 15.9
0.2  0.3  0.2 LCLR  D3R  0.0(.233)  0.0
R  4 15.9
Observation  .23315.8
15.9
3

UCL x  x  A 2 R  15.92  0.73.233  16.09

LCL x  x  A 2 R  15.92  0.73.233  15.75

Copyright©stc bangladesh 43
Example practice……..

Copyright©stc bangladesh 44
Exercise……..
Use the below data to develop x̅-R control charts.

Observation-1 Observation-2 Observation-3 Observation-4 Observation-5

11.1 10.1 9.8 11.3 11.2


9.2 11.2 10.2 10.1 9.4
11.3 9.9 9.9 10.1 8.9

Copyright©stc bangladesh 45
Solution……..
R-bar=1.47; UCL=3.75; LCL=0

X-bar-bar=10.24; UCL=11.73; LCL=8.75

Copyright©stc bangladesh 46
S-chart
─ An S-chart is a type of control chart used to monitor the process variability
(as the standard deviation) when measuring subgroups (n ≥ 10) at regular
intervals from a process.

─ The major difference is that the subgroup standard deviation is plotted


when using the x̅-s chart, while the subgroup range is plotted when using
the x̅-R chart.

─ One advantage of using the standard deviation instead of the range is that
the standard deviation takes into account all the data, not just the maximum
and the minimum.

Copyright©stc bangladesh 47
Example practice……..
Use the below data to develop x̅-s control charts.

S1 11.1 10.1 9.8 11.3 11.2 10.1 11.5 11.0 10.8 11.1 9.5 0.68
Stv
S2 9.2 11.2 10.2 10.1 9.4 9.3 10.2 9.7 9.6 10.2 10.8 0.62
s3 11.3 9.9 9.9 10.1 8.9 11.0 9.8 9.9 11.2 9.3 11.5 0.86

x S

Copyright©stc bangladesh 48
Example Solution……..
X-bar-bar=10.31; S-bar=0.72

UCLX=10.31+1.954*0.72=11.71
LCLx=10.31-1.954*0.72=8.91

UCLs=2.568*0.72=1.85;
LCLs=0*0.72=0

Copyright©stc bangladesh 49
Exercise……..
Use the below data to develop x̅-s control charts.

Copyright©stc bangladesh 50
Solution……..

Copyright©stc bangladesh 51
Individual & Moving Range(XmR,ImR) chart
─ There are two charts – one with Individual data points and the other chart has
data points that are range values derived between consecutive individual
points.

─ Very slow processes or processes in which the measurement is very expensive


to obtain.

Copyright©stc bangladesh 52
Example practice……..
Use the below data to develop XmR control charts.

Reading Individual data element


1 290
2 288
3 285
4 290
5 291
6 287
7 284
8 290
9 290
10 288

Copyright©stc bangladesh 53
Example practice……..
Read Individual mR
ing data element
1 290
2 288 2
3 285 3
4 290 5
5 291 1
6 287 4
7 284 3
8 290 6
9 290 0
10 288 2

Copyright©stc bangladesh 54

You might also like