Scribd
Upload
20 views12 pages

ANNOVA

The document discusses the Design of Experiments (DOE) and its application in analyzing experiments using ANOVA to compare means and assess significant variations among groups. It presents two problems involving technicians and workers, detailing hypotheses, calculations, and conclusions drawn from ANOVA tables. The aim is to determine performance differences among technicians and productivity differences among workers and machines.

Uploaded by

kokilamani.m
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
20 views12 pages

ANNOVA

The document discusses the Design of Experiments (DOE) and its application in analyzing experiments using ANOVA to compare means and assess significant variations among groups. It presents two problems involving technicians and workers, detailing hypotheses, calculations, and conclusions drawn from ANOVA tables. The aim is to determine performance differences among technicians and productivity differences among workers and machines.

Uploaded by

kokilamani.m
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

Design of Experiments - ANOVA

Introduction to Design of Experiments

▶ Design of Experiments (DOE) is a statistical method.


▶ It is used to plan and analyze experiments efficiently.
▶ It studies the effect of multiple factors.
▶ It improves accuracy of experimental results.
▶ It reduces time, cost and effort.
▶ ANOVA is a key technique in DOE.
▶ It compares means of different groups.
▶ It checks whether variation is significant.
▶ It is widely used in engineering and research.
Problem 1: Statement

The following are the numbers of mistakes made in 5 successive


days by 4 technicians working in a photographic laboratory.
6 14 10 9
14 9 12 12
10 12 7 8
8 10 15 10
11 14 11 11

Test at α = 1% whether the differences among means are


significant.
Problem 1: Hypothesis

Null Hypothesis: H0 : All technician means are equal


Alternative Hypothesis: H1 : At least one mean differs
Level of Significance: α = 1%
Method Used: One-way ANOVA
Objective: To check whether variation is due to chance or real
difference.
Problem 1: Calculations

Correction Factor:
(Total)2
CF =
N
Total Sum of Squares:
X
TSS = x 2 − CF

Between Columns:
Cj2
P
SSC = − CF
r
Error:
SSE = TSS − SSC
Problem 1: ANOVA Table

Source SS df MS F
Technicians SSC 3 MSC F
Error SSE 16 MSE
Decision Rule: Compare with F0.01 (3, 16)
Conclusion: Accept or reject H0
Problem 1: Conclusion

▶ Calculate CF, TSS, SSC and SSE.


▶ Find mean squares.
▶ Compute F-ratio.
▶ Compare with table value.
▶ If F is small, accept H0 .
▶ If F is large, reject H0 .
▶ It shows whether technicians differ.
▶ Helps to analyze performance.
▶ Final conclusion based on F-test.
Problem 2: Statement

The following data represent number of units produced per day by


5 workers using 4 machines.
44 38 47 36
46 40 52 43
34 36 44 32
43 38 46 33
38 42 49 39

Test: (i) Workers differ in productivity (ii) Machines differ in


productivity
Problem 2: Hypothesis

Null Hypotheses:
▶ H01 : Workers have equal productivity
▶ H02 : Machines have equal productivity
Alternative Hypotheses:
▶ Workers differ
▶ Machines differ
Method: Two-way ANOVA
Problem 2: Calculations

Correction Factor:
(Total)2
CF =
N
Total SS: X
TSS = x 2 − CF
Row SS:
Ri2
P
SSR = − CF
c
Column SS:
Cj2
P
SSC = − CF
r
Problem 2: ANOVA Table

Source SS df MS F
Workers SSR 4 MSR F1
Machines SSC 3 MSC F2
Error SSE 12 MSE
Decision: Compare F 1 and F 2 with table values
Problem 2: Conclusion

▶ Calculate SSR, SSC and SSE.


▶ Compute mean squares.
▶ Find F-values for rows and columns.
▶ Compare with critical values.
▶ Accept or reject hypotheses.
▶ Workers variation shows performance difference.
▶ Machines variation shows efficiency difference.
▶ Helps in selecting best worker and machine.
▶ Final conclusion based on ANOVA.

You might also like