You are on page 1of 31

FORMULAS:

Source of Variation df Sum Sq Mean Sq F-value


Treatments k-1 SST MST = SST / (k-1) MST / MSE
Error n-k SSE MSE = SSE / (n-k)
Total n-1 Total SS
FORMULAS:
Source of Variation df Sum Sq
Treatments k-1 SST
Blocks b-1 SSB
Error (b - 1)(k - 1) SSE
Total n - 1 = bk -1 Total SS
Mean Sq F-value
MST = SST / (k-1) MST / MSE
MSB = SSB / (b-1) MSB / MSE
MSE = SSE / [(b - 1)(k - 1)]
FORMULAS:
Source of Variation df Sum Sq
Factor A a -1 SSA
Factor B b -1 SSB
Interaction AB (a - 1)(b - 1) SS(AB)
Error (Within) ab (r -1) SSE
Total abr - 1 Total SS
Mean Sq F-value
MSA = SSA / (a-1) MSA / MSE
MSB = SSB / (b-1) MSB / MSE
MSAB = SSE(AB) / [(a - 1)(b - 1)] MSAB / MSE
MSE = SSE / [ab (r -1)]
Question:
It was on June of 2019 when the new batch of Mechanical Technician Trainees were hired to supplement the manpower requirem
the company in adherence to the plant's expansion projects. The group was composed of 23 trainees and they were deployed to
teams such as Polymer, NCP ISBL, and OSBL. After a year of being trained in their respective teams, these technicians underwe
revalidation tests (written, oral, and hands-on) to evaluate them if they were qualified to become a Certified Mechanical Technicia
average percentage scores of each technician were tabulated in the proceeding table. This analysis aims to determine whether th
difference in the effectiveness of the training being offered by each team in reference to the performance of the technicians in the
revalidation tests.
Initial Hypothesis:
Null Hypothesis (Ho): There is no statistically significant difference among the means of the treatment (areas of assignment).
Alternative Hypothesis (Ha): There is a statistically significant difference among the means of the treatment (areas of assignment)

Areas of Assignment Average Percentage Scores


Polymer Team 88 90 90 92 93 91 86
NCP ISBL Team 90 91 91 94 91 91 92
OSBL Team 91 89 91 89 87 92
ment the manpower requirement of
and they were deployed to different
these technicians underwent three
tified Mechanical Technician 1. The
ims to determine whether there is a
nce of the technicians in the

(areas of assignment).
ment (areas of assignment).

89
92 91
Anova: Single Factor

SUMMARY
Areas of Assignment Count Sum Average Variance
Polymer Team 8 719 89.875 4.982
NCP ISBL Team 9 823 91.444 1.278
OSBL Team 6 539 89.833 3.367

ANOVA
Source of Variation df Sum Sq Mean Sq F-value P-value
Treatment (Areas of Assignment) 2 13.809 6.904 2.230 0.134
Error 20 61.931 3.097
Total 22 75.739
F-critical
3.493
Question:
Several FPSO (Floating Production Storage and Offloading) projects depend heavily on the quality of steel structures anchoring th
steady onto the seabed while in operation. Any failure on the material can lead to explosion or oil spill into the ocean, leading to
catastrophic environmental impacts. This is why engineering companies deeply invest in research about material composition and
geological inspection before constructing these structures.
One material vendor is supplying steel for differrent offshore projects, and the client wants to be assured that these materials will
equally among the different locations. A study is done to test if corrosion rate (nanometer/year) among the different locations are u
for the material supplied.
Several samples of the material are submerged into the waters of the different project locations, and after one year, corrosion is m
Null Hypothesis (Ho): There is no statistically significant difference among the means of the treatment (test locations).
Alternative Hypothesis (Ha): There is a statistically significant difference among the means of the treatment (test locations).

Test Locations Measured Corrosion Rates


FPSO A Location 85 93 102 78 83 94 77
FPSO B Location 92 74 88 97 73 91 82
FPSO C Location 97 91 75 88 94 79 81
FPSO D Location 86 77 84 79 90 82 87
ality of steel structures anchoring them
oil spill into the ocean, leading to
rch about material composition and

e assured that these materials will behave


among the different locations are uniform

, and after one year, corrosion is measured.


atment (test locations).
he treatment (test locations).

Rates
79 100 85
91 81 76
79 94 86
81 79 91
Anova: Single Factor

SUMMARY
Test Locations Count Sum Average Variance
FPSO A Location 10 876 87.6 82.711
FPSO B Location 10 845 84.5 71.389
FPSO C Location 10 864 86.4 57.822
FPSO D Location 10 836 83.6 23.156

ANOVA
Source of Variation df Sum Sq Mean Sq F-value P-value
Treatment (Test Locations) 3 98.275 32.758 0.557 0.647
Error 36 2115.7 58.769
Total 39 2213.975
F-critical
2.866
Anova: Single Factor

SUMMARY
Groups Count Sum Average Variance
FPSO A Location 10 876 87.6 82.711111111
FPSO B Location 10 845 84.5 71.388888889
FPSO C Location 10 864 86.4 57.822222222
FPSO D Locatio 10 836 83.6 23.155555556

ANOVA
Source of Variation SS df MS F P-value F crit
Between Group 98.275 3 32.758333333 0.5574041688 0.6465593145 2.8662655509
Within Groups 2115.7 36 58.769444444

Total 2213.975 39
Question:
In the Gas Turbine Integration Project of the company, there was a requirement to determine the type of grouting material that wil
in the foundation of the equipment. The OEM vendor provided three options in their manual for us to evaluate and the choices inc
cementitious grout, epoxy grout, and furan resin grout. As a way forward, the PMC of the project tasked the EPC to conduct a com
strength testing by incorporating the different grouting materials into a concrete cylinder. There were 12 concrete cylinder samples
for each material type and every three samples of which were subjected to varied drying/curing temperatures (25degC, 30degC, 3
and 40degC). The average compressive strength (MPa) obtained from the 3 samples for each type and scenario were tabulated i
proceeding table.
Initial Hypothesis:
Null Hypothesis (Ho): There is no statistically significant difference among the treatment (types of grouting material) means.
Null Hypothesis (Ho): There is no statistically significant difference among the block (drying/curing temperature) means.
Alternative Hypothesis (Ha): There is at least one of the means that has a statistically significant difference from the others.

Drying/Curing Temperature
Types of Grouting Material
25 deg C 30 deg C 35 deg C

Cementitious Grout 18.6 19.4 20.5

Epoxy Grout 16.5 17.5 18.2

Furan Resin Grout 13.8 15.9 21.5


e of grouting material that will be used
evaluate and the choices includes
ed the EPC to conduct a compressive
12 concrete cylinder samples made
eratures (25degC, 30degC, 35degC,
nd scenario were tabulated in the

uting material) means.


mperature) means.
rence from the others.

e
40 deg C

19.8

17.3

22.2
Anova: Two-Factor Without Replication

Summary Count Sum Average Variance


Cementitious Grout 4 78.300 19.575 0.629
Epoxy Grout 4 69.500 17.375 0.489
Furan Resin Grout 4 73.400 18.350 17.150

25 deg C 3 48.900 16.300 5.790


30 deg C 3 52.800 17.600 3.070
35 deg C 3 60.200 20.067 2.863
40 deg C 3 59.300 19.767 6.003

ANOVA
Source of Variation df Sum Sq Mean Sq F-value
Rows (Types of Grouting Material) 2 9.722 4.861 1.133
Columns (Drying/Curing Temperature) 3 29.073 9.691 2.260
Error 6 25.732 4.289
Total 11 64.527
P-value F-critical
0.382 5.143
0.182 4.757
Question:
A refrigerant supplier for HVAC equipment manufacturer was recently invited to a bidding to submit a comparative report of their a
products as applied to four different brands of chiller refrigeration system. The report requires the Coefficient of Performance (CO
determine the behavior and effects of the different refrigerants when introduced to the system. The available products of the comp
which met the requirements of the client are R410A, R32, and R22 while the brands of the chiller considered in the comparative a
will be designated as Brand A, B, C, and D for confidentiality purposes.
Initial Hypothesis:
Null Hypothesis (Ho): There is no statistically significant difference among the treatment (types of refrigerant) means.
Null Hypothesis (Ho): There is no statistically significant difference among the block (different brands of chiller) means.
Alternative Hypothesis (Ha): There is at least one of the means that has a statistically significant difference from the others.

Different Brands of Chiller


Types of Refrigerant Brand A Brand B Brand C

R410A 2.7 2.6 3.5

R32 2.9 3.2 4.8

R22 2.5 2.2 3.4


comparative report of their available
efficient of Performance (COP) to
vailable products of the company
sidered in the comparative analysis

igerant) means.
of chiller) means.
rence from the others.

Brand D

2.6

2.7

2.4
Anova: Two-Factor Without Replication

Summary Count Sum Average Variance


R410A 4 11.4 2.85 0.19
R32 4 13.6 3.4 0.913
R22 4 10.5 2.625 0.282

Brand A 3 8.1 2.7 0.04


Brand B 3 8 2.667 0.253
Brand C 3 11.7 3.9 0.61
Brand D 3 7.7 2.567 0.023

ANOVA
Source of Variation df Sum Sq Mean Sq F-value
Rows (Types of Refrigerants) 2 1.272 0.636 6.559
Columns (Different Brands of Chiller) 3 3.576 1.192 12.295
Error 6 0.582 0.097
Total 11 5.429
P-value F-critical
0.031 5.143
0.006 4.757
Anova: Two-Factor Without Replication

SUMMARY Count Sum Average Variance


R410A 4 11.4 2.85 0.19
R32 4 13.6 3.4 0.913333
R22 4 10.5 2.625 0.2825

Brand A 3 8.1 2.7 0.04


Brand B 3 8 2.666667 0.253333
Brand C 3 11.7 3.9 0.61
Brand D 3 7.7 2.566667 0.023333

ANOVA
Source of VariationSS df MS F P-value F crit
Rows 1.271667 2 0.635833 6.558739 0.030914 5.143253
Columns 3.575833 3 1.191944 12.29513 0.005667 4.757063
Error 0.581667 6 0.096944

Total 5.429167 11
Question:
In our university, one group of my mechanical engineering students conducted a research to investigate how the
axis wind turbines (VAWT) will perform when it comes to energy harnessing at the various locations in the provinc
available. The study generally aims to assess the feasibility of erecting wind turbines for power generation in the
which type is most efficient and to which location will it be best installed. In this experiment, three types of VAWT
investigated and four locations were considered such as onshore, offshore, mountain top, and plain field. The ins
turbines were observed for a month and the power generated per week (KW) were tabulated in the proceeding ta
Initial Hypothesis:
For Interaction:
Null Hypothesis (Ho): There is no interaction between Factors A and B.
Alternative Hypothesis (Ha): There is a significant interaction between Factors A and B.
For Factor A:
Null Hypothesis (Ho): There is no statistically significant difference among the Factor A (types of VAWT) means.
Alternative Hypothesis (Ha): There is a statistically significant difference to at least two of the Factor A (types of V
For Factor B:
Null Hypothesis (Ho): There is no statistically significant difference among the Factor B (installation locations) me
Alternative Hypothesis (Ha): There is a statistically significant difference to at least two of the Factor B (installatio

Installation Locations
Types of Vertical Axis Wind Turbines
Onshore Offshore
11.6 12.5
10.4 12.4
Savonius
10.7 12.8
11.1 12.2
16.2 19.4
14.9 19.8
Darrieus
15.4 18.9
15.1 19.2
12.4 16.3
12.1 15.9
Giromill
12.6 15.3
12.9 15.7
a research to investigate how the different types of vertical
the various locations in the province where strong winds were
bines for power generation in the area and to determine
experiment, three types of VAWT (of the same size) were
untain top, and plain field. The installed prototypes of wind
ere tabulated in the proceeding table.

A and B.

Factor A (types of VAWT) means .


ast two of the Factor A (types of VAWT) means.

Factor B (installation locations) means.


ast two of the Factor B (installation locations) means.

Installation Locations
Mountain Top Plain Field
10.2 9.8
9.8 8.7
10.5 10.2
10.6 9.4
17.6 15.8
17.3 16.1
17.8 15.2
17.2 15.1
13.5 11.8
13.2 12.2
14.1 12.5
13.6 11.7
Anova: Two-Factor with Replication

SUMMARY Onshore Offshore Mountain Top


Savonius
Count 4 4 4
Sum 43.8 49.9 41.1
Average 10.95 12.475 10.275
Variance 0.27 0.063 0.129

Darrieus
Count 4 4 4
Sum 61.6 77.3 69.9
Average 15.4 19.325 17.475
Variance 0.327 0.143 0.076

Giromill
Count 4 4 4
Sum 50 63.2 54.4
Average 12.5 15.8 13.6
Variance 0.113 0.173 0.140

Total
Count 12 12 12
Sum 155.4 190.4 165.4
Average 12.950 15.867 13.783
Variance 3.905 8.637 9.538

ANOVA
Source of Variation df Sum Sq Mean Sq
Samples (Types of VAWT) 2.00 302.314 151.16
Columns (Installation Locations) 3.00 84.142 28.05
Interaction 6.00 9.495 1.58
Error (Within) 36.00 6.628 0.18
Total 47 402.578
Plain Field Total

4 16
38.1 172.9
9.525 10.806
0.409 1.435

4 16
62.2 271
15.55 16.938
0.230 2.896

4 16
48.2 215.8
12.05 13.488
0.137 2.353

12
148.5
12.375
6.869

F-value P-value F-critical


821.071 0.000 3.259
152.351 0.000 2.866
8.596 0.000 2.364
Question:
Material engineers in an engineering design firm wanted to determine whether the material choices of the vendors have a statistic
supported impact or effect in terms of the purchased material's tensile strength (MPa). In addition, they also want to investigate if
is a significant difference on the tensile strength when these vendors' supplied materials will be subjected to three different operat
temperatures with which these materials are intended to be used for.
For Interaction:
Null Hypothesis (Ho): There is no interaction between Factors A and B.
Alternative Hypothesis (Ha): There is a significant interaction between Factors A and B.
For Factor A:
Null Hypothesis (Ho): There is no statistically significant difference among the Factor A (material vendors) means.
Alternative Hypothesis (Ha): There is a statistically significant difference to at least two of the Factor A (material vendors) means.
For Factor B:
Null Hypothesis (Ho): There is no statistically significant difference among the Factor B (intended operating temperatures) means
Alternative Hypothesis (Ha): There is a statistically significant difference to at least two of the Factor B (intended operating
temperatures) means.

Intended Operating Temperatures


Material Vendors
15 deg C 50 deg C
417.5 413.7
409.3 409.9
Vendor A 415.4 410.9
414.1 408.4
417.3 400.8
410.4 408.6
414.8 412.4
Vendor B 418.7 411.8
411.6 407.1
420.7 413.3
407.7 407.1
409.1 410.7
Vendor C 419.4 416.3
415.9 411.4
412.8 413.2
hoices of the vendors have a statistically
ition, they also want to investigate if there
be subjected to three different operating

rial vendors) means.


Factor A (material vendors) means.

ded operating temperatures) means.


Factor B (intended operating

g Temperatures
80 deg C
412.6
405.7
410.3
406.4
409.1
406.7
403.7
411.6
408.5
411.2
407.8
409.6
407.7
411.1
410.1
Anova: Two-Factor with Replication

SUMMARY 15 deg C 50 deg C 80 deg C


Vendor A
Count 4 4 4
Sum 43.8 49.9 41.1
Average 10.95 12.475 10.275
Variance 0.27 0.063 0.129

Vendor B
Count 5 5 5
Sum 2076.2 2053.2 2041.7
Average 415.24 410.64 408.34
Variance 19.663 7.05299999999994 10.7630000000001

Vendor C
Count 5 5 5
Sum 2064.9 2058.7 2046.3
Average 412.98 411.74 409.26
Variance 23.1769999999999 11.413 2.19300000000003

Total
Count 15 15 15
Sum 6214.7 6155.6 6132.1
Average 414.313 410.373 408.807
Variance 16.427 13.619 6.149

ANOVA
Source of Variation df Sum Sq Mean Sq
Samples (Material Vendors) 2 3.728 1.864
Columns (Intended Operating Temps) 2 241.507 120.754
Interaction 4 35.432 8.858
Error (Within) 36 467.576 12.988
Total 44 748.243
Total

16
172.9
10.806
1.435

15
6171.1
411.407
19.525

15
6169.9
411.327
13.072

F-value P-value F-critical


0.144 0.867 3.259
9.297 0.001 3.259
0.682 0.609 2.634

You might also like