Design of Experiment
Design of Experiment 11 January 2022
Tasks:
1. Full Factorial design data analysis
- Effect of single factors and thier rankings
- Interaction effects
2. Fractional Factorial design data analysis
- Effect of single factors and thier rankings
Case Study 2:
Excel File:DOE Excel (download file to view graph - website unable to view graph)
Full factorial design data analysis:
Effect of single factors
Table 1: Factor Levels
Level |
Factor A –
concentration of coagulant added |
Factor B –
Treatment Temperature |
Factor C –
Stirring Speed |
+ |
1% |
72 oF |
200 rpm |
- |
2% |
100 oF |
400 rpm |
Figure 1: Full factorial Data for Runs
Figure 2: Full factorial Calculation for Effect of Factors
|
Unit |
Factor A – Concentration
of coagulant added |
Factor B –
Treatment Temperature |
Factor C –
Stirring Speed |
+ |
Ib/day |
17.5 |
12 |
4 |
- |
5 |
10.5 |
18.5 |
|
Difference |
12.5 |
1.5 |
-14.5 |
The gradient of both lines are different by a little margin.
Therefore there’s an interaction between A and B, but the interaction is small.
At LOW B, when the concentration of coagulant added increases from 1% to 2%,
the amount of pollutant discharged increases by 12 from 4.5 lb/day to 16.5
lb/day. At HIGH B, when the concentration of coagulant added increases from 1%
to 2%, the amount of pollutant discharged increases by 13 from 5.5 lb/day to
18.5 lb/day. Since the total increase of pollutant discharged is higher at HIGH
B than LOW B by a little margin, the gradient of HIGH B is a little steeper
than at LOW B.
A x C
The gradient of both lines are different (one is + and the
other is -). Therefore there’s a significant interaction between A and C. At
LOW C, when the concentration of coagulant increases from 1% to 2%, the amount
of pollutant discharged increases by 26 from 5.5 lb/day to 31.5 lb/day. At HIGH
C, when the concentration of coagulant increases from 1% to 2%, the amount of
pollutant discharged decreases by 1 from 4.5 lb/day to 3.5 lb/day. Since the
total change of pollutant discharged at LOW C is a increase, while at HIGH C,
the total change of pollutant discharged is a decrease. Therefore the gradient
of both lines are of opposite values and opposite directions.
B x C
The gradient of both lines are different by a little margin.
Therefore, there’s an interaction between B and C, but the interaction is
small. At LOW C, when the Treatment temperature increases from 72 oF
to 100 oF, the amount of pollutant discharge increases by 2 from
17.5 lb/day to 19.5 lb/day. At HIGH C, when the Treatment temperature increases
from 72 oF to 100 oF, the amount of pollutant discharge
increases by 1 from 3.5 lb/day to 4.5 lb/day.
Conclusion for Full Factorial Data Anaylsis:
In conclusion, Factor C Stirring Speed is the most significant factor effecting the amount of pollutant discharged followed by Factor A Concentration of coagulant added then Factor B Treatment Temperature. Factor A and B both increased amount of pollutant discharged when their levels were increased from low to high, only Factor C decreased amount of pollutant discharged when the level was increased from low to high. In terms of interactions between the factors, there is significant interaction for A x C but small interaction for A x B and B x C.
Fractional factorial design data analysis:
Effect of single factors
Figure 10: Data for fractional runs
Figure 11: Calculation for fractional factorial factor effect
|
Unit |
Factor A –
Concentration of coagulant added |
Factor B –
Treatment Temperature |
Factor C –
Stirring Speed |
+ |
Ib/day |
18 |
19 |
4 |
- |
5 |
4 |
19 |
|
Difference |
13 |
15 |
-15 |
- When Concentration of coagulant added increases from 1% to 2%, the amount of pollutant discharged increases from 5 lb/day to 18 lb/day.
- When Treatment Temperature increases from 72 oF to 100 oF, the amount of pollutant increases from 4 lb/day to 19 lb/day.
- When Stirring Speed increases from 200 to 400 rpm, the amount of pollutant decreases from 19 lb/day to 4 lb/day.
Ranking:
1. Stirring Speed/Treatment temperature -> 3. Concentration of coagulant added
Conclusion for Fractional Factorial Data Analysis:
In conclusion, the most significant factor remains C Stirring Speed but it is tied to Factor B Treatment Temperature then Factor A concentration of Coagulant added. There is a difference between the ranking for full fractional and fractional fractorial as factor B treatment temperature increased its ranking from least significant to most significant. This means that some of the data collected may be wrong hence further study is necessary.
·
·
Comments
Post a Comment