1 prof. indrajit mukherjee, school of management, iit bombay blocking a replicated design consider...
DESCRIPTION
3 Prof. Indrajit Mukherjee, School of Management, IIT Bombay Experiment from Example 6.2 Suppose only 8 runs can be made from one batch of raw material Pilot Plant Filtration Value Experiment Run Number FactorsRun label Filtration Rate (gal/h) ABCD 1----(1) a b ab c ac bc abc d ad bd abd cd acd bcd abcd96TRANSCRIPT
1Prof. Indrajit Mukherjee, School of Management, IIT Bombay
Blocking a Replicated DesignConsider the example from Section 6-2; k = 2 factors, n = 3 replicates
This is the “usual” method for calculating a block sum of squares
2 23...
1 4 126.50
iBlocks
i
B ySS
Chemical Process Experiment in Three BlocksBlock 1 Block2 Block 3(1)=28 (1)=28 (1)=28a=36 a=36 a=36b=18 b=18 b=18ab=31 ab=31 ab=31
Block Totals B1=113 B2=113 B3=113
2Prof. Indrajit Mukherjee, School of Management, IIT Bombay
ANOVA for the Blocked DesignPage 267
Analysis of variance for the chemical process experiment in the three blocks
Source of variation
Sum of squares
Degrees of freedom
Mean square F0 P-value
Blocks 6.50 2 3.25
A(Concentration)208.33
1 208.33 50.32 0.0004
B(Catalyst)75.00
1 75.00 18.12 0.0053
AB8.33
1 8.33 2.01 0.2060
Error24.84
6 4.14
Total323.00
11
3Prof. Indrajit Mukherjee, School of Management, IIT Bombay
Experiment from Example 6.2
Suppose only 8 runs can be made from one batch of raw material
Pilot Plant Filtration Value ExperimentRun
NumberFactors Run
labelFiltration Rate
(gal/h)A B C D1 - - - - (1) 452 + - - - a 713 - + - - b 484 + + - - ab 655 - - + - c 686 + - + - ac 607 - + + - bc 808 + + + - abc 659 - - - + d 43
10 + - - + ad 10011 - + - + bd 4512 + + - + abd 10413 - - + + cd 7514 + - + + acd 8615 - + + + bcd 7016 + + + + abcd 96
4Prof. Indrajit Mukherjee, School of Management, IIT Bombay
The Table of + & - Signs, Example 6-4
A B AB C AC BC ABC D AD BD ABD CD ACD BCD ABCD1 - - + - + + - - + + - + - - +a + - - - - + + - - + + + + - -b - + - - + - + - + - + + - + -ab + + + - - - - - - - - + + + +c - - + + - - + - + + - - + + -ac + - - + + - - - - + + - - + +bc - + - + - + - - + - + - + - +abc + + + + + + + - - - - - - - -d - - + - + + - + + - + - + + -ad + - - - - + + + - - - - - + +bd - + - - + - + + + + - - + - +abd + + + - - - - + - + + - - - -cd - - + + - - + + + - + + - - +acd + - - + + - - + - - - + + - -bcd - + - + - + - + + + - + - + -abcd + + + + + + = + - + + + +r + +
Contrast constant for the 24 design
5Prof. Indrajit Mukherjee, School of Management, IIT Bombay
ABCD is Confounded with Blocks (Page 279)
Observations in block 1 are reduced by 20 units…this is the simulated “block effect”
Block 1 Block 2(1)=25 a=71ab=45 b=48ac=40 c=68bc=60 d=43ad=80 abc=65bd=25 bcd=70cd=55 acd=86
abcd=76 abd=104(b) Assignment of the 16
runs to two blocks
(a) Geometric ViewA
BC = Runs in Block 1 = Runs in Block 2
D - +
6Prof. Indrajit Mukherjee, School of Management, IIT Bombay
Effect EstimatesEffected Estimate for the Blocked 2k Design in Example
Model TermRegression Coefficient
Effect Estimate
Sum of Squares
Percent Contribution
A 10.81 21.625 1870.563 26.3B 1.56 3.125 39.0625 0.55C 4.94 9.875 390.0625 5.49D 7.31 14.625 855.5625 12.03AB 0.062 0.125 0.0625 <0.01AC -9.06 -18.125 1314.063 18.48AD 8.31 16.625 1105.563 15.55BC 1.19 2.375 22.5625 0.32BD -0.19 -0.375 0.5625 <0.01CD -0.56 -1.125 5.0625 0.07ABC 0.94 1.875 14.0625 0.2ABD 2.06 4.125 68.0625 0.96ACD -0.81 -1.625 10.5625 0.15BCD -1.31 -2.625 27.5625 0.39
Block (ABCD) -18.625 1387.563 19.51
7Prof. Indrajit Mukherjee, School of Management, IIT Bombay
The ANOVA
The ABCD interaction (or the block effect) is not considered as part of the error termThe reset of the analysis is unchanged from the original analysis
Analysis of variance for ExampleSource of variation
Sum of squares
Degrees of freedom
Mean square F0 P-value
Blocks 1387.5625 1A
1870.56251 1870.5625 89.76 <0.0001
C390.0625
1 390.0625 18.72 0.0019
D855.5625
1 855.5625 41.05 0.0001
AC1314.0625
1 1314.0625 63.05 <0.0001
AD1105.5625
1 1105.5625 53.05 <0.0001
Error187.5625
9 20.8403
Total711.4375
15
8Prof. Indrajit Mukherjee, School of Management, IIT Bombay
Another Illustration of the Importance of Blocking
Now the first eight runs (in run order) have filtration rate reduced by 20 units
The Modified Data From Example
Run Order Std OrderFactor A
TemperatureFactor B Pressure
Factor C Concentration
Factor D Stirring
Rate
Response Filtration
Rate8 1 -1 -1 -1 -1 25
11 2 1 -1 -1 -1 711 3 -1 1 -1 -1 283 4 1 1 -1 -1 459 5 -1 -1 1 -1 68
12 6 1 -1 1 -1 602 7 -1 1 1 -1 60
13 8 1 1 1 -1 657 9 -1 -1 -1 1 236 10 1 -1 -1 1 80
16 11 -1 1 -1 1 455 12 1 1 -1 1 84
14 13 -1 -1 1 1 7515 14 1 -1 1 1 8610 15 -1 1 1 1 704 16 1 1 1 1 76