This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentThis preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 2 k Factorial Experiments • k Factors • 2 Levels Each x i = level of factor i (1 or +1), i=1...k • 2 k Treatments: total number of combinations Notation • Treatments, Replicates, Runs • Labeling for Treatments ex) k=3 and x 1 =+1, x 2 =1, and x 3 =+1 label: ac • example: 2 3 experiment with 6 replicates 48 RUNS 2 DESIGN MATRIX FOR 2 3 EXPERIMENT Design Matrix specifies the treatments in STANDARD ORDER Treat ment x 1 x 2 x 3 Label for treatment Randomize Run Order 1111 (1) 6 2 +111 a 4 31 +11 b 7 4 +1 +11 ab 3 511 +1 c 2 6 +11 +1 ac 8 71 +1 +1 bc 1 8 +1 +1 +1 abc 5 What might be the consequence of not randomizing these runs? 3 The Router Experiment (Montgomery 1991, pp. 498) • A router cuts registration notches in printed circuit boards. The boards are fixed into position using the notches and components are inserted automatically. Variability in the notch dimension causes insertion problems. • Two factors: Factor Levels Bit Size (X 1 ) 1/16" and 1/8" Speed (X 2 ) 40 and 80 rpm • Performance Measure: An accelerometer measures vibration allowed on the X, Y, and Z coordinate axes. The resultant vector is the performance measure in units of cps, cycles per second....
View
Full
Document
This note was uploaded on 09/20/2011 for the course ENG 300 taught by Professor Albin during the Fall '11 term at Rutgers.
 Fall '11
 Albin

Click to edit the document details