Applied Statistics and Probability for Engineers

(Chris Devlin) #1
552 CHAPTER 14 DESIGN OF EXPERIMENTS WITH SEVERAL FACTORS

For all three pairs of effect estimates, we would obtain the following results:
Effect,i from (lili) from (lili)
i A
i B
i C
Thus, by combining a sequence of two fractional factorial designs, we can isolate both the
main effects and the two-factor interactions. This property makes the fractional factorial de-
sign highly useful in experimental problems since we can run sequences of small, efficient ex-
periments, combine information across severalexperiments, and take advantage of learning
about the process we are experimenting with as we go along. This is an illustration of the con-
cept of sequential experimentation.
A 2k^1 design may be constructed by writing down the treatment combinations for a full
factorial with k 1 factors, called the basic design,and then adding the kth factor by identi-
fying its plus and minus levels with the plus and minus signs of the highest order interaction.
Therefore, a 2^3 ^1 fractional factorial is constructed by writing down the basic design as a full
22 factorial and then equating factor Cwith the ABinteraction. Thus, to construct the prin-
cipal fraction, we would use C ABas follows:

(^1)  21 CABCAB 2 C (^1)  23 CAB 1 CAB 24 AB
(^1)  21 BACBAC 2 B (^1)  23 BAC 1 BAC 24 AC
(^1)  21 ABCABC 2 A (^1)  2 3 ABC 1 ABC 24 BC
(^1) 
2
(^1) 
2
Table 14-25 The 2^4 ^1 Design with Defining Relation I ABCD
Treatment Etch
ABC D ABC Combination Rate
 550
 ad 749
 bd 1052
 ab 650
 cd 1075
 ac 642
 bc 601
 abcd 729
112
Basic Design Fractional Design
Full 2^223 ^1 ,IABC
AB A BC AB




To obtain the alternate fraction we would equate the last column to C AB.
EXAMPLE 14-7 To illustrate the use of a one-half fraction, consider the plasma etch experiment described in
Example 14-5. Suppose that we decide to use a 2^4 ^1 design with I ABCDto investigate the
four factors gap (A), pressure (B), C 2 F 6 flow rate (C), and power setting (D). This design
would be constructed by writing down as the basic design a 2^3 in the factors A, B, and Cand
then setting the levels of the fourth factor D ABC. The design and the resulting etch rates
are shown in Table 14-25. The design is shown graphically in Fig. 14-29.
c 14 .qxd 5/9/02 7:54 PM Page 552 RK UL 6 RK UL 6:Desktop Folder:TEMP WORK:MONTGOMERY:REVISES UPLO D CH112 FIN L:

Free download pdf